No title

The world’s path to the better mousetrap: myth or reality? An empirical investigation into the launch strategies of high and low advantage new products Erik Jan Hultink and Susan Hart

The authors Erik Jan Hultink is an Associate Professor of Marketing at Delft University of Technology, The Netherlands. Susan Hart is a Professor of Marketing at the University of Strathclyde, Scotland. Abstract Focuses on product advantage, a major contributing factor to new product performance, by examining the launch strategies associated with high and low levels of product advantage. Views a launch strategy as integrating protocol decisions, which have steered the course of a product’s development, with the tactical marketing mix decisions. Data confirm all associations between key elements of new product protocol and product advantage. Growthrelated objectives guide the development of new products with high advantage, while the speedy development and early timing of the projects, the focus on growth markets, and the use of a niche targeting strategy are the hallmarks of products with high advantage. Contends that companies offering the world a better mousetrap do not believe the myth that a path to its door will be beaten; the better mousetrap requires and receives a different launch treatment from more pedestrian competitors.

European Journal of Innovation Management Volume 1 · Number 3 · 1998 · pp. 106–122 © MCB University Press · ISSN 1460-1060

Introduction Among the many factors shown to be critical to new product performance by the literature, product advantage recurs as a powerful differentiating factor between successful and unsuccessful new products (Cooper, 1979, 1993; Link, 1987; Montoya-Weiss and Calantone, 1994). However, it is equally well acknowledged by the literature that the world will not automatically beat a path to the door of the better mousetrap as Waldo Emerson would have us believe, but that the communication of the advantages offered are required via a proficient launch of the new product (Cooper, 1979, 1980; Edgett et al., 1992; Green and Ryans, 1990; Green et al., 1995; Hultink et al., 1997; Maidique and Zirger, 1984). This said, there have been few studies focusing on the details of a launch strategy, leaving an important gap in our understanding of how better (or for that matter, worse) mousetraps get launched. This paper takes as its focus product advantage, a major contributing factor to new product performance, by examining the launch strategies associated with high and low levels of product advantage. To date, the relationships between product advantage and other recurrent success factors in the literature are uncharted, raising questions such as: How does product advantage affect product development cycle time, the targeting strategy, marketing communication options, or the training of the salesforce to “sell in” the new product? In the present paper, we seek to explore how products offering different degrees of advantage are introduced with different sets of launch strategies. The organization of this paper is as follows. First, we review the components of a launch strategy and develop propositions of how launch decisions are associated with products of high and low levels of advantage. Then, we describe the research method used to examine these propositions. Finally, we present the results of our study and draw conclusions. The authors are grateful to Rudy Moenaert, Sylvia Mooy, Walle Oppedijk van Veen and Henry Robben for their useful comments on a previous draft of this paper, and to Kirsty Garrett for research assistance.

106

The world’s path to the better mousetrap: myth or reality?

European Journal of Innovation Management

Erik Jan Hultink and Susan Hart

Volume 1 · Number 3 · 1998 · 106–122

Defining a launch strategy The process of formulating a launch strategy is not a trivial matter. Of all the steps in the new product development (NPD) process, the product launch often requires the largest commitment in time, money, and managerial resources (Urban and Hauser, 1993). A launch strategy describes those marketing decisions that are necessary to position a new product to its target market. These decisions are referred to under the collective terms of market entry, launch strategy, product launch, commercialization and introduction. Surprisingly, few research studies operationalize the term used to describe these decisions from which we can derive a precise definition of the phase, because previous studies have focused on the market launch in a number of contexts. Yoon and Lilien (1985), Urban et al. (1986) and Green and Ryans (1990) consider the market launch within the context of NPD, while other studies focus on a strategic entry to a new market (Biggadike, 1979; Lambkin, 1988; Lieberman and Montgomery, 1988), or are related to market entry to international (export) markets (Root, 1994; Ryans, 1988). In addition, there is no consistency of decisions belonging to a launch strategy in the new product literature. For example, Choffray and Lilien (1984, 1986) included pricing, timing, and distribution decisions, whereas Green and Ryans (1990) included timing, and marketing and R&D expenditures. Specifically, it appears that researchers include as variables pertinent to the launch, both those decisions which are set to guide the development process, such as the likely target markets, as well as the more detailed considerations such as the level of promotion expenditures and pricing (Hultink et al., 1997). Crawford (1984, 1994) refers to the former set in discussing the product innovation charter or protocol. A new product’s protocol defines the new product’s objectives to the firm, its technologies, target markets and competitive stance. These protocol decisions take place before the development process begins, and serve as benchmarks against which evaluations about the direction of the development can be made as the process unwinds. They are set apart in time from the actual launch, but are very much a part of it as they set the parameters in which the new product will compete: as a technological

innovation in the category, as a cost-reduced version for certain market segments or geographic regions, or as a set of solutions for a special segment of the market, to give a few examples. Table I shows the key decisions in launching a new product culled from the major studies in the subject area. Among the authors contributing to the “market entry” studies (studies 1 to 11), their focus on pricing, distribution, and marketing expenditures is in most instances marginal. On the other hand, the seven studies (12 to 18) that focus on the tactical launch decisions do not examine protocol related launch decisions (as opposed to communication, distribution, pricing and so on) to a full extent. Clearly, the protocol decisions govern key elements of the tactical launch: pricing decisions, promotional messages and channels of distribution will be engaged in order to capitalize on these elements of the protocol (Biggadike, 1979; Hultink et al., 1997; Urban et al., 1986). The more detailed launch considerations are, in effect, the elements of the marketing mix. To include both levels of decisions is valid, as both sets of decisions affect the nature of the overall launch strategy. However, we consider it important to distinguish between both levels because of the way in which the strategic position defined by the protocol definition sets the context for the more tactical launch decisions (Hultink et al., 1997). In other words, once the protocol is fulfilled, its dimensions are much harder to reverse than the tactical marketing mix decisions. This notion resembles the path-dependence and lock-in phenomena provided by Ghemawat (1991). Our view of the launch integrates protocol decisions, which have steered the course of a product’s development (Biggadike, 1979; Booz-Allen and Hamilton, 1982; Crawford, 1984, 1994) with the tactical launch decisions. The pertinent decisions for the protocol are: the objectives set for the new product launch, timing, targeting, and characteristics of the market entered. The tactical launch decisions involve the marketing mix decisions: product, promotion, pricing and distribution.

The nature of product advantage in new product launch decisions Product advantage is defined as the outcome of the NPD process comprising the degree of

107

The world’s path to the better mousetrap: myth or reality?

European Journal of Innovation Management

Erik Jan Hultink and Susan Hart

Volume 1 · Number 3 · 1998 · 106–122

Table I Key decisions in launching a new product

Variables

Previous study

Strategic launch decisions (protocol) Objectives Timing NPD cycle time Targeting Target market growth rate Target market PLC stage

1, 7, 18 2, 3, 4, 5, 9, 10, 11, 12, 16, 18 5, 9, 18 1, 2, 5 9, 12, 17 1, 12, 18

Tactical (marketing mix) launch decisions Breadth of product line Promotion Advertising Salesforce Marketing expenditures Pricing strategy Distribution Key: Market entry studies 1. Biggadike (1979) 2. Lambkin (1988) 3. Urban et al. (1986) 4. Ryans (1988) 5. Robinson and Fornell (1985) 6. Schmalensee (1982) 7. Roberts and Berry (1985) 8. MacMillan and Day (1987) 9. Green and Ryans (1990) 10. Lieberman and Montgomery (1988) 11. Glazer (1985) unique benefits not previously available, the extent to which customer needs are better satisfied, the product’s relative quality and innovativeness, and the extent to which the new product solves customer problems better (Cooper, 1979). Product advantage has consistently been shown to be a key differentiator between success and failure in the development of new products and services alike (Craig and Hart, 1992; Montoya-Weiss and Calantone, 1994), although its relationship with other differentiators of success and failure in NPD has not been the focus of investigation. With the exception of Cooper (1979), few researchers have defined the details of what is meant by “product advantage”. Cooper’s (1979) conceptualization embraces the extent to which the product offers unique benefits to the customer, whether the new product is of higher quality than competitive offerings, the extent to which it reduces customer costs, how innovative the product is, whether it is

1, 2 13, 14, 15, 16 13, 14, 15, 16, 17 15, 16, 17 1, 2, 3, 9, 16 1, 2, 13, 14, 16, 17, 18 1, 2, 13, 14, 15, 16, 18

Studies of tactical launch variable 12. Yoon and Lilien (1985) 13. Little (1975) 14. Wind (1982) 15. Crawford (1984) 16. Urban and Hauser (1993) 17. Cooper (1993) 18. Choffray and Lilien (1986)

superior in the eyes of the customer, and whether the new product solves a customer problem. His definition, therefore, is one that emphasizes both the viewpoints of the customer and competitor. Little attention, however, has been paid to the nature of the launch strategy given the achieved level of product advantage, although the launch is said to be, in itself, an important contributory factor to new product performance (Green et al., 1995; Hultink et al., 1997). Yet, it is intuitively logical that the launch strategy will differ, depending on whether the new products developed have achieved a high or low level of advantage. Below, we formulate propositions regarding the different launch strategies in conditions of high and low product advantage. We start with the propositions on the protocol issues. Then, we will formulate the propositions on the tactical launch decisions. 108

The world’s path to the better mousetrap: myth or reality?

European Journal of Innovation Management

Erik Jan Hultink and Susan Hart

Volume 1 · Number 3 · 1998 · 106–122

Protocol decisions Protocol issues encompass the objectives set for the new product launch, timing, targeting, and characteristics of the market targeted (Biggadike, 1979; Cooper, 1993; Crawford, 1984). According to Mahajan and Wind (1992), “increasing market penetration” dominated all other new product objectives for the firms in their sample. Yoon and Lilien (1985) found that original new products had different objectives from reformulated new products, with the former being more oriented towards diversification. There is, however, little in the way of empirical research to reveal which objectives are set for new product programmes, nor how these might relate to corporate strategy. This is surprising, given the centrality of product development to the implementation of strategic options presented by frameworks such as Ansoff’s (1965) growth vector matrix, Porter’s (1980) generic strategies, or Miles and Snow’s (1978) strategic configurations. Specifically, new product objectives such as “increasing market penetration”, “gaining a foothold in a new market”, “extending the seasonality of the product” or “penetrating a new market segment” may exemplify options chosen to pursue growth. Indeed, Cardozo et al. (1993) whose focus was on which product-market strategies provided highest growth, found that this was achieved where companies build on their initial core offering, that is, pursued a strategy of market penetration. To build effectively on a core offering, it is logical to develop new (or improved) products with higher levels of advantage than those already on the market. Accordingly, we propose that: P1. New products with high advantage will be driven by objectives that focus on growth, while new products with low advantage will be driven by objectives that are intended to defend current market positions.

(Dumaine, 1989; Rosenau, 1990; Smith and Reinersten, 1991), the ability to create greater barriers to entry for potential competitive offerings (Bain, 1956; Porter, 1980), and enhanced market image due to being a technological leader. There are counterarguments, however. Wensley (1982) and Aaker and Day (1986) suggest that the advantages of early entry are not automatic due to the technological and market uncertainties inherent in the development of new products. These may allow later entrants to develop superior skills for the market, thereby allowing them to outplay the first movers with their own offering. Of concern to the current study, however, is the extent to which the timing decision is related to the degree of product advantage. Yoon and Lilien (1985) found that for reformulated new products, being a pioneer may be more beneficial than being a fast follower. According to Urban et al. (1986) an early entrant will not be able to reap any of the hypothesized advantages of being early to market, if the product itself is of little market appeal or suffers from poor positioning. Given these findings, we propose: P2.The swifter the product’s development, the higher the degree of product advantage. P3. The earlier the market entry in comparison with competitors, the higher the degree of product advantage.

Timing is related to the order of entry issue. This order of entry issue is a huge area of debate in its own right, with opposing views on the advantages of being “first-mover” (Golder and Tellis, 1993). The advantages thought to accrue to early entrants include the acquisition of market knowledge (Stigler, 1981), the freedom to charge a premium price until competitive products are launched

Characteristics of the market into which the new product will be launched should be a decision with its roots early in the NPD process, at the time of the initial definition of protocol (Cooper, 1993; Crawford, 1984). Protocol involves specification of the degree of innovativeness being pursued, together with a specification of the final markets into which the product will be launched. In other words, the objectives set in terms of the advantage to be achieved are intrinsic to the choice of launch markets. Cooper (1984) highlighted different combinations of protocol, concluding that strategies which targeted high potential growth markets, together with those avoiding markets where competitors are dominant, were more successful. Where a high level of advantage has been developed in a new product, maximizing return from the development effort may be easier to achieve in high growth markets, where competitors are not

109

The world’s path to the better mousetrap: myth or reality?

European Journal of Innovation Management

Erik Jan Hultink and Susan Hart

Volume 1 · Number 3 · 1998 · 106–122

dominant. These observations lead to the proposition that: P4. New products with high advantage are launched into high growth markets where competitors have not achieved dominance.

and image. In the case of a superior product with discernible advantages, the communication of advantages may require a distinctive brand name. Doyle (1992) suggests that branding is only successful if built on clear product differentiation and advantage. In other words, product advantage develops the brand. The number of versions, models or variations in the new product will depend on the strategic nature of the development; whether the development is an addition to an existing line of products or a completely new line (or range) of products. The number of products included in the launch is related to the breadth of the target segments and the desired position of the company in the particular product-market. Biggadike (1979) found that new entrants generally launched fewer line items into a product-market than the incumbent players, but also that the length of the product line was positively related to the number of customers being targeted. Lambkin (1988) found that pioneering firms offered a broader product line and achieved higher market shares and long-term profit advantages over their rivals. This discussion of the product decisions leads us to propose that: P6. Products displaying high levels of advantage will adopt a brand name and will be offered as a broader range than products of less advantage.

Linked to these decisions is the fundamental choice between attempting to sell the new product to everyone (a mass-marketing strategy), and targeting the product to a specific group of customers who, it is felt, have a predisposition to the product’s attributes (a differentiated targeting strategy). Since the concept of product advantage embodies customer-relevance, it is to be expected that companies seeking high levels of product advantage will eschew an “undifferentiated” approach to product launch. The specific launch decision relates to the targets for new product sales. We therefore propose that: P5. New products with high levels of advantage will be launched using a differentiated targeting strategy. Tactical (marketing mix) launch decisions Having outlined the protocol issues that are in place before the final scheduling of the launch is considered, we turn attention to the tactical launch decisions. At this stage, the traditional marketing mix becomes relevant: product, promotion, pricing, and distribution, all of which should be calibrated in such a way as to make operational the posture taken via the protocol decisions outlined above. In short, having created “the better mousetrap”, how does the launch effort ensure that “the world will beat a path to its door”? We deal with each of these sets in turn to examine their relationships with product advantage. Product decisions extend to the choice of branding and the breadth of the product line, also called product assortment. Branding is more than product identification; it is also a powerful positioning tool and is equally related to issues governing the choice of product assortment. Where identification is the only purpose, a combination of letters and models may well suffice. If the identification of the product is linked to the position of the product with respect to other products offered by the company, then the brand name may be used to communicate this relationship. In the case of a strong company identity, the brand name chosen may echo the company identity

Promotion decisions encompass the range of communication and motivation instruments needed to raise awareness and precipitate purchase of the new product (Calantone and Montoya-Weiss, 1993; Moore and Pessemier, 1993). This encompasses both communication, sales promotion and deployment of the salesforce. We do not attempt to summarize the content and wisdom from research into advertising, a task well beyond the scope of the current paper. However, in conjunction with sales promotion, advertising is a crucial element in the launch of a new product, and is also linked to the push versus pull approach. Total advertising expenditures have been shown to impact the performance of new product introductions (Biggadike, 1979; Lambkin, 1988), which is unsurprising, given its role in positioning the product and creating awareness, interest and trial (Wind, 1982). The proportion of investment in marketing communications and sales promotion given

110

The world’s path to the better mousetrap: myth or reality?

European Journal of Innovation Management

Erik Jan Hultink and Susan Hart

Volume 1 · Number 3 · 1998 · 106–122

over to trade and consumers, is linked to the extent to which the new product is following a push (intermediary-focused) or a pull (final customer-focused) strategy. Although there is not much empirical literature on which to draw when making normative statements regarding this division of resources, lower levels of awareness for the new product have been suggested as requiring a pull, rather than a push strategy for advertising and promotion (Hisrich and Peters, 1991). Following on from our earlier discussions on the nature of product advantage, it is to be expected that customers are not as aware of products with high advantage as they might be with products of lower advantage, which are more related in functionality and appeal to their predecessors. The communication of the benefits of the new product is crucial to its adoption, but again, in the “effective” NPD process, the key facets of the communication mix are evolving alongside the technical development. Indeed, as Urban and Hauser (1993) point out, advertising testing may be seen as part of the overall “product testing” cycle which takes place at an earlier point in the NPD process, since advertising is a component of the product’s overall design and positioning. Several of the tenets underpinning the testing of advertisements are commonly used in concept and product testing: likeability, believability and so on. This point underlines the one made by Cooper (1993, p. 228) when he states that “marketing planning is an ongoing activity that occurs formally and informally throughout much of the NPD process. Informally, it begins during the first few stages of the game plan, right after the idea stage”. Although there is a comparative lack of attention given to sales promotion in the marketing literature, it is important to marketing managers, particularly in new product launch (Moore and Pessemier, 1993; Wind, 1982). Promotional techniques include those aimed at the intermediaries – retailers and wholesalers – and include discounts, training, point of sale material, direct mail, mail coupons, off-pack discounts and other incentive offerings. These techniques are introduced to increase stocking by the chosen distributors and trial by end users. Obviously, the availability of any new product to the customer is vital to its eventual purchase and is crucial to ensure that communication expenditures are not wasted. For new products with high advantage, where, arguably,

there is a greater task involved in communicating the benefits and encouraging trial, there is an accompanying requirement for higher levels of input across the range of sales promotion techniques. In this respect, Rogers (1983) suggests that innovators are dependent upon the sources and information which operate via the formal media of mass communication, while later purchasers rely on word of mouth communication. Cooper (1993) contends that for the majority of new products, decisions regarding the salesforce will be straightforward. There is little empirical research to confirm or refute this view. Where the new product is aimed at markets already served by the company, the existing salesforce will require training in the new product and sales management will be required to plan the inclusion of the new product in call schedules and targets. In addition, sales aids will have to be produced and time must be devoted to motivating the salesforce (Cooper, 1993; Crawford, 1984). It is logical, however, to suggest that the salesforce will require specialist training in the advantages offered by the new product to sell them effectively to both distributors and end customers. To sum up on the foregoing observations, new products with high advantage will merit more attention on all promotion techniques. We therefore propose that: P7. New products with high levels of advantage will be launched with higher levels of marketing expenditures, and with a higher proportion being aimed at the trade. P8. New products with high levels of advantage will receive greater coverage by all types of marketing communication channels and will be given greater levels of attention by the salesforce. Pricing a new product, at the time of launch, is an integral element in its appeal (or lack of appeal). The price reflects its competitive positioning and for customers it may be a measure of the product’s advantage. The pricing decision for a new product extends to the choice between skimming and penetration (Kotler, 1994; Woodside, 1995). Until recently, a skimming strategy was advised as the most profitable route, especially for high advantage, innovative products (Wind, 1982). The initial high price was deemed appropriate for products with a clear, unique advantage and was thought to allow for greater recovery

111

The world’s path to the better mousetrap: myth or reality?

European Journal of Innovation Management

Erik Jan Hultink and Susan Hart

Volume 1 · Number 3 · 1998 · 106–122

of development costs. In addition, the assumption was reinforced associating advantage with being first to market, so a high initial price was tenable until the arrival of serious competition on the market. This said, the tendency toward shorter life cycles and decreasing time lags between first movers and early “me-too” products have caused a rethink in pricing strategy for new products. Specifically, where a product’s diffusion into the market is considered likely to follow the typical diffusion curve, there is an argument for employing penetration pricing, to hinder competitive product launches and to benefit from increasing economies of scale as volume sales of the products increase along with diffusion (Choffray and Lilien, 1984, 1986). This view requires a longer-term perspective of the recovery of development costs. In addition, the decision regarding skimming or penetrating is also dependent upon the scale of entry. Where the scale of entry is small, skimming is advised, where it is large, then penetration is preferred (Abell, 1975). To date, however, there is little in the way of empirical evidence to suggest which pricing strategies are associated with high or low levels of product advantage. Observations of the literature to date suggest that: P9. Products with high levels of advantage will be priced at a higher level than comparable competitive products and tend to follow a skimming strategy.

intended positioning of the product (Cooper and Kleinschmidt, 1987). In particular, the role of distributors in providing customer services is a crucial aspect of positioning. Bowersox (1990) suggests three facets of customer service: availability, capability and quality. The availability of the product relates to the levels of inventory carried by the distributor, which have an impact on the delivery lead times for orders. Capability refers to the organization and procedures of the distributor in processing and delivering the orders, once placed. Quality is concerned with the core of the distributor’s own business performance: error free order processing, delivery, invoicing and backup services such as complaints handling. It is thought desirable that these elements feed into the choice of a distributor for a new product and that they are monitored as part of the launch in order to maximize the impact of the launch. It is generally recognized, however, that the task of promoting the product through the distribution channel is one which requires careful management: distributors must agree on their role. It is important that objectives are set for the distribution so that this element of the marketing mix can be monitored effectively during launch. There is very little in the NPD literature that examines the role and impact of the distribution strategy on product launch. Biggadike (1979) found that the degree of newness in new products was related to the level of after-sales service. Specifically, incremental innovations offered more aftersales support, although this finding is not confined to the level of service provided by distributors. Lambkin (1988) reports that pioneering companies which generally perform better in terms of market share and return on sales, also display higher levels of customer services, but she does not relate customer services specifically to distribution. This discussion leads us to propose that: P10. Products with high advantage will be distributed more selectively than lower advantage new products, and will be supported by higher levels of distribution expenditures. P11. For products with high advantage, greater attention will be given to the choice of distributors and the motivation toward achieving quality standards in their roles.

Distribution is crucial in the eventual acceptance and sales of a new product in the market as it governs the availability of the new product to customers (Calantone and MontoyaWeiss, 1993; Moore and Pessemier, 1993). It goes without saying that the distribution channels chosen must reflect the target market’s buying behaviour and allow for maximum availability to the target market. The distribution channels chosen may reinforce or dilute the intended message of the product’s positioning in the marketplace. Thus, a product may benefit from extensive distribution if it is intended for mass markets, whereas selective or exclusive distribution will be more appropriate for products being aimed at differentiated or niche market segments. In addition to the selectivity chosen for the distribution of the new product, the quality of the distribution will also help, or hinder, the

112

The world’s path to the better mousetrap: myth or reality?

European Journal of Innovation Management

Erik Jan Hultink and Susan Hart

Volume 1 · Number 3 · 1998 · 106–122

advantage new products across industries (χ2 = 7.3, df = 4; p > 0.10).

The thrust of these 11 propositions is that the level of product advantage achieved, seen as the outcome of the proficiency of the NPD process, will be evident in the nature of the launch, at the protocol and at the tactical launch levels.

Research method Industries and sample To examine these propositions, we chose to investigate a variety of industries since studies which suggest that product advantage is a key determinant of new product performance have been conducted in both consumer and industrial markets (Craig and Hart, 1992; Montoya-Weiss and Calantone, 1994). The data used in the present study were drawn from a mail survey of 293 firms across five industries covering both consumer and industrial goods manufacturers in the UK. The industries (consumer durables, packaged goods, transport, chemicals and construction/ installation) were chosen due to their reputation for a large number of new product introductions annually. From an original sample frame of 1,906 firms listed in the major manufacturing directories, after initial contact by telephone, 533 were excluded due to a company policy of confidentiality regarding new product information. A further 271 were unable to identify a recent new product and were therefore excluded. Of the remaining 1,102 potential respondents, 497 agreed to participate in the survey and the effective number of usable questionnaires returned was 293, providing an effective response rate of 27 per cent. It is worth noting that during this selection process, we spoke personally to managers who identified themselves as having been in charge of NPD in the companies and directly involved with the particular products about which they chose to answer questions. In addition, prenotification by phone was used to solicit cooperation, to explain the purpose of the study, and to increase the response rate (Yu and Cooper, 1983). Respondents provided information on a total of 493 new products. To evaluate whether there was any industry bias with respect to the proportions of high and low advantage new products, we crosstabulated the level of advantage with industry category. The results showed that there is an even distribution of low and high

Measured variables An overview of the measures used to describe the product advantage construct and the launch strategy decisions for the new products is presented below. Product advantage was measured using a summated scale comprising six items derived from the literature (Cooper, 1979): the extent to which the new product offered unique benefits, was higher quality than competitors, reduced customers costs, was innovative, was superior in the eyes of the customers, and solved a customer problem. Each item was measured on a five-point scale, where 1 = “does not apply at all” and 5 = “applies completely”. The reliability of the scale was high (Cronbach’s α = 0.85). We then constructed three categories of product advantage: low advantage (1-2.67, N = 155), medium advantage (2.68-3.66, N = 173), and high advantage (3.67-5.00, N = 165). In order to provide better discrimination between product advantage and the launch strategies used, we excluded from the analyses the medium advantage category. Therefore, we will contrast the low and high product advantage categories below. The product innovation charter, or protocol, as specified in the literature review above has four sets of components: commercialization objectives, timing, targeting and characteristics of the target market selected. The objectives set were measured as dichotomous variables where 1 = “objective set” and 0 = “objective was not set”. The objectives, again taken from the relevant literature (Mahajan and Wind, 1992) included growth-oriented examples such as increasing market penetration, gaining a foothold in new markets, and penetrating a new market segment, as well as defensive objectives such as using excess capacity and producing existing products at lower costs. Timing was measured in two ways. First, the time lapsed between ideation and launch (i.e. the swiftness of the product’s development). Three categories were used: less than one year, one to three years, and more than three years. The second variable described the number of competitors on the market at the time of launch, thereby capturing when, relative to competitors, products were launched.

113

The world’s path to the better mousetrap: myth or reality?

European Journal of Innovation Management

Erik Jan Hultink and Susan Hart

Volume 1 · Number 3 · 1998 · 106–122

Market characteristics were described in terms of two variables. The first variable sought to capture market growth by asking respondents to classify the market growth rate at the time of launch as being negative, between 0 per cent and 5 per cent, between 6 per cent and 10 per cent, and over 10 per cent. The second variable indicated the life cycle stage of the product category. The variable focusing on the targeting strategies employed was measured using three categories: niche, selective or mass market. Comprehensive definitions of these terms were given to the respondents on the questionnaire. The tactical launch decisions relate to the four basic elements of the marketing mix. The first tactical launch decisions are productrelated. Branding options for the new product were investigated by asking respondents to identify whether the product was given a new brand name, adopted one already in existence or emphasized the company name. In the analyses, we reduced this further to two categories: those using a brand name and those using the company name. The product assortment chosen for the launch was examined using a categorical question, where respondents indicated the breadth of the product range compared to competitors. Promotion decisions included the amount of investment in the promotional effort, measured in comparison to competitors and the orientation of this expenditure to the end customers or distributors, which were expressed as percentages. In addition, we examined nine types of promotion for their use as part of the launch promotion mix, and calculated the breadth of the mix of communication types. Finally, the intensity of the salesforce was investigated by asking respondents to indicate whether this was highly-, moderately-, or non-intensive. These three options were described in terms of specialist training, special incentives and targets for the launch of the new product. Pricing decisions were examined in two ways. First, the pricing strategy was categorized as being a skimming, penetration or an other strategy. Second, respondents classified the price level relative to competitive products. Distribution decisions were examined along four dimensions. First, distribution intensity was categorized by respondents as intensive, selective or exclusive, with each label being explained in the

questionnaire. Second, distribution expenditures were assessed in relation to competitive spending. Third, four factors which described distributors’ approach to quality were examined as to their impact on the decision to appoint distributors, and respondents were asked to say whether they used current, new, or a combination of distributors for the new product. Fourth, the questionnaire examined whether distributors were set targets for the launch period in respect of: stock levels held, flexibility in response to customers, error-free order processing, and after sales service. Throughout the questionnaire, we were careful to avoid the possibility of systematic biases in attributions. Given that respondents were asked to provide information about product advantage, it was important to avoid “leading” respondents into selecting the “appropriate” response which might explain the advantage achieved (Curren et al., 1992; Weiner, 1986). We therefore chose not to measure launch variables in terms such as “How well was the pricing strategy executed for this product?”; 1 = “not at all well”; 5 = “extremely well”. Instead, we simply asked for a description of the decision made. While the possibility of attribution bias cannot be ruled out completely in retrospective research, there is reason to expect that asking respondents to record decisions (rather than the proficiency of decisions) is an improvement on the standard methodology which has been the subject of criticism (Craig and Hart, 1992).

Analyses and results Protocol and product advantage In order to examine the associations between product advantage and protocol, we employed a number of bivariate techniques. Table II shows the results of the analyses examining the associations between product advantage and the commercialization objectives set for the new products. Of the six results shown, four are significant, a higher proportion than would have been expected by chance. The majority of the results are consistent with our expectations. Specifically, with products of high advantage, the objectives set were more often to increase market penetration, to establish a foothold in a new market, and to offset a seasonal cycle. Low advantage new products, on the other

114

The world’s path to the better mousetrap: myth or reality?

European Journal of Innovation Management

Erik Jan Hultink and Susan Hart

Volume 1 · Number 3 · 1998 · 106–122

Table II Product advantage and protocol: objectives

Product advantage High Low N = 165 N = 155 (54%) (46%)

Objective Increase market penetration (88%) Utilise excess capacity (21%) Lower costs (42%) Foothold in new market (17%) Offset seasonal cycle (8%) Penetrate a new segment (9%)

92% 13% 45% 26% 12% 11%

83% 29% 39% 7% 4% 7%

χ2 (df = 1) 5.90* 11.90*** 1.49 18.80*** 5.24* 1.99

Notes: ***p < 0.001,* p < 0.05 hand, had a significantly increased tendency to be driven by the need to use excess capacity. The second element in the protocol is the timing and swiftness of the new product’s launch. Table III shows the results of the investigation into the relationship between NPD cycle time, the number of competitors on the market at the time of launch, and product advantage. These results confirm propositions 2 and 3. Taken together with the number of competitors on the market at the time of launch, where products with high advantage tend to be launched into markets with fewer competitors, our data confirm the findings of Yoon and Lilien (1985) who suggested that high advantage new products tend to be among the first to market and those with low advantage tend to be followers.

Turning to market-related protocol decisions, Table IV shows the associations between product advantage, market growth rate and the life cycle stage of the product category. The findings relating to market growth suggest that products with high advantage are launched into faster growing markets than those with low advantage. When market growth is looked at from the perspective of the stages in the product category life cycle, a different pattern is observed. Of interest here is that products of high and low advantage are not launched in different stages of the product category life cycle. Taken together these findings partially support proposition 4. The issue of targeting was also examined as part of the protocol design. Table V shows clear associations between advantage levels and targeting strategy; niche strategies are most commonly employed where the levels of

Table III Product advantage and protocol: timing

Product advantage High Low N = 165 N = 155 (54%) (46%)

χ2 (df = 2)

Timing Less than 1 year (23%) 1 to 3 years (51%) More than 3 years (26%)

40% 55% 6%

5% 47% 48%

110.2***

No competitors (4%) 1-3 competitors (49%) More than 4 competitors (47%)

5% 60% 35%

3% 38% 59%

18.5***

Note: ***p < 0.001 115

The world’s path to the better mousetrap: myth or reality?

European Journal of Innovation Management

Erik Jan Hultink and Susan Hart

Volume 1 · Number 3 · 1998 · 106–122

Table IV Product advantage and protocol: market characteristics

Product advantage High Low N = 165 N = 155 (54%) (46%)

Product life cycle Introduction/growth (47%) Maturity/decline (54%) Market growth Mean rank scoreª

43% 57%

51% 49%

χ2(1) = 2.2

177.8

142.0

U = 9607.0***

Notes: ª ranked on an ordinal scale; ***p < 0.001 Table V Product advantage and protocol: targeting

Product advantage High Low N = 165 N = 155 (54%) (46%)

Targeting strategy Niche (24%) Selective (49%) Mass market (27%)

32% 47% 21%

15% 51% 34%

χ2 (df = 2)

15.7***

Note: ***p < 0.001 product advantage are high, while low advantage new products have used mass market (undifferentiated) strategies in greater proportion to their expected numbers. These results confirm proposition 5. Tactical launch (marketing mix) decisions and product advantage The analyses of the tactical launch decisions are carried out for product, promotion, pricing and distribution decisions. Product decisions at this stage include the branding policy and the product assortment breadth chosen for the launch, the analyses of which are shown in Table VI. Table VI suggests that a higher proportion of products with high advantage adopt existing brand identities, whereas a higher proportion of low advantage products are more likely to build on the overall company name for the launch. Results regarding product assortment paint a consistent picture: high advantage products tend to introduce a broader range, with low advantage products launching a smaller range. Thus, proposition 6 is generally

confirmed by the data, although, given that a large number of all products are launched with a brand name, it is prudent to note a cautious confirmation. Promotion decisions for the launch encompassed the relative level of marketing expenditures, its orientation vis-à-vis push or pull, the types and breadth of communication channels used, and the attention given by the salesforce to the new products. Results of the analyses of promotion decisions for the two levels of product advantage are shown in Table VII. The data only partially confirm proposition 7. Products with high levels of advantage are generally launched with promotion expenditures higher than those of the competition, but there are virtually no differences in the division of the expenditures between trade and end customers. Thus these results suggest that both push and pull strategies are equally often used for products of high and low advantage, but a further indication of how the products are presented to the market is provided below. Proposition 7 finds some

116

The world’s path to the better mousetrap: myth or reality?

European Journal of Innovation Management

Erik Jan Hultink and Susan Hart

Volume 1 · Number 3 · 1998 · 106–122

Table VI Product advantage and tactical launch: product decisions

Product advantage High Low N = 165 N = 155 (54%) (46%)

χ2 (df = 2)

Branding strategy New/existing brand name (80%) Company name (20%)

85% 15%

74% 25%

5.06*

Product assortment Broader than competitors (15%) Equal to competitors (60%) Narrower than competitors (25%)

22% 66% 13%

7% 55% 38%

33.8***

Notes: ***p < 0.001, * p < 0.05 Table VII Product advantage and tactical launch: promotion decisions

Product advantage High Low N = 165 N = 155 (54%) (46%)

Promotion expenditures Higher than competitors (17%) About the same (57%) Lower than competitors (27%)

21% 61% 18%

12% 53% 36%

Customer oriented expenditure Mean score

48%

48%

t = 0.15

Trade oriented expenditure Mean score

52%

52%

t = 0.15

Types of communication Salesforce promotion (91%) Trade promotion (87%) Customer promotion (70%) Personal selling (46%) Direct marketing (38%) Print advertising (26%) Public relations (25%) Radio advertising (4%) TV advertisng (5%)

92% 91% 62% 48% 45% 33% 28% 7% 6%

90% 82% 78% 42% 31% 23% 20% 2% 5%

χ2(1) = 0.61 5.64* 9.09*** 1.30 6.31* 10.6 3.01 4.22* 0.36

4.1

3.6

24% 58% 18%

19% 58% 23%

Breadth of communication mix Mean # types in mix Salesforce intensity Highly intensive (22%) Moderately intensive (58%) Non-intensive (21%) Notes: ***p < 0.001, * p < 0.05

117

χ2(2) = 14.6***

t = –3.4***

χ2(2) = 1.86

The world’s path to the better mousetrap: myth or reality?

European Journal of Innovation Management

Erik Jan Hultink and Susan Hart

Volume 1 · Number 3 · 1998 · 106–122

tentative support through the interpretation of the types of communication media used. Where products display high advantage, the media used are those associated with a push strategy such as trade promotion. Low advantage products rely to a greater extent on customer promotions, indicative of a pull strategy. In addition, the high advantage products tend to make greater use of radio coverage, although the number of companies involved is small (N= 14). The breadth of the communication mix is significantly wider for high advantage products than for products of low advantage, lending partial support for proposition 8. The final element in the promotion strategy is the “intensity of salesforce effort”. Here, no differences between the two groups of advantage were found. A majority of the sample (58 per cent) indicated a moderate salesforce effort, perhaps confirming Cooper’s (1993) view that “decisions regarding the salesforce will be straightforward”. Pricing decisions were investigated by focusing on the pricing strategy used for the launch and the relative price compared to competitive offerings. Results are shown in Table VIII. These results produce some surprising insights. The question regarding pricing strategy used, as options, the two principal strategies of the marketing literature: skimming (charge a high price to reap maximum benefits) and penetration (charge a lower price to achieve long-term market share). With this question, we offered respondents the opportunity to give

responses that reflected their views by including an “other” category. For this question, two “other” categories emerged: pricing guided by economic objectives and pricing guided by the objective of “early return on investment”. While it is logical and proper to assume that early ROI is the respondents’ view of what theory refers to as “skimming”, we are unsure of what is meant by “economic” objectives. We are sure, however, that on this issue there is a diminished correspondence between academic and practitioner terminology. We must, therefore, treat these results cautiously. That said, products with high advantage use both “skimming” and early “ROI” more frequently than products with low levels of advantage, and a far smaller proportion of products with high advantage were described as having the spurious “economic objective”. No significant differences among the two categories of product advantage were found with respect to the price of the new product compared to competitors. The data partially confirm proposition 9. Distribution was examined by detailing the expenditure level, distribution intensity, the channels used and the factors used to choose and motivate distributors. Results of the analyses are found in Table IX. There were no significant differences in the distribution strategies (intensity), nor were there variations with regard to the choice of distributors used or the factors influencing this choice. Products with high advantage, however, tended to be distributed more

Table VIII Product advantage and protocol: market characteristics

Product advantage High Low N = 165 N = 155 (54%) (46%)

Pricing strategy Skimming (16%) Penetration (50%) Other – ”economic” (29%) Other – ”early ROI” (4%)

24% 53% 17% 6%

8% 47% 42% 3%

χ2(3) = 31.3***

Price level Higher (16%) Same (42%) Lower(42%)

18% 45% 38%

15% 40% 46%

χ2(2) = 1.94

Note: ***p < 0.001 118

The world’s path to the better mousetrap: myth or reality?

European Journal of Innovation Management

Erik Jan Hultink and Susan Hart

Volume 1 · Number 3 · 1998 · 106–122

Table IX Product advantage and tactical launch: distribution decisions

Product advantage High Low N = 165 N = 155 (54%) (46%)

Distribution intensity Intensive (42%) Selective (43%) Exclusive (16%)

44% 39% 17%

39% 47% 14%

χ2(2) = 2.27

Channels used Current (72%) New/both (28%)

79% 21%

64% 36%

χ2(1) = 8.46***

Distribution expenditures Higher (14%) Same (75%) Lower (11%)

19% 75% 6%

8% 75% 16%

Distribution choice factors Willing to hold extra stock (56%) Flexibility to customers (27%) Quality in after sales service (27%) Quality order processing (33%) Motivation of distributors (targets set) Stock levels (52%) After sales service quality (31%) Flexibility in response (22%) Error free processing (33%)

55% 31% 29% 36% 48% 37% 20% 35%

χ2(2) = 13.1***

56% 22% 25% 30%

χ2(1) 0.46 2.90 0.42 0.93

56% 24% 24% 31%

χ2(1) 2.51 5.60* 0.50 0.56

Notes: ***p < 0.001, * p < 0.05 through current channels, with products of low advantage more often using new channels of distribution. It appears that, along with salesforce decisions, the options available may be limited, requiring the use of established distributors. In considering the structure of several of the industries covered by this research, where distributors are concentrated, and account for large proportions of sales to end customers, this finding is perhaps not so surprising. On the other hand, manufacturers do have some influence with regard to the amount of investment in, and motivation of, the particular distributive channel. The data suggest that where products are of high advantage, not only are the levels of investments in the distribution higher, but targets are only set for distributors in respect of the desired levels of after sales service. Thus, while proposition 10 is partially confirmed, there is very little to confirm proposition 11.

Discussion and implications The research study presented in this paper has taken as its focus the notion of product advantage, to investigate the associations of high and low levels of this phenomenon with issues recognized in the new product literature as crucial to NPD performance: protocol and tactical launch decisions. It is not, however, an alternative view of which factors contribute to success. Rather, it has sought to describe and explain the way in which product advantage relates to these other vital elements. Table X summarizes the findings for the 11 propositions. In discussing the implications, we first consider some managerial applications and then review the findings with a view to their methodological issues. Managerial implications Our data confirm most of the elements comprising the proposed associations between protocol and product advantage. We found

119

The world’s path to the better mousetrap: myth or reality?

European Journal of Innovation Management

Erik Jan Hultink and Susan Hart

Volume 1 · Number 3 · 1998 · 106–122

Table X Summary of results for the research propositions

Launch variables

Product advantage Low High

Proposition

Protocol Objectives Time to market Number of competitors Market growth PLC stage Targeting strategy

Defensive Longer time to market More competitors Lower Late Mass marketing

Growth-oriented Shorter time to market Fewer competitors Higher Early Niche/selective

Partially confirmed Confirmed Confirmed Confirmed Not confirmed Confirmed

Marketing mix Branding Product assortment Promotion expenditures Push versus pull Types of communication Mix of communication Salesforce input Pricing strategy Price level Distribution intensity Distribution expenditures Quality of distributors Current/new distributors Targets for distributors

No specific brand Narrower Lower Pull Pull oriented Fewer types mixed Non-intensive Penetration Equal or lower Intensive Lower Fewer factors considered Current distributors Fewer targets used

Brand name Broader Higher Push Push oriented More types mixed Highly intensive Skimming Higher than competitors Selective/exclusive Higher More factors considered New distributors More targets used

Partially confirmed Confirmed Confirmed Not confirmed Partially confirmed Confirmed Not confirmed Partially confirmed Not confirmed Not confirmed Confirmed Not confirmed Confirmed Partially confirmed

that growth-related objectives, early entry, high growth markets and a niche targeting strategy are all significantly associated with high levels of product advantage. Of the six propositions forwarded regarding the associations between product advantage and tactical launch decisions, a majority of their constituent elements are confirmed or partly confirmed. Products with high advantage are launched in a broader assortment, are supported by higher levels of promotion and distribution expenditures, exhibit a broader mix of promotional types, and adopt a skimming or “early ROI” pricing strategy. More products with high advantage use brand names and engage promotional techniques associated with a “push” strategy, while products of low advantage tend to use customer promotion and new distributors more. On the negative side, we expected products with high advantage to have higher price levels, which our data do not confirm. This may be explained by the maturity of the targeted markets. After all, in markets where customers already have experience with a product category, they may need extra incentives to “trade up” or “trade on”. In addition, it should be remembered that the data have

been collected at a time of a manufacturing recession in the UK, when investment in R&D was low and there was a good deal of market nervousness regarding competitive strategy. The finding that a skimming or early ROI strategy is used more by products with high advantage suggests that respondents see a clear distinction between skimming and what might be termed “over-pricing”. Research implications This research focuses on two issues that consistently prove themselves as differentiators of NPD success in bivariate tests: product advantage and launch strategy. We show a considerable number of relationships which raises questions regarding their respective and combined impact on new product success rates. Future research needs to turn attention to the need to examine inter-relationships among the factors influencing NPD performance. As mentioned earlier in the paper, we avoided, where possible, attribution bias and did not measure either the protocol or tactical launch decisions using scales anchored by positive and negative poles. We simply asked which decisions had been taken. Partly as a result of this, and partly to maintain clarity in

120

The world’s path to the better mousetrap: myth or reality?

European Journal of Innovation Management

Erik Jan Hultink and Susan Hart

Volume 1 · Number 3 · 1998 · 106–122

the measurement instrument, our data are measured mostly at the nominal level, precluding parametric and data reduction techniques. Future research in this field, particularly where the number of independent variables is large, must strive to overcome the self-fulfilling prophecies caused by attribution bias and, at the same time, devise “neutral” scales more amenable to multivariate, explanatory analysis. Given these findings, what can we say about the better mousetrap – the product with high advantage? First, they are clearly planned and outlined in their development protocol, an interesting finding in itself contrasted with Page’s (1993) observation that just over half the sample of firms in his survey claimed to have a strategy which might be likened to protocol. In this paper, we provide some detail regarding the dimensions of protocol which are associated with product advantage: the better mousetrap is well planned across the dimensions of objectives, timing of market entry and selection of target markets. When it comes to the tactical launch, better mousetraps have higher levels of both promotional and distribution expenditures, their advantage is more often ticketed with a brand name, a bigger entry splash is made with a broader assortment of products, more direct marketing and trade promotion extol its virtues to the end user, a broader mix of promotional types is used and pricing is set to provide early ROI (over-pricing is avoided). The lowadvantage mousetraps rely more on “pull” promotional techniques; perhaps customers need more persuading to buy the low-advantage mousetrap. In conclusion, we contend that companies offering the world a better mousetrap do not believe the myth that a path to its door will be beaten: the better mousetrap requires and receives a different launch treatment than more pedestrian competitors.

Biggadike, E.R. (1979), Corporate Diversification: Entry, Strategy and Performance, Harvard University Press, Cambridge, MA.

References

Booz-Allen and Hamilton (1982), New Product Management for the 1980s, Booz-Allen and Hamilton, New York, NY. Bowersox, D.J. (1990), “The strategic benefit of logistics alliances”, Harvard Business Review, July-August, pp. 36-45. Calantone, R.G. and Montoya-Weiss, M.M. (1993), “Product launch and follow on”, in Souder, W.E. and Sherman, J.D. (Eds), Managing New Technology Development, pp. 217-48. Cardozo, R. et al. (1993), “Product-market choices and growth of new business”, Journal of Product Innovation Management, Vol. 10, pp. 331-40. Choffray, J.M. and Lilien, G.L. (1984), “Strategies behind the successful industrial product launch”, Business Marketing, pp. 82-95. Choffray, J.M. and Lilien, G.L. (1986), “A decision-support system for evaluating sales prospects and launch strategies for new products”, Industrial Marketing Management, Vol. 15, pp. 75-85. Cooper, R.G. (1979), “The dimensions of industrial new product success and failure”, Journal of Marketing, Vol. 43, pp. 93-103. Cooper, R.G. (1980), “Project newprod: factors in new product success”, European Journal of Marketing, Vol. 14 No. 5/6, pp. 277-91. Cooper, R.G. (1984), “How new product strategies impact on performance”, Journal of Product Innovation Management, Vol. 1 No. 1, pp. 5-18. Cooper, R.G. (1993), Winning at New Products: Accelerating the Process from Idea to Launch, AddisonWesley, Reading, MA. Cooper, R.G. and Kleinschmidt, E.J. (1987), “New products: what separates winners from losers?”, Journal of Product Innovation Management, Vol. 4, pp. 16984. Craig, A. and Hart, S. (1992), “Where to now in new product development research?”, European Journal of Marketing, Vol. 26 No. 11, pp. 1-49. Crawford, C.M. (1984), “Protocol: new tool for product innovation”, Journal of Product Innovation Management, Vol. 2, pp. 85-91. Crawford, C.M. (1994), New Product Management, Richard D. Irwin, Homewood, IL. Curren, M.T., Folkes, V.S. and Steckel, J.H. (1992), “Explanations for successful and unsuccessful marketing decisions: the decision makers’ perspective”, Journal of Marketing, Vol. 56, pp. 18-31.

Aaker, D.A and Day, G.S. (1986), “The perils of high-growth markets”, Strategic Management Journal, Vol. 7, pp. 409-21.

Doyle, P. (1992), “Good brands are built on competitive superiority”, Journal of Product & Brand Management, Vol. 1 No. 2.

Abell, D. (1975), Competitive Marketing Strategies, Marketing Science Institute, Cambridge, MA.

Dumaine, B. (1989), “How managers can succeed through speed”, Fortune, 13 February, pp. 54-59.

Ansoff, I.H. (1965), Corporate Strategy, McGraw-Hill, New York, NY.

Edgett, S., Shipley, D. and Forbes, G. (1992), “Japanese and British companies compared: contributing factors to success and failure in NPD”, Journal of Product Innovation Management, Vol. 9, pp. 3-10.

Bain, J. (1956), Barriers to New Competition, Harvard University Press, Cambridge, MA.

121

The world’s path to the better mousetrap: myth or reality?

European Journal of Innovation Management

Erik Jan Hultink and Susan Hart

Volume 1 · Number 3 · 1998 · 106–122

Ghemawat, P. (1991), Commitment: The Dynamics of Strategy, The Free Press, New York, NY.

Moore, W.L. and Pessemier, E.A. (1993), Product Planning and Management: Designing and Delivering Value, McGraw-Hill, New York, NY.

Glazer, A. (1985), “The advantages of being first”, American Economic Review, Vol. 75, pp. 473-80.

Page, A.L. (1993), “Assessing new product development practices and performance: establishing crucial norms”, Journal of Product Innovation Management, Vol. 10 No. 4, pp. 273-91.

Golder, P.N. and Tellis, G.J. (1993), “Pioneering advantage: marketing logic or marketing legend”, Journal of Marketing Research, Vol. 30, pp. 158-70. Green, D.H. and Ryans, A.B. (1990), “Entry strategies and market performance: causal modelling of a business simulation”, Journal of Product Innovation Management, Vol. 7, pp. 45-58. Green, D.H., Barclay, D.W. and Ryans, A.B. (1995), “Entry strategy and long-term performance: conceptualization and empirical examination”, Journal of Marketing, Vol. 59, pp. 1-16.

Roberts, E.B. and Berry, C.A. (1985), “Entering new businesses: selecting strategies for success”, Sloan Management Review, Vol. 26 No. 3, pp. 3-17. Robinson, W.T. and Fornell, C. (1985), “Sources of market pioneer advantage in consumer goods industries”, Journal of Marketing Research, Vol. 22, pp. 305-17.

Hisrich, R.D. and Peters, M.P. (1991), Marketing Decisions for New and Mature Products, Macmillan Publishing Company, New York, NY. Hultink, E.J., Griffin, A., Hart, S. and Robben, H.S.J. (1997), “Industrial new product launch strategies and product development performance”, Journal of Product Innovation Management, Vol. 14 No. 4, pp. 243-57.

Rogers, E.M. (1983), Diffusion of Innovations, 3rd ed., The Free Press, New York, NY. Root, F.R. (1994), Entry Strategies for International Markets, Lexington Books, New York, NY. Rosenau, M.D. (1990), Faster New Product Development: Getting the Right Products to Market Quickly, AMACOM, New York, NY. Ryans, A.B. (1988), “Strategic market entry factors and market share achievement in Japan”, Journal of International Business Studies, Fall, pp. 389-409.

Kotler, P. (1994), Marketing Management: Analysis, Planning, Implementation and Control, 8th ed., Prentice-Hall, Englewood Cliffs, NJ.

Schmalensee, R. (1982), “Product differentiation advantages of pioneering brands”, American Economic Review, Vol. 27, pp. 349-65.

Lambkin, M. (1988), “Order of entry and performance in new markets”, Strategic Management Journal, Vol. 9, pp. 127-40. Lieberman, M.B. and Montgomery, D.B. (1988), “Firstmover advantages”, Strategic Management Journal, Vol. 9, pp. 41-58. Link, P.L. (1987), “Keys to new product success and failure”, Industrial Marketing Management, Vol. 16, pp. 109-18. Little, J.D.C. (1975), “Brandaid: a marketing mix model. Structure, implementation, calibration, and case study”, Operations Research, Vol. 23 No. 4, pp. 628-73.

Smith, P.G. and Reinersten, D.G. (1991), Developing Products in Half the Time, Van Nostrand Reinhold, New York, NY. Stigler, G. (1981), “The Economics of Information”, Journal of Political Economy, Vol. 69, pp. 213-25. Urban, G.L. and Hauser, J.R. (1993), Design and Marketing of New Products, Prentice-Hall, Englewood Cliffs, NJ. Urban, G.L., Carter, T., Gaskin, S. and Mucha, Z. (1986), “Market share rewards to pioneering brands: an empirical analysis and strategic implications”, Management Science, Vol. 32 No. 6, pp. 645-59.

MacMillan, I.C. and Day, D.L. (1987), “Corporate ventures into industrial markets: dynamics of aggressive entry”, Journal of Business Venturing, Vol. 2, pp. 29-39.

Weiner, B. (1986), An Attributional Theory of Motivation and Emotion, Springer-Verlag, New York, NY. Wensley, R. (1982), “PIMS and BCG: new horizons or false dawn?”, Strategic Management Journal, Vol. 3, pp. 147-58.

Mahajan, V. and Wind, J. (1992), “New product models: practice, shortcomings and desired improvements”, Journal of Product Innovation Management, Vol. 9, pp. 128-39.

Wind, Y. (1982), Product Policy: Concepts, Method and Strategy, Addison-Wesley, Reading, MA.

Maidique, M.A. and Zirger, B.J. (1984), “A study of success and failure in product innovation: the case of the US electronics industry”, IEEE Transactions on Engineering Management, EM-31(4), pp. 192-203.

Woodside, A.G. (1995), “Pricing a technological innovation: a case study”, Industrial Marketing Management, Vol. 24, pp. 145-50.

Miles, R.E. and Snow, C.C. (1978), Organizational Strategy, Structure and Process, McGraw-Hill, New York, NY. Montoya-Weiss, M.M. and Calantone, R.G. (1994), “Determinants of new product performance: a review and a meta-analysis”, Journal of Product Innovation Management, Vol. 11 No. 5, pp. 397-417.

Porter, M.E. (1980), Competitive Strategy, The Free Press, New York, NY.

Yoon, E. and Lilien, G.L. (1985), “New industrial product performance: the effect of market characteristics and strategy”, Journal of Product Innovation Management, Vol. 3, pp. 134-44. Yu, J. and Cooper, H. (1983), “A quantitative review of research design effects on response rates to mail questionnaires”, Journal of Marketing Research, Vol. 20, pp. 36-44.

122

1. Introduction

Innovating in reverse: services and the reverse product cycle Faïz Gallouj

The author Faïz Gallouj is an Associate Professor at the University of Lille, France. Abstract As they account for the largest share of employment and value added, services do not (or cannot) lie outside a Schumpeterian view of innovation phenomena. Of the various attempts at shedding more light on the mechanisms of innovation in service industries and firms, we consider the “reverse product cycle” to warrant special attention because of its highly thought-provoking nature and its theoretical ambition. This article has two objectives: first, to present this interesting and still neglected theoretical study, and second, to evaluate on a theoretical and empirical level the extent to which Barras’ model meets the objective of a “theory of innovation in services”.

European Journal of Innovation Management Volume 1 · Number 3 · 1998 · pp. 123–138 © MCB University Press · ISSN 1460-1060

Since they account for the largest share of employment and value added, service industries are at the heart of contemporary economies. From a Schumpeterian perspective, innovation phenomena must be at work in service firms. In an economic sector which is, admittedly, highly heterogeneous, but whose output remains largely “imperceptible”, the question, therefore, is what are the specifities of innovation in services? The different answers given to this question by economic and management literature can be schematically classed into three categories (Gallouj, 1994; Gallouj and Gallouj, 1996): (1) “Service-oriented” approaches (Gadrey et al., 1995; Gallouj, 1991; Sundbo, 1993; Van der Aa and Elfring, 1993) which emphasise the specificities of innovation in services and show particular innovation modalities in these activities (particularly high frequency of ad hoc innovation, “intangible” service trajectories as opposed to technological trajectories, etc.). In management science, and particularly, service industries marketing, these service-oriented approaches towards innovation are defined particularly by the distinction between basic services and peripheral services. Viewed from this angle, a new service corresponds to setting up a new basic service, and the extension of an existing service occurs through the addition of a new peripheral service (Eiglier and Langeard, 1987; Flipo, 1984; Jallat, 1992; Lovelock, 1992). The course taken by financial innovation (e.g. Desai and Low, 1987; Hardouin, 1973) represents, at a more subtle level (that of demand for service characteristics), this is the view of “product/service”. (2) Integrative approaches (Barcet et al., 1987; Belleflamme et al., 1986; Gallouj and Weinstein, 1997) whose aim is to reconcile goods and services in a single innovation theory. These approaches are based on a functional conception (or one in terms of characteristics) of the product,

This paper is based upon empirical materials derived from a EU financed project (DG XII, TSER programme) called SI4S (Services in innovation and innovation in services).

123

Innovating in reverse: services and the reverse product cycle

European Journal of Innovation Management

Faïz Gallouj

Volume 1 · Number 3 · 1998 · 123–138

and they propose typologies of forms of innovation valid for both goods and services. (3) Technologist approaches, by far the most numerous, which can be summarised as being concerned with the introduction of equipment and technical systems into service firms and industries.

sources of raw materials, new organization of production.

2. Presentation of the model

We consider Barras’ reverse product cycle model, on which this study is centred, to belong to the last category. We do not intend to examine the first two approaches. For a survey of the literature on innovation in services, cf. particularly Gallouj (1994), Miles et al. (1995), Gallouj and Gallouj (1996), Gallouj, F. (1997); it is, however, necessary to justify our particular interest in Barras’ model. Even though it is, as we shall see, fundamentally technologist, this model nevertheless exceeds a simple analysis of the assessment and consequences of adopting technological innovation in services, unlike the majority of the economic and management literature on this theme. Its consideration of innovation processes in services is not, as is often the case in other works, the “by-product” of other analytical priorities. Indeed, this model aims to draw up a study on the production of innovations by services themselves. Consequently, it is an interesting theoretical advance and one which we still feel to be somewhat neglected. Richard Barras’ theoretical objective is clearly set out in the title of one of his reference articles published in the journal Research Policy in 1986: “Towards a theory of innovation in services”. The objective of our article is twofold: first, to present Barras’ theory as it appears not only in Barras’ own article, but also in other earlier and later works; second, to evaluate the extent to which the model meets the objective of a “theory of innovation in services”, and what needs to be retained from it for our own perception of innovation in services. It can already be said that, in services more than elsewhere, innovation cannot be reduced, as is the case in Barras’ theory, to its technological manifestations. The definition of innovation put forward by Schumpeter at the beginning of the century, therefore, through its broadness, remains the best reference. Indeed, Schumpeter’s definition contains five categories: new goods, new production methods, new markets, new

In a series of empirical studies covering the fields of banking, insurance, accounting and local public administration, Barras (1986, 1990) highlights a product life cycle which he describes as the reverse of the cycle at work in manufacturing, where the initial phase, predominantly product innovation, is followed by a second phase, dominated by process innovation (Abernathy and Utterback, 1978). Following this model, the evolution in services will be the reverse, because the incremental phase of process innovation will be followed by radical phases of process innovation, then of product innovation, the respective purposes of which are improving the efficiency of the service, improving its quality, and conceiving a new service. The empirical assessment of this reversal of the life cycle is the core of Barras’ theory of innovation in services. Each stage of the reverse cycle is initiated by the introduction of a particular technological system: respectively, mainframe computers, minicomputers and microcomputers and networks (see Table I). Unlike some typological approaches (Lakshmanan, 1987; Pavitt, 1984; Soete and Miozzo, 1990), Barras’ model does not restrict firms to a given technological trajectory, but considers the nature of the trajectory to vary from one phase of the cycle to another. He thus places the debate on service innovations in a dynamic perspective. This model, the usefulness and limitations of which we are examining here, is without doubt the first economic study which explicitly aims to devise a theory of innovation in services in the Schumpeterian tradition. 2.1. Incremental process innovation and improvement of service efficiency The first stage of the reverse cycle (see Table I) is initiated by service-providing firms adopting back-office mainframe computers. The setting up of these mainframe computers is the opportunity for many forms of learning, resulting in incremental improvements in the service provided. These incremental innovations concern the process: they reduce the cost of the service provided without affecting its quality. Examples of these incremental process innovations are the computerisation

124

Innovating in reverse: services and the reverse product cycle

European Journal of Innovation Management

Faïz Gallouj

Volume 1 · Number 3 · 1998 · 123–138

Table I The main characteristics of the reverse cycle

Phase of the cycle

Main forms of innovation

Competitive effort

Enabling technologies

Phase I

Incremental process innovation

Improvement of service efficiency (cost decrease)

Mainframe

Phase II

Radical process innovation

Improvement of service quality

Mini and micro The computers computerised management of housing waiting lists in local public administration, on-line insurance policy quotations, ATMs

Phase III

Product innovation

New services

Networks

of insurance policy records, local government personnel records and payroll, audit techniques and internal time recording in accountancy firms. According to Pavitt’s taxonomy (1984), at this stage of the cycle, firms are “technologically dominated by suppliers”. This does not, however, mean that no interaction takes place; quite the opposite. Indeed, Barras stresses the idea of an interactive innovation process, which introduces a feedback loop between the incremental process innovation produced by the service provider, and the producer of new technologies. Indeed, incremental process innovations affect not only the technological trajectories in the field of equipment sectors (leading to other innovations in this field: e.g. superior software applications), but also the institutional structure of the service activity and the nature and volume of demand for the service (which costs less).

Examples The computerisation of insurance policy records, personnel

Home banking

Impact of technical advances on production factors Labour-saving technical advances which increase the amount of capital used Technical advances which are neutral in terms of labour, and which encourage an increase in the quantity and particularly the quality and variety of capital Technical advances which save capital whilst improving its quality

In terms of impact on production factors, the first stage of the reverse cycle is characterised by technical progress which saves labour and increases the amount of capital used. 2.2. Radical process innovation and improvement of service quality The following phase coincides with the introduction of a new generation of computer systems: mini and microcomputers. These new systems benefit from the knowledge and experience base accumulated during the previous phase, and, in turn, initiate new learning and innovation opportunities which are more centred on the front office and client satisfaction. These innovations, however, are of a more radical nature. Their purpose is no longer to lower costs, but to enhance the quality of the service provided. The best known examples of this type of innovation are automatic teller machines, other examples being the computerised management of

125

Innovating in reverse: services and the reverse product cycle

European Journal of Innovation Management

Faïz Gallouj

Volume 1 · Number 3 · 1998 · 123–138

housing waiting lists in local public administration, on-line insurance policy quotations, and computerised bookkeeping services in accountancy firms. With the acceleration of the speed at which technical change in equipment spreads, its impact on production factors changes. Indeed, technical advances have progressively less effect on the labour factor, and, instead, promote the improvement of quantity and, above all, the quality and variety of capital.

concepts from the field of manufacturing (product, life cycle). Indeed, it could be said, paradoxically, that “in services, the product is a process”. It is therefore difficult to make the distinction between product innovation and process innovation. By the same token, it becomes difficult to follow the “product” life cycle. But other problems, which to our mind are more important, must be resolved if Barras’ model is truly to constitute a theory of innovation in services. Indeed, in this model, innovation is not envisaged independently of technological possibilities. It is also a model whose field of application (in terms of fields of activity) merits discussion.

2.3. “Product” innovation The purpose of the third phase of the cycle, the product innovation phase, is to open up new markets. It is accompanied by a radical change in the service firm’s structure and strategies. The firm becomes freed from subordination to technological suppliers, and can produce its own innovations autonomously, within specialised departments, or by calling upon external service providers, particularly those belonging to knowledgeintensive business services, but where the balance of power is in the service firm’s favour. On the face of it, this phase seems to be the most interesting, as it corresponds to producing new services, rather than simply improving the efficiency or quality of existing services. However, two limitations of the model are already evident, to which we will return in more detail: (1) as it is a technologist vision, it is a restrictive view of “product” innovation, as is borne out by the examples given, such as interactive and fully-automated auditing and accounting processes in accountancy firms, complete on-line services in insurance firms, home banking, and all “home” services made possible by new information and telecommunication technologies; (2) moreover, the third phase would barely get under way, and in all cases, its progress would require an informational infrastructure to be set up.

3. A model which is limited in terms of type of activity

This “product” innovation phase has a positive effect both on output and employment. It is associated to technical advances which save capital whilst improving its quality. Barras’ model, the main points of which we have just presented, suffers from using

Two factors at the centre of Barras’ model lead to its domain of validity being questioned: (1) The enabling technologies which we have seen to be at the root of the different stages of the reverse cycle. In the main, these enabling technologies are information and telecommunications technologies. It is worth mentioning the confusion which arises (frequently in studies) between enabling technologies (mainframe computers, mini and microcomputers, networks) and the innovations made possible by these enabling technologies (which can themselves be incorporated into equipment and technical systems (e.g. automatic teller machines (ATMs) and cashpoints)). Indeed, automatic teller machines and cashpoints are not enabling technologies, but radical process innovations whose enabling technologies are particularly networks of dumb terminals. (2) The vanguard sectors which, in terms of their contribution to economic growth, are particularly dynamic. Barras would class these as financial and business and professional services, and these served as the empirical field of investigation for his model. We would like to raise the following two questions: (1) Are there not other enabling technologies, other than information and telecommunications technologies; can the validity

126

Innovating in reverse: services and the reverse product cycle

European Journal of Innovation Management

Faïz Gallouj

Volume 1 · Number 3 · 1998 · 123–138

of the model therefore not be extended to other technological systems? (2) Is the analysis valid for all vanguard services, and can its field of application be extended further than vanguard services?

Our conclusion, therefore, is that Barras’ model applies mainly to those vanguard services and other services which are most affected by technological evolution, and that it cannot be applied to most other cases. It does cover some (but, as we shall see, not all) innovation modalities in “pre-industrial” services, i.e. essentially mass informational services (banks, insurance, public administration, large audit firms). Other assessments of Barras’ model’s field of validity draw similar conclusions. Some service marketing specialists (Langeard and Eiglier, 1990) consider Barras’ model to be valid only for services with a substantial backoffice (insurance, banking), as it is dominated by back-office technologies. It is not valid for services where the service relationship (servuction) occupies a central position, as is the case in consultancy and most knowledge intensive business services. Economists interested in Solow’s paradox (Petit, 1990), consider the model’s main field of validity to be household services, since they have a strong element of self-service. An additional question, which we will only touch upon here, is whether, from the perspective of a functional approach, Barras’ thesis could be extended to service functions internal to firms, particularly when the latter adopt, as is often the case, informational technologies.

3.1. Is Barras’ model valid for all vanguard services and beyond? First, this question can be considered on the basis of concrete examples. Legal consultancy to firms has seen a relatively high growth rate in France, which has been encouraged by a number of recent institutional changes. According to Barras’ definition, as a “knowledge intensive business service”, it belongs to the vanguard services. Does it, however, display a reverse product life cycle? The answer is negative for different reasons: first, in France, this activity is currently not particularly open to information technology; the second reason (closely linked to the first) is that it is an activity which does not deal with codified information, but with expertise. This does not, however, prevent it from innovating, although its innovations do not slot neatly into the product innovation/process innovation typology. Indeed, what we see here (and we will come back to this point in section 5.1) are different forms of innovation lying outside Barras’ model: ad hoc innovations arising from the need to come up with a new solution to a client’s problem; the opening up of new legal fields through accumulation of knowledge and expertise; formalisation innovations through the introduction of methods, and through procedures of combining/dissociating existing services (architectural or recombination innovations). Another legal profession, that of notary, very rapidly embarked on the road to office computerisation. Microcomputers and fax machines were introduced under the impetus given by professional institutions. Indeed, the information processed by notaries is often more standardised. This innovation does not, however, seem to have followed Barras’ reverse cycle. Admittedly, incremental process innovations occurred, but no radical process innovations or, indeed, product innovations. Obviously, in both cases, the strength of institutional rigidities must be pointed out, as must the degree of complexity and instability of the environment and of the problems to be solved.

3.2. Can Barras’ model be applied further than informational technologies? The informational paradigm on which Barras’ model is based is certainly dominant in our economies, but other technological systems also occupy an important place in them, sometimes by merging with the previous paradigm. This is the case with technologies involved not only in storing, processing and circulating information, but also in material logistics, i.e. storing, handling and circulating materials (transport, refrigeration, cooking and cleaning technologies etc.). It is also the case with new technologies such as medical instrumentation, genetics and biotechnology, etc. Before Barras’ model could be applied to technologies other than informational technologies, it would have to be verified that the adoption of these technologies initiates a reverse cycle whose initial phases are dominated by process innovation, before the following phases of “quasi-product”

127

Innovating in reverse: services and the reverse product cycle

European Journal of Innovation Management

Faïz Gallouj

Volume 1 · Number 3 · 1998 · 123–138

innovation occur. There are indicators that such an evolution occurs: (1) There appear to be good examples of new “service-products”, new “formulas” or “concepts” in activities that utilize material transformation and logistics technologies, in distribution, for instance, or in restaurant chains (Pizza Hut, the French “Courte-Paille” or Spizza 30’...). The automation of petrol pumps in service stations, for instance, may be considered a radical process innovation similar to automatic teller machines. Moreover, the opening of sales points in these service stations, using all the techniques of the supermarket and open permanently, is related to “product” innovation in the sense used by Barras. (2) Moreover, situations can be envisaged in which the Barras cycle is based on a combinatorial adoption of information technologies and material transformation and logistics technologies. This is the case with firms like Federal Express, Chronopost and mail order companies.

systems can be seen as radical process innovations (by analogy with the automatic teller machines of banks, which are considered so by Barras). But even then, supposing that Barras’ model could have a wider field of application than the informational technologies, the fact remains that it would not necessarily accommodate all the diverse forms of innovation in service firms and industries.

An example is containerized transport. While this technology is relatively old (Ernst, 1985), it has been a source of process innovations in Barras’ sense; in the first place it improved the efficiency of transport without changing the nature of the service itself. The later standardization of container sizes and development of technologies involving the unloading cranes and their standardization have been factors in improving service quality in terms of a greater availability and so on (radical process innovation). With the introduction in recent years of information and telecommunications technologies into maritime container transport, the quality of the service has been improved so much that it is possible to speak of a “new service” in Barras’ sense. It is now possible to know at every moment to whom each container belongs, what it contains, where it is located, where it comes from and where it is going, where it should (optimally) go once empty, what kind of container it is, if it needs to be repaired and at what price, etc. (Ernest, 1985). Another example is fast food in the USA. In certain fast food restaurants, cooking and refrigeration technologies are permitting incremental process innovations (affecting the “back-office”: the central kitchen). On the other hand, computerized menu ordering

4. A model determined by technology Barras’ model’s main limitation is that it reduces the degree of variety of innovation in services, which is paradoxical on two counts: (1) Barras claims to follow Schumpeter, but Schumpeter’s definition of innovation is, in fact, broad and open, and can accommodate intangible “products” and “processes”. (2) The technological bias of the analysis means that recent important advances made in theories of financial and commercial innovation are not taken into account. We will indicate several important points of these advances here. Analyses of financial innovation based on the demand for certain characteristics have developed independently with a view to providing a theory that applies solely to financial services. They take account of facets of innovation which are not accommodated by Barras’ model. The basic hypothesis is that any financial product can be broken down into a certain number of service characteristics by which it is defined. From this point of view, any change in the topography of these characteristics, whether this involves the emergence of new characteristics or of new combinations of existing characteristics, constitutes an innovation. Hardouin (1973) formalises this analysis as follows. A monetary and financial instrument Ti can be defined a priori or a posteriori by a finite set of “n” characteristics and can therefore be written in the form of a vector with n dimensions in which the tij indicate the extent to which property j is incorporated in instrument i. Ti = (ti1....tij....tin). Thus if the instrument Ti does not have property j, tij = 0. Innovation appears in the following two cases: a variation in tij, i.e. a variation in the extent to which the existing property j is incorporated in the instrument (e.g. the instrument is more

128

Innovating in reverse: services and the reverse product cycle

European Journal of Innovation Management

Faïz Gallouj

Volume 1 · Number 3 · 1998 · 123–138

liquid), and the activation of a property that did not previously exist (transition from tij = 0 to tij ≠ 0). For example assuming that any financial service, from the simplest to the most complex, can be represented by a given combination of the following three functions or characteristics (Niehans, 1983): the exchange of current money against future money, the linking of borrowers and lenders, the making of payments in the name of a client, this author defines innovation as any new way of combining these three aspects. Niehans also introduces an interesting distinction between irreversible combinations (innovations), which he describes as “technological” but which are not limited to material technologies since they include double-entry bookkeeping, which was invented at the end of the Middle Ages, and those that are more reversible and cyclical, which he terms “adaptive innovations”. Innovations in this latter category disappear as soon as the conditions that encouraged their development have themselves disappeared. Like specialists in the financial services industry, students of retailing have sought to develop theories of innovation adapted to their particular field. The most important of them relate to the dynamic of shop formats, which are conceived of in terms of life cycle. Thus the “wheel of retailing” model (McNair, 1958) can be summarised as follows: (1) All new forms of retailing appear first in a “discount” version, i.e. outlets offer a limited range of goods and services and the main objective is to maximise sales volumes. (2) Their success causes the “wheel” to revolve as retailers gradually “trade up” by adding new products and services to the original ranges; this leads in turn to increased operating costs and higher prices. (3) This “bourgeoisification” of the retail form opens up the market for new, more “Spartan” entrants (to borrow the terms used by Tarondeau and Xardel, 1992).

of “trading up” or ways of causing the wheel to revolve by the degree of innovation in goods or service they introduce into the range; • in the “accordion theory” (Hollander, 1966) the retailing dynamic is characterised by alternation between outlets offering a wide, non-specialist range of products and those with a narrow, specialised product range.

Other analyses couched in terms of cycles, which cannot be outlined in any detail here (Gallouj, C., 1997), have extended the “wheel of retailing” model: • Goldman’s analyses (Goldman, 1975) distinguish between various possible forms

However, the cycle model in its various forms, as well as Barras’ reversed cycle model, cannot account adequately for the wide diversity of forms of innovation in the retail sector. These retail cycle models are concerned only with innovation in shop format (i.e. organisational innovation). However, even in this particular case, they are trapped within a binary logic (low/high prices; wide/restricted product range) and fail to take full account of the diversity of new shop formats and of new forms and new channels of distribution. Nor do these models take account of the following forms or areas of innovation, most of which require detailed investigation if they are to yield up their secrets: • new methods of selling (mail order, doorto-door selling …); • new products and services retailed in stores; • new processes (or new forms of organisation and operation) within the same format, whether based on the introduction of new technologies or not (within the same form of retail outlet or within the environment – customers, suppliers, other stores – of the form under consideration). Contrary to financial innovation theories and commercial innovation theories, in Barras’ model, innovation does not exist outside of technological possibilities. This is true not only of process innovations, but also of socalled “product” innovations, occurring in the third phase of the cycle. In the following section, we will attempt to illustrate, through concrete examples, the diverse forms which innovation in services can take.

5. Barras’ model put to the test: consultancy, transport and insurance Despite his affiliation to Schumpeter, we have seen that Barras adopts a reductionist view of “new services” (new “products”). The object of this section is to illustrate the multiplicity of

129

Innovating in reverse: services and the reverse product cycle

European Journal of Innovation Management

Faïz Gallouj

Volume 1 · Number 3 · 1998 · 123–138

innovation forms and trajectories in the three service industries, which display differing relationships with materiality and technology: • consultancy (the very epitome of a “pure” service); • road haulage (the tangible dimensions of which are obvious); • but, above all, insurance and financial services (which are situated somewhere between the other two service industries in terms of their relationship with materiality and technical systems, and which are one of Barras’ main fields of investigation).

hoc innovation can, to a certain extent, be reproduced, and to establish a boundary between ad hoc innovation and the ad hoc nature of all consultancy service transactions, it is necessary for the service provider, at the outset of the service and innovation, to embark upon a process of formalisation, i.e. codification of certain elements of the service, which can then be reused elsewhere, and which will help enrich the organisational memory of the firm. This is the interface which constitutes the “laboratory” where this “non-programmed” (Zaltman et al., 1973) and somewhat emergent form of innovation is conceived. Consequently, the probability of an ad hoc innovation occurring, and the quality and nature of the innovation, depend heavily on the nature and quality of the interface. Apart from the service provider’s own intrinsic elements, the “quality” of the interface depends on: • The quality of the experts from the client organisation who are involved in the interface. Indeed, these professionals partly determine how well the request is formulated and the “true need” (which often differs from the request) is (re)built (Gallouj, 1994). Furthermore, the success of the innovative solution and its assimilation by the firm depends on their ability to absorb the new ideas. • The quality of the problem raised. Original and unprecedented problems are potential sources of ad hoc innovation. That said, the innovation potential arising from curative and preventative problems (according to Kubr’s (1988) terminology), must not be underestimated. Moreover, strategic problems are often more fruitful sources of innovation than more operational or routine problems. These strategic problems are most often the subject of sparring-type interfaces (Gadrey and Gallouj, 1998). They are rarely contracted out. The terms of the analysis can thus be reversed, concluding that sparring type interfaces are more frequent sources of innovation than jobbing-type interfaces.

5.1. Consultancy Consultancy activities have much in common with research activities: their high content of “grey matter”, a similar purpose, namely “solving problems”, etc. However, paradoxically, it is difficult to study and evaluate innovation activities in this field using traditional analytical tools. Some researchers are quick to conclude that little or no innovation takes place in this type of activity, apart from, at the very most, the technical systems adopted. Consultants themselves are often split into two camps: those who underestimate their capacity for innovation and those who consider every service transaction to be an innovation, as each transaction is new and original. For our part, we do not share either of these conclusions, which prove how unfit our analytical tools are for understanding the nature of innovation in service firms. The difficulty stems in particular from the intangible and interactive nature of this type of activity. These characteristics also call into question the traditional distinction between product innovation and process innovation (Gallouj and Gallouj, 1996; Gallouj and Weinstein, 1997). Innovation does exist in this type of activity, but it can take different forms. Our empirical investigations allow three different types of innovation to be envisaged, which we will designate by the following terms: ad hoc innovation, anticipatory innovation (or new expertise field innovation) and formalisation innovation. 5.1.1 Ad hoc innovation Ad hoc innovation is the conception of a new solution to a client’s problem (which is, itself, often completely novel), with the client’s participation. It can be a solution to an organisational, strategic, legal, fiscal, social, or human problem, etc. To ensure that the ad

On a qualitative level, it can be said that opportunities for ad hoc innovation seem to increase with the size of the service provider and their client. Indeed, the multiplication of “contact” zones (i.e. interfaces) provides 130

Innovating in reverse: services and the reverse product cycle

European Journal of Innovation Management

Faïz Gallouj

Volume 1 · Number 3 · 1998 · 123–138

multiple opportunities for reciprocal learning and possibly ad hoc innovation. The existence of this interface raises two important problems: (1) In particular, it is partly responsible for limiting the extent to which the ad hoc innovation can be reproduced in its current form. In our opinion, this is what leads a number of researchers to refuse to class this particular form of mobilising expertise as an innovation. However, knowledge, experience (codifiable or not), tacit and idiosyncratic techniques stemming from practical experience, and methods used for their production and transfer can, for their part, be reproduced. (2) It poses serious problems of protection and appropriation. Indeed, as the ad hoc innovation stems from the interface, during a process of reciprocal learning, and as it has a strong tacit and contextual dimension, it is difficult to designate its inventor or owner. It is impossible, if the need arises, to ensure protection.

expertise. In this way, IT has given rise to experts in IT consultancy and IT law, etc., and ecological and environmental concerns, European construction and the opening up of Eastern countries have given rise to many “new fields of expertise”, shared by different types of service providers according to their main field of activity (technical, commercial, legal, political, etc.). These new fields of expertise, which have constituted innovations for those who anticipated these changes, are the equivalent of “product” innovations in the field of knowledge-intensive services. However, until an interface has been established with the client, anticipatory innovation will remain potential. Consequently, this presupposes some marketing and communication efforts which, in the field of consultancy, usually take the form of publications, participation in conferences, etc. As a “new field of expertise”, this form of innovation is particularly difficult to protect. Its appropriation can sometimes be facilitated by the realisation of another form of innovation: formalisation innovation.

5.1.2. Anticipatory innovation (or new expertise field innovation) This particular form of innovation could also be called a “new field of knowledge and/or expertise”. It can be considered as a particular manifestation (i.e. adapted to knowledgeintensive business services), of what Barcet et al. (1987) call functional innovation (the appearance of a new function). The ideas at the root of such an innovation can stem from the interface (i.e. direct exchanges with the client, the expression of the client’s needs), but they more generally stem from what we call the “abstract need” (i.e. the “diffuse” background noise emitted by the environment), which is complex and unclear, and not linked to any particular client (Gallouj, 1994). As the environment and the client’s needs are monitored and listened to, new needs emerge, which must be satisfied. Anticipatory innovation consists of collecting and accumulating new knowledge and expertise relevant to the “problem” or anticipated need, stemming from technological, economic, social or institutional change. Faced with particularly novel problems, i.e. problems for which there is little available expertise, the consultant will have to turn to outside experience, to similar situations. He may, in some cases, carry out research which creates genuinely new

5.1.3. Formalisation innovation Formalisation innovation consists of “putting order” into service functions, which are often vague and unformatted. They must be given form, specified and made concrete. This does not, however, mean that the desired “materiality” necessarily has to be tangible. These objectives can be met by introducing “boundaries”, reference points into the “vague” service. The components of this genuine service framework can be tangible (back-office or interface technical systems, software, etc.) or intangible: methods that constitute the scripts for the “live performance” that services are, organisations embodying service innovations, toolboxes, etc. In service activities, this “ordering” of service characteristics and functions is very often the preliminary to implementing mechanisms for architectural- or combinatory-type innovations, which consist of producing new services by combining existing services or, conversely, by dissociating an existing service. In many services, formalisation innovation constitutes a truly “natural trajectory” in the sense of the evolutionary theory of innovation. That said, we cannot, in the case of professional services, talk of industrialisation.

131

Innovating in reverse: services and the reverse product cycle

European Journal of Innovation Management

Faïz Gallouj

Volume 1 · Number 3 · 1998 · 123–138

These three forms of innovation can appear autonomously, or they can be combined and interact. Anticipatory innovation is the central element of this system of interaction, in that it is very often followed by ad hoc and/or formalisation innovations. It must not, however, be reduced to a single particular strategy, a stage in a process whose purpose is one of the other two forms of innovation. Indeed, in consultancy activities, it is just as often an autonomous and viable form of innovation. It is a particular form (one adapted to knowledge-intensive business services) of functional innovation, in the sense of Barcet et al. (1987). Anticipatory innovation and formalisation innovation can overlap, as is the case when a new field of expertise is detected and exploited at the same time as methods and tools are developed and autonomous services within the new field of expertise are differentiated. Moreover, ad hoc innovations can be a source of ideas both for improving methods (formalisation innovation) and for new fields of expertise to be detected (anticipatory innovation).

In transport, initially, the material logistic element was dominant (Djellal, 1998), so the activity naturally evolved according to the trajectory of logistical technologies (transport equipment and the handling of materials). These technologies do not seem to have behaved in exactly the same way as mainframe-based computer systems, which are described by Barras in the first stages of development of the (reverse) life cycle of financial, and, more generally, informational services. However, if back-office mainframe systems have played no role whatsoever in the development of the transport activity, the same cannot be said of decentralised computer systems. It would seem that transport can be considered as an activity involved not only in handling materials, but also in processing information, and even in processing knowledge and the service relationship. Transport is therefore an activity where several technological and innovation trajectories connect: (1) A material logistics technological trajectory (improvement of vehicles and material-handling systems). (2) An informational and communicational technological trajectory (computer systems, telecommunications). (3) An immaterial logistical trajectory (organisational knowledge and expertise, service relation, etc.).

5.2. Transport It is unlikely that Barras’ reverse cycle model can explain the dynamic of innovation in road haulage. Indeed, as we have already stressed, the model applies, above all, to activities whose most important element is the informational element. Transport can be broken down into three types of operation (as per Djellal, 1998; Gadrey, 1991): (1) those which consist of “handling” tangible objects, i.e. changing, moving or maintaining them (material logistics and transformation operations); (2) those which consist of “processing” codified information, i.e. producing, capturing, circulating it, etc. (informational logistics operations); (3) those whose main medium is the client and which consist of a direct (contact) service. This third type of operation can apply to transport firms which provide “tour operator” services, i.e. whose main operations are immaterial logistics operations, such as organising the flow of information between firms, monitoring and commitment to the client (quality, trust, guarantee) and, more generally, social innovations.

To these three trajectories must be added an organisational or “infrastructural” trajectory (the setting up of computerised roadside information points), which depends mainly on the public authorities. The first three trajectories can occur independently, when they characterise firms evolving according to only one of the trajectories (specialised trajectory). But most of the time, they connect and merge, becoming inseparable. This is particularly the case with material logistics technological trajectories and informational technological trajectories. Moreover, as we saw above, the evolution of the transport activity is characterised by two movements: (1) enrichment of each of the trajectories, i.e. innovation within each of the functional components of the activity. (2) movement from one trajectory to the other. Whether they are of a logistical or informational nature, technologies are not the only 132

Innovating in reverse: services and the reverse product cycle

European Journal of Innovation Management

Faïz Gallouj

Volume 1 · Number 3 · 1998 · 123–138

dimensions of innovation in transport. There is also room for truly intangible services (“tour operator”-type services, set-ups of transport and logistical consultancy firms).

innovations (in accordance with Barras’ model), the conception of new “product/services” such as life insurance, damage/injury insurance or assistance policies quite often entails innovative changes in computer systems. A point which Barras’ model, however, does not cover. Innovation in insurance companies and banks may take four generic forms (Gadrey and Gallouj, 1994), which are summarised in Table II. In order to facilitate analysis, these various forms of innovation are presented here separately. In reality, of course, they are

5.3. Insurance In this domain, our field studies (they were carried out together with Jean Gadrey; this section is indebted to him), highlight the importance of “product/service” forms of innovation, which are absent from the reverse cycle model. It even seems that in some situations, if computerisation brings about process Table II The main forms of innovation in insurance services

Types of innovation A: Product/service innovations

Sub-categories A1: ”Absolute” product/service innovations A2: “Relative” product/service innovations A3: Tailor-made products/services innovations 1) Adaptive tailor-made innovations

2) Fully tailor-made innovations 3) Cover for special risks B: Architectural innovations

C: Innovations based on modifications to a product or service D: Process and organizational innovations, innovations in methods and management

B1: Product/service bundling innovations B2: Product/service unbundling innovations

D1: Innovations introduced in support of product/service innovations D2: Innovations associated with a product/service that remains unchanged in terms of both formal specifications and mode of delivery D3: Innovations associated with a product/service whose formal specifications remain unchanged but whose mode of delivery, perceived quality and marketing are to be improved D4: Formal management innovations D5: Informal management innovations (ad hoc or makeshift innovation) 133

Definition New service, concept or policy for the whole market New service, concept or policy for the company concerned

Adaptation of a standard policy for a particular client through changes in pricing or the addition of certain supplementary clauses Design of a genuinely specific policy for a given client Cover for a new risk affecting only small populations Recombination of existing products/ services Isolation of one element in a product/ service Certain specifications and options are modified, leaving the basic formula unchanged Process and organisational innovation following a product/service of type A, B or C and indissociable from it Significant change in process (technology, work organisation) leaving the final service unchanged Significant change in process (technology, work organisation) leaving the product “formally” identical but improved in quality

Innovations relating to financial, actuarial, legal, HR management Differentiated from the forms outlined above by their informal nature

Innovating in reverse: services and the reverse product cycle

European Journal of Innovation Management

Faïz Gallouj

Volume 1 · Number 3 · 1998 · 123–138

frequently indissociable from each other, both in the way they are produced and in the effects they have. Thus many process and product innovations are simply two facets of the same phenomenon, and process and organisational innovations are often indissociable from each other. Furthermore, as Lasfargue (1995) rightly points out, the effects of different innovations on the firm, its specialities, skills and jobs, etc., cannot be isolated, but must be systematically comprehended.

Product/service innovations can be tailormade (A3) to suit a particular kind of risk or client. This is especially the case in group life insurance, the covering of major industrial risks and certain aspects of assistance or support services (e.g. Europ Assistance). However, a distinction must be made between three different kinds of tailor-made innovations. Adaptive tailor-made innovations adjust an existing policy to a given clientele by modifying the premiums or introducing additional clauses. This form of innovation is relatively common, particularly in the SME market. Fully tailor-made innovations are common in the insuring of risks faced by large firms, and involve the drafting of a genuinely specific policy for each client. Finally, special risk policies provide cover for risks for which there are no actuarial statistics available since they affect only very small populations. To conclude this section on product/service innovations (type A), several subtle differences can be brought into the description. The distinction between radical product/service innovations, relative product/service innovations and tailor-made product innovations may be difficult. A new product does not intrinsically belong to one particular category, but can belong to one or another according to circumstances: • Thus, an A3-type innovation (tailor-made) can become an A2-type innovation because of changes in regulations. Today, for example, insurance against computer fraud is a tailor-made product usually offered to large firms. If, tomorrow, a change in regulations made it compulsory, it could become a genuine mass product. • Similarly, tailor-made products developed in the context of brokerage relationships can be distributed by general agents, and reciprocally, products designed for the latter (by technical departments, etc.) can be offered to brokers, in return for, if need be, a “tailor-made” adaptation process. • A1 (absolute innovation) and A3 (tailormade innovation) are not necessarily in opposition. For example, the specific contract drawn up for Harley Davidson could be said to be an absolute product/service innovation (A1) since the competitors do not have such a contract; a relative product/service innovation since Harley Davidsons are motorcycles, and there are long-standing motorcycle contracts; a tailor-made innovation (A3) in the

5.3.1. Product/service innovations (type A) It is a new service, a new “formula”, a new concept, a new policy which we ambiguously tend to call a new “product”. This is a service (a formula) of contractually making available methods and competencies for managing clients’ insurance problems under conditions which are novel. The characteristic of newness must be assessed from the point of view of the user, i.e. the client. Thus, if the clients obtain the same results or guarantees, the same benefits, but the processes differ, then it is the same “product/service” (and therefore a D-type innovation). Furthermore, if “product/service” innovations can fit the existing management system, the latter often needs to be modified, sometimes innovatively. The “product/service” innovation is then accompanied by a D1-type innovation and Barras’ reverse cycle is reversed, as the product innovation (which, admittedly, is not used here in Barras’ sense) precedes the process innovation. This category partly covers what Lasfargue (1995) calls “product, service and mission innovation”. Unlike that innovation, however, it excludes innovations such as new distribution channels, which belong to category D3 (front-office innovations). Product/service innovations may be “absolute” (A1) or “relative” (A2), depending on whether the products or services involved are new to the market as a whole or just to the insurance company concerned. This latter case (imitation) is obviously more frequent than the former, particularly since product/service innovations in the insurance industry are not patentable. Nevertheless, it should be noted that the particular structural characteristics of a company (in terms of technologies, distribution channels, etc.) will very often endow the imitation with a certain degree of originality.

134

Innovating in reverse: services and the reverse product cycle

European Journal of Innovation Management

Faïz Gallouj

Volume 1 · Number 3 · 1998 · 123–138

new service sold separately or as an option (type B2).

sense that it addresses a particular target, for which it adapts a specific product. Similarly, certain innovations, which can be classed as tailor-made (A3) in the sense that they address a large firm which has a monopoly over a given activity, can also, given their originality, be radical innovations. The absolute/relative innovation distinction, however, will be reserved only for “general public” “product/service” innovations.

Individual elements such as guarantees, options and premiums may be recombined or detached from existing products. Such innovations may also involve changes in the mix of services offered (this is the case in assistance/support policies).

5.3.2. Architectural innovations (type B) This is a frequent form of innovation in services, and has been highlighted by Bressand and Nicolaïdis (1988). It must also be noted that this form of innovation is becoming more and more important in manufacturing (particularly in electronic and biotechnology industries). It is thus at the centre of an innovation model entitled “recombination model” (Foray, 1993), or architectural innovation model (Henderson and Clark, 1990). This model contrasts with the radical innovation model (which is governed by the principle of “absolute originality”) and with the incremental innovation model (governed by the principle of “first improvement”, which preserves the main characteristics of the product, but replaces some secondary characteristics with new characteristics). The recombination model can be defined in the following terms (Foray, 1993): (1) it maintains all the known characteristics of a product; (2) it recombines these different characteristics; (3) it encourages systematically reusing “components”; (4) it may add a slight difference. Architectural innovation can be divided into two different types according to the following processes: (1) the bundling or integration of services, consisting of offering formulas or contracts in which the service provider commits itself to treat a bundle of problems or operations on behalf of the client which were previously dealt with by separate formulas or contracts (type B1); (2) inversely, the separation of services by isolating a type of service or a sub-set of operations which previously formed part of an integrated service, offering it as a

5.3.3. Innovations based on modifications to the product/service (type C) In this case the core of the service, as seen from the client’s perspective, is unchanged, which is most often revealed in the fact that its “denomination” remains the same. At the same time, modifications are explicitly introduced into the formulas and the contracts. The main difference between this type of innovation and the tailor-made innovations referred to above is that innovations based on modifications are supply-driven, whereas tailor-made solutions are demand-driven. In conclusion, it must be remembered that, in the case of the insurance industry, the traditional notion of product innovation covers a wider reality than is usually imagined. Not only does it correspond to categories A, B and C, but, moreover, some of these can be split into subcategories which themselves are pertinent. 5.3.4. Process and organisational innovations, innovations in methods and management (type D) This generic category can be divided into four subcategories, each of which will be discussed as follows: (1) (Process and organisational) innovations associated with product/service innovations (type D1). In the insurance industry, product/service innovations, whether architectural in nature or based on modifications of existing products/services, almost always require changes in processes and organisation, some of which may be innovative. It is partly for this reason that it is generally considered difficult, in services, to distinguish between “product innovation” and “process innovation”. Barras’ theory is obviously reversed here, as it is the conception of a new “product” which leads to computer systems being modified. This innovation is sometimes entrusted to external service providers, such as IT service companies or manufacturers of transmission

135

Innovating in reverse: services and the reverse product cycle

European Journal of Innovation Management

Faïz Gallouj

Volume 1 · Number 3 · 1998 · 123–138

To conclude this point, we can say that: (1) Most of the process innovations taken into account by Barras’ model belong to categories D2 and D3. (2) “Product” innovations in the sense of new contracts and new services (categories A, B, C) which are the core of insurance activity are mostly not accommodated by Barras’ model, which contains a very restrictive definition of “new products”. (3) “Product” innovations (in the sense of new contracts, new services) can give rise to process innovations. In this case, Barras’ cycle is reversed in that product innovation precedes process innovation (a return to the traditional cycle).

and monitoring equipment in the case of assistance/support services. (2) (Process and organisational) innovations associated with a product/service that remains unchanged in terms of both formal specifications and mode of delivery (unchanged quality criteria) (type D2). This is what Barras calls an “incremental innovation process”. In his view, it equates to the first phase of the “reverse product cycle” that illustrates the dynamic of innovation in services. It improves the efficiency of an existing service (i.e. reduces the cost of providing it) without affecting its quality. There is a significant (non-incremental) change in the process (new technologies, new work organisation) while the final service remains unchanged. This category is a back-office innovation. (3) (Process and organisational) innovations associated with a product/service whose formal specifications remain unchanged but whose mode of delivery, perceived quality and marketing are to be improved (type D3).

6. Conclusion

Innovations of this kind involve a significant change in the process (technology, work organisation) while leaving the final product unchanged in formal terms but improved in quality. They affect the front office, i.e. they improve the quality of relationships with customers. Examples: improvements in advice and information; reduction in payment or response times; reduction in waiting times at counters. (4) Innovations in management (type D4). This category includes innovations relating to financial, actuarial, legal and HR management and, in particular, certain innovations in financial management. For example, assetsliabilities model, innovations in risk analysis methods, particularly relating to technical risks in the industrial sphere; legal innovations as applied to insurance, such as the setting up of bancassurance policies; innovations in HRM. We will also include in this category another form of innovation that might be described as informal management or makeshift innovation, in which solutions are found for certain local problems, sometimes in a secretive (even, paradoxically, disreputable) way, particularly when the innovation involves bypassing central computer systems.

Barras’ model constitutes, in our opinion, a neo-Schumpeterian theoretical synthesis of many studies in terms of “the impact of information technology and telecommunications technology on services”. Scattered materials, empirical and theoretical results have been drawn together into a synthetic and dynamic model, with its own internal coherence. Consequently, Barras has effectively succeeded in what he set out to do, i.e. devise a “theory of innovation”. But it is less a theory of innovation in services than a theory of the spread of technological innovation from manufacturing to services. In other words, the reverse product cycle model remains fundamentally technologist: innovation is not really considered to occur outside of “technological possibilities”. It does not take into account, for example, the appearance of new functions which are independent of technology. Part of the problem stems from the fact that Barras does not alter the conceptual frame of reference. The model is, indeed, concerned primarily with material technologies, and no other form of technology is taken into consideration. This bias is doubtlessly due to the choice of technology-intensive service industries as a field of investigation. Product and process innovations are then considered, but the question of the validity of transposing these concepts onto services is not addressed. Finally, the notion of life cycle, reversed or not, also merits the same investigation as regards its true scope in services. Once again, this does not mean that Barras’ model should be rejected. Indeed, his model studies, in a highly thought-provoking

136

Innovating in reverse: services and the reverse product cycle

European Journal of Innovation Management

Faïz Gallouj

Volume 1 · Number 3 · 1998 · 123–138

and unprecedented fashion, a certain (important) aspect of innovation in services. Rather than rejecting the model, we must seek to complement the model through studies which place emphasis on the least technologist aspects of this type of innovation.

Gadrey, J. and Gallouj, F. (1998), “The provider-customer interface in business and professional services”, The Service Industries Journal, Vol. 18 No. 2, April, pp. 1-15.

References

Gadrey, J., Gallouj, F. and Weinstein, O. (1995), “New modes of innovation: how services benefit industry”, International Journal of Service Industry Management, Vol. 6 No. 3, pp. 4-16. Gallouj, C. (1991), “Les formes de l’innovation dans les services de conseil”, (The forms of innovation in consultancy services), Revue de l’ Économie Industrielle, No. 57, pp. 25-45.

Abernathy, W. and Utterback, J. (1978), “Patterns of industrial innovation”, Technology Review, Vol. 80, June-July, pp. 41-7. Barcet, A., Bonamy, J. and Mayère, A. (1987), Modernisation et innovation dans les services aux entreprises (Modernisation and innovation in business services), Report for Commissariat Général du Plan. Barras, R. (1986), “Towards a theory of innovation in services”, Research Policy, Vol. 15, pp. 161-73. Barras, R. (1990), “Interactive innovation in financial and business services: the vanguard of the service revolution”, Research Policy, Vol. 19, pp. 215-37.

Gallouj, C. (1997), Innovation in French Retailing, preliminary report for EC, DG XII, TSER program, March. Gallouj, C. and Gallouj, F. (1996), “L’innovation dans les services”, (Innovation in services), Editions Economica, Paris. Gallouj, F. (1994), Economie de l’Innovation dans les Services, (Economics of innovation in services), Editions L’Harmattan, Logique Économique, Paris.

Belleflamme, C., Houard, J., Michaux, B. and Ruyssen, O. (1986), Innovation and Research and Development Process Analysis in Service Activities, IRES, Report for EC, FAST programme.

Gallouj, F. (1997), “Towards a neo-Schumpeterian theory of innovation in services”, Science and Public Policy, Vol. 24 No. 6, December, pp. 405-20.

Bressand, A. and Nicolaïdis, K. (1988), “Les services au coeur de l’économie relationnelle”, Revue d’Economie Industrielle (Services in the Heart of the Relational Economy), Vol. 43, pp. 141-63.

Gallouj, F. and Weinstein, O. (1997), “Innovation in services”, Research Policy, Vol. 26, pp. 537-66.

Desai, M. and Low, W. (1987), “Measuring the opportunity for product innovation”, in De Cecco, M. (Ed.), Changing Money: Financial Innovation in Developed Countries, Basil Blackwell. Djellal, F. (1998), Innovation in French Road Haulage, preliminary report for EC, DG XII, TSER programme. Eiglier, P. and Langeard, E. (1987), Servuction: Le Marketing des Services (Servuction: Services Marketing), McGraw-Hill.

Goldman, A. (1975), “The role of trading up in the development of retail system”, Journal of Marketing, Vol. 39 No. 1, pp. 54-62. Hardouin, J.C. (1973), “L’apparition de l’innovation financière. Contribution à l’étude de ses éléments explicatifs” (The appearance of financial innovation: a contribution to the analysis of its determinants), Complementary Thesis, University of Rennes. Henderson, R.M. and Clark, K.B. (1990), “Architectural innovation: the reconfiguration of existing product technologies and the failure of established firms”, Administrative Science Quarterly, Vol. 35 No. 1, March, pp. 9-30.

Ernst, M.L. (1985), “Electronics in commerce”, in Forester, T. (Ed.), The Information Technology Revolution, MIT Press. Flipo, J.-P. (1984), Le Management des Entreprises de Service (Service Firms Management), Les Editions d’Organisation.

Hollander, S.C. (1966), “Notes on the retail accordion”, Journal of Retailing, Vol. 42 No. 2, pp. 24-33. Jallat, F. (1992), “Le management de l’innovation dans les entreprises de services au particulier : concepts, processus et performances” (Innovation management in household services: concepts, processes and efficency), PhD Thesis, University of Aix-Marseille III.

Foray, D. (1993), Modernisation des Entreprises, Coopération Industrielle Inter et Intra-firmes et Ressources Humaines (Firms Modernisation, Inter and Intra-Firm Cooperation and Human Resources), report for the French Ministry of Research and Technology, June. Gadrey, J. (1991), “Le service n’est pas un produit : quelques implications pour l’analyse économique et pour la gestion” (The service is not a product : some consequences for economic analysis and management), Politiques et Management Public, Vol. 9 No.1, March.

Kubr, M. (1988), Management Consulting: A Guide to the Profession, ILO, Geneva.

Gadrey, J. and Gallouj, F. (1994), L’Innovation dans l’Assurance: le Cas de l’UAP (Innovation in Insurance : The Case of UAP), report for UAP and the French Ministery of research, University of Lille I, September.

Lakshmanan, T. (1987), Technological and Institutional Innovation in the Service Sector, Conference “R-D, industrial change and economic policy”, University of Karlstad, Sweden, June. Langeard, E. and Eiglier, P. (1990), “L’innovation de service (Service innovation)”, Working paper, University of Aix-Marseille III, March.

137

Innovating in reverse: services and the reverse product cycle

European Journal of Innovation Management

Faïz Gallouj

Volume 1 · Number 3 · 1998 · 123–138

Lasfargue, Y. (1995), “Quelques réflexions sur les innovations et l’évolution des métiers” (Some thoughts about innovations and jobs evolution), INTEPF seminar ,“Compétitivité des services: quel avenir pour le travail et l’emploi?” (Services competitiveness: what kind of future for work and employment?, Marcy l’Etoile, 16 November.

Pavitt, K. (1984), “Sectoral patterns of technical change: towards a taxonomy and a theory”, Research Policy, Vol. 13, pp. 343-73.

Lovelock, C. (1992), “A basic toolkit for service managers”, in Lovelock, C. (Ed.), Managing Services: Marketing, Operations, and Human Resources, Prentice-Hall International Editions. McNair, M.P. (1958), “Significant trends and developments in the post-war period”, in Smith, A.B. (Ed.), Competitive Distribution in a Free High Level Economy and its Implication for the University, University of Pittsburgh Press, Pittsburgh, pp. 1-25. Miles, I., Kastrinos, N., Flanagan, K., Bilderbeek, R., Den Hertog, P., Huntik, W. and Bouman M. (1995), Knowledge-Intensive Business Services: Users, Carriers and Sources of Innovation, Report for DG13 SPRINT-EIMS, March. Niehans, J. (1983) “Financial innovation, multinational banking, and monetary policy”, Journal of Banking and Finance, Vol. 7, pp. 537-51.

Petit, P. (1990), “Emploi, Productivité et technologies de l’information: le cas des services (Employment, productivity and information technologies: the case of service industries)”, IRIS Conference, University of Paris XI Dauphine “Competence and competitiveness”, April, p. 5. Soete, L. and Miozzo, M. (1990), Trade and Development in Services: A Technological Perspective, MERIT, The Netherlands. Sundbo, J. (1993), Innovative Networks, Technological and Public Knowledge Support Systems in Services, Roskilde University, mimeo. Tarondeau, J.C. and Xardel, D. (1992), La distribution, (Retailing), Que sais-je?, Paris, PUF. Van der Aa, W. and Elfring, T. (1993), “A typology of Innovations in service firms”, Conference on “Managing Innovations in Services”, Cardiff Business School, 5-7 April. Zaltman, G., Duncan, R. and Holbeck, J. (1973), Innovation and Organization, John Wiley & Sons, New York, NY.

138

Introduction

Investigation of the relationship between total quality and innovation: a research study involving small organisations Rodney McAdam Gren Armstrong and Brigitta Kelly The authors Rodney McAdam is Senior Lecturer at Ulster Business School, Belfast, Northern Ireland. Gren Armstrong is a Lecturer at Ulster Business School, Belfast, Northern Ireland. Brigitta Kelly is a Research Student at Ulster Business School, Belfast, Northern Ireland. Abstract Investigates how organisations can progress from total quality (TQ) to business innovation and represents the first part of an EU sponsored research programme in total quality and innovation. First, definitions and underlying assumptions are analysed which enables a definition of TQ and innovation to be derived that can accommodate a natural organisational progression in terms of implementation. Second, TQ and innovation are compared and contrasted by analysing models in each of the respective fields. The main findings were that, in general, innovation models were based more on organisational learning and appreciation of human capital than TQ models, which were based more on mechanistic process based continuous improvement. Finally, the results of a research study into innovation and total quality are presented and discussed. The study found that organisations which have a history of continuous improvement are more likely to go on and build a successful innovative culture.

European Journal of Innovation Management Volume 1 · Number 3 · 1998 · pp. 139–147 © MCB University Press · ISSN 1460-1060

For many people, both in organisations and academia, total quality (TQ) and innovation appear to have remained as separate fields or disciplines. Whilst this perception may be true in some respects, it is too general an assumption to make since both fields cover a wide spectrum of philosophy and methodologies. The possibility exists that synergy could exist between the fields to enhance business improvement/change management efforts in organisations. Kanji (1996) implies that such an alliance could be a basis for creating and sustaining competitiveness. The aim of this paper is to seek to define and explore both TQ and innovation within organisations to establish possible relationships. The objectives are to: • investigate the definitions and underlying assumptions of TQ and innovation to see if there can be a natural organisational progression in terms of implementation; • compare and contrast existing models of TQ and innovation to further understand the underlying assumptions of both fields; • to describe a research study investigating the progression from TQ to innovation (or lack of) in organisations.

Definitions Total quality – definitions Broadly speaking the TQ literature divides into two broad overlapping areas, namely holistic TQ and continuous improvement TQ (Collins, 1994). The holistic view defines TQ as an all embracive philosophy covering concepts such as business process re-engineering (BPR), benchmarking, total productive maintenance (TPM) etc. For example, Hutchins (1992) defines TQ as “everything that an organisation does which in the eyes of others determines its reputation on a comparative basis with the best alternatives”. This view of TQ includes both business efficiency and proficiency. The continuous improvement (CI) approach to TQ is defined by Hill (1994) as both a mechanism and a cultural definition of TQ. Continuous improvement is one of the fundamental tenets of the quality movement. Despite many wider definitions TQ, in the opinion of many, remains associated with mechanistic tools and techniques associated with change management. Zairi (1994b) attempts to balance the argument by

139

Investigation of the relationship between total quality and innovation

European Journal of Innovation Management

Rodney McAdam, Gren Armstrong and Brigitta Kelly

Volume 1 · Number 3 · 1998 · 139–147

describing TQ as “essentially a whole array of techniques, management principles, technologies and methodologies which are put together for the benefit of the end customer ...”. Broadly speaking, definitions of TQ have remained somewhat mechanistic and while the European Quality Model and other quality models have developed much wider frameworks these have not always been reflected in actual TQ practice. To try and advance the discussion on the relationship between TQ and innovation, the definition of TQ will be limited to that of CI for the rest of this paper. As already noted this does not limit the underlying assumptions of this definition, which are (based on Kanji and Asher, 1993): • Customer satisfaction is vital. • Internal customers are real. • All work is a process. • Measurement is essential (both hard and soft). • Teamwork is the preferred work mode. • People achieve CI. • The CI cycle is sacrosanct (plan, do, check, act). Innovation – definitions Innovation is the process of taking new ideas effectively and profitably through to satisfied customers. It is a process of continuous renewal involving the whole company and is an essential part of business strategy and every day practice (DTI, CBI and National Manufacturing Council, 1993). Innovation is also considered to be the new way of delivering quality to the customer both consistently and with economic viability in mind (Zairi, 1994a). Innovation is the successful production, assimilation and exploitation of novelty. It offers new solutions to problems and thus makes it possible to meet the needs of both the individual and society. The opposite of innovation is “ archaism and routine”. That is why innovation comes up against so many obstacles and encounters such fierce resistance. Innovation has a variety of roles: • the renewal and enlargement of the range of products and services and the associated markets; • the establishment of new methods of production, supply and distribution; • the introduction of changes in management, work organisation, and the working

conditions and skills of the workforce (European Commission, 1995). Innovation in work organisation and the exploitation of human resources, together with the capacity to anticipate techniques and trends in demand and the market, are frequently necessary preconditions for the success of the other forms of innovation. Since the life-cycle of products and services is becoming ever shorter, and generations of technologies are succeeding each other at a faster rate, firms are often under pressure to innovate as fast as possible. The time of entry into the market and the moment of introducing a new product are becoming crucial factors in competition (European Commission, 1995).

Innovation and TQ The distinction between innovation and invention is a necessary prerequisite to the understanding of innovation as a natural progression path from TQ for organisations. It is important to make the distinction between “radical” and “incremental” innovation. Radical innovations refer to product and processes that result from advances in knowledge whereas incremental innovation refers to the continual process of improvement of techniques (Mole and Elliot, 1987). Organisations tend to adopt an incremental approach to innovating because it is much easier than concentrating on radical breakthroughs (Zairi, 1994a). Incremental innovation fits best with an existing TQ system that is designed to improve various evolutionary aspects of an organisation that are of an incremental nature (Ireland and Hitt, 1997). The TQ process can reinforce incremental innovation. Therefore, while TQ can simplify or streamline a process, incremental innovation requires doing it slightly differently. Quality is doing things better; innovation is doing things differently. Both are needed. When a company is an industry leader, quality processes can produce incremental improvements that will help maintain its leadership position – for a time. However, to maintain competitive advantage over the long term, companies need to push ahead relentlessly, always innovating (Samaha, 1997). Table I shows the qualities of managers in both TQ and innovation environments.

140

Investigation of the relationship between total quality and innovation

European Journal of Innovation Management

Rodney McAdam, Gren Armstrong and Brigitta Kelly

Volume 1 · Number 3 · 1998 · 139–147

Table I Comparison of managers in a TQ and an innovation environment

Managers who foster an environment conducive to innovation do most, if not all of the following: (Samaha, 1997) 1. Encourage a learning organisation 2. Create long-term goals 3. Manage innovation proactively 4. Make innovation part of the strategy 5. Create a consistent recognition system 6. Create opportunities for cross-functional collaboration 7. Teach teamwork 8. Encourage the use of problem-solving skills 9. Teach people to assess their creative potential 10. Stop treating TQM as the only solution 11. Take a step to overcome barriers to innovation In many ways TQ can be seen as laying the foundation of a cultural environment that encourages innovation. Success depends on skills and the ability to select and implement – at the right time – the most important factors that comprise the innovation process. The ability to adapt to change and capitalise on the opportunities presented by change, i.e. innovation, is central to competitiveness. The process of TQ as outlined in Figure 1 instils a culture that is a pre-requisite for adoption of an innovation process. The organisational culture that exists in a TQ environment is a tool for innovation and for reaching strategic goals. As Zairi (1994a) remarked, TQ has “given organisations the impetus and commitment required for establishing climates of never-ending innovation or innovativeness”. In a changing world, we Figure 1 Innovation ratings for the organisations Innovation Score 5 4 3 2 1 0

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Organisation

Managers who foster a TQ environment do most if not all of the following: (Luthans and Kessler, 1993) 1. Self-image themselves as a team leader, sponsor or internal consultant 2. Cut across functional lines dealing with anyone necessary to attain quality goals 3. Change the composition of teams in response to customer needs and needed innovation 4. Act and make decisions as part of a team 5. Share and supplement information with team or anyone else 6. Becomes an expert and has significant assignments in many different functions 7. Demand quality results and loyalty not only to the organisation and one’s boss, but also to subordinates, team-mates in other departments and especially customers

must be willing to re-examine our values and update them if necessary. The TQ process creates such an environment and in many instances has been acknowledged as the vehicle by which successful innovating activity has been instigated (Zairi, 1994a).

Model comparisons and analysis To further investigate the relationship between total quality and innovation, as defined earlier, it is helpful to compare and contrast existing respective models. The European Quality model The European Business Excellence Model is a widely used total quality model. A detailed description of the European Quality Model is given in EFQM (1997). Appendix 1 gives a brief summary of the model and its criterion parts. For this paper only the “enabler” parts of the model are considered for the present analysis as the “results” parts are by definition, rather than causal factors. The Centrim innovation model The Centrim innovation model was developed by the Centre for Research in Innovation Management based at the University of Brighton. It comprises six main sections which are each sub-divided into three further sub-sections (see Appendix 2). Each of the 18 141

Investigation of the relationship between total quality and innovation

European Journal of Innovation Management

Rodney McAdam, Gren Armstrong and Brigitta Kelly

Volume 1 · Number 3 · 1998 · 139–147

sub-sections have been arrived at through extensive review of literature in the relevant area (e.g. Heraculeous and Langham, 1996; Taylor, 1995; Teal, 1996). The actual audit itself is conducted through the use of a three tier questionnaire that is administered to individuals via a diagonal slice through the organisational structure.

leadership/direction and people focus. This is in contrast to the findings of the similar analysis on the European model in Table III. The European model is predominantly associated with “traditional mechanical innovation” which has little innovation in the area of people or leadership/direction focus. In general the European model does assess the efficiency of the organisation, i.e. are they doing things right. At this point the innovation model takes on the onus of establishing whether the right thing is being done in the most up-to-date manner.

Method of comparisons The underlying assumptions of total quality and innovation, based on our current definitions in the above discussion, are summarised in Table II. All parts of the two models were compared against these assumptions and categorised as either more TQ based, more innovation based or somewhere in-between. The “in-between” category was considered acceptable as traversing from TQ to innovation is viewed as a continuum rather than a series of distinct discontinuities. When the comparisons were made with Table II, the results for each model were compared under the headings of product/ process, leadership/direction and people (Table III). These headings were sufficiently generic to both models to use as a means of summarising the comparisons.

Research methodology A research study was carried out to further investigate the relationship between TQ and innovation. Fifteen small organisations (all less than 100 employees) took part in the study. They were from Northern and Southern Ireland. The study was EU funded as part of a “cross-border” study. The organisations were selected on the following criteria: • equal numbers from Northern and Southern Ireland; • high growth organisations; • undergoing substantial change.

Analysis of the models The European Quality model comparisons are summarised under the five enabler criteria of the model and are down to “areas to address” level (see Appendix 1 for a fuller description of the European model). Below, each of the 18 key creativity questions are analysed to ascertain where their orientation lies in relation to TQ or innovation. As Table IV illustrates, the innovation model inclines towards the innovation orientation while showing an aptitude towards the

This information was obtained from Government Agencies which support small organisations and from existing university experience in working with a large number of small organisations over many years. Each organisation was evaluated in regard to innovation and continuous improvement. The Centrim innovation model (see Appendix 2) was used to evaluate the organisations in regard to innovation and continuous improvement. First, the Management Director was interviewed followed by the management team (usually an owner manager and

Table II Summary of underlying assumptions

Total quality management

Innovation

Continuous improvement Cause and effect/enablers and results Improving past performance and systems Limited empowerment Predictable outcomes Limited size step improvements Emphasis on organisational structures Reactive activity Extrapolate improving trends

Continuous renewal Importance of “meaning” and “being” New ideas, not extrapolation of past Creativity of employees unleashed Risks are more acceptable New approach, large or small Better developed knowledge systems Proactive activity Anticipate trends and techniques 142

Investigation of the relationship between total quality and innovation

European Journal of Innovation Management

Rodney McAdam, Gren Armstrong and Brigitta Kelly

Volume 1 · Number 3 · 1998 · 139–147

Table III Analysis of the European model

TQ 1a 1b 1c 1d 2a 2b 2c 2d 3a 3b 3c 3d 3e 3f 4a 4b 4c 4d 4e 5a 5b 5c 5d 5e

TQ /I

Product/ process

I

* * * * * * * * * * * * * * * * * * * * * *

*

* *

* * * * * *

*

*

*

*

*

* *

People

* * * * * * * *

*

*

Leadership/ direction

* * * * * * * * * *

*

Table IV Analysis of the Centrim innovation model

CI 1.1

TQ /I

Product/ process

I

*

1.2

Leadership/ direction

People

* *

*

1.3

*

2.1

*

2.2

*

2.3

*

3.1

*

3.2

*

3.3

* * * * * *

*

*

4.1

*

*

4.2

*

*

4.3

*

*

5.1

*

5.2

*

*

5.3

*

*

6.1

*

6.2

*

6.3

*

*

* * * 143

Investigation of the relationship between total quality and innovation

European Journal of Innovation Management

Rodney McAdam, Gren Armstrong and Brigitta Kelly

Volume 1 · Number 3 · 1998 · 139–147

two or three direct reports) using the six areas of the model (as shown in Appendix 2) and the respective questions for each section in a semi-structured format. These interviews were recorded and transcribed. Second, focus groups representing a “diagonal slice” through the organisation were asked to complete a questionnaire based on the model. Third, the data (quantitative and qualitative) were analysed to produce scores for each organisation in regard to innovation and continuous improvement. Feedback reports for the organisations were also prepared at this stage. Finally each organisation had a feedback meeting at which joint organisation/university action plans were developed.

Continuous improvement and innovation Quantitative results Each organisation was given a score for each of the questions in the questionnaire using a Likert type scale (1-6 rating, 1 is poor and 6 is excellent). The resultant scores were then averaged to give an overall innovation score for each organisation. Similarly the questions relating to continuous improvement were computed and an overall continuous improvement rating was obtained for each organisation. Figure 1 shows the overall innovation ratings for the 15 organistions. The scores range from just over 2.5 to almost 5.0. Considering that all of the organisations are in a growth phase this is a considerable range. Further analysis showed that in general almost all of the organisations scored higher on certain groups of questions and lower on others. The higher scoring questions related to: • strategy to be better than competitors; • decisions taken quickly when the need arises; • management carefully considers major decisions; • management takes responsibility for major decisions; • investing in capabilities for future success.

• inspirational management developing creativity; • appraisal in regard to developing new ideas. Thus while in general there was strong directive competitor driven leadership, it was mostly top down. There was a lack of inspirational leadership leading to innovation and creativity in the workforce. Organisations such as Org1, Org2 and Org3 demonstrated an ability to combine bottom up creativity as manifested in large numbers of new ideas generated by the workforce, with strong enabling and inspirational leadership. The lower scoring organisations such as Org13, Org14 and Org15 while demonstrating some degree of innovativeness relied on directive top management rather than employee participation in suggesting innovative ideas. The overall continuous improvement scores are shown in Figure 2. The scores range from just over 2.5 to over 4.5. Although the order of some of the organisations is reversed in comparison with Figure 1, there is a remarkable similarity. For example Org1 – Org6 retain the same order, as do Org13 – Org15. Only Org7 – Org12 have their orders reversed when comparing both figures, (the differences in scores for these organisations is very small). Those organisations which scored higher in continuous improvement (e.g. Org1, Org2, Org3) scored higher on questions which related to having mechanisms in place whereby employees could become involved in problem-solving schemes and making suggestions. There were also recognition schemes in place whereby such activity could be rewarded or recognised. Importance was also placed on measuring such activity. Figure 2 Continuous improvement ratings for the organisations Continuous Improvement Score 5 4 3 2

The lower scoring questions related to: • rewarding creative ideas; • setting targets for new ideas; • investing resource in developing creativity in the workforce;

1 0

144

1 2 3 4 5 6 8 7 10 9 12 11 13 14 15 Organisation

Investigation of the relationship between total quality and innovation

European Journal of Innovation Management

Rodney McAdam, Gren Armstrong and Brigitta Kelly

Volume 1 · Number 3 · 1998 · 139–147

Organisations which scored lower in regard to continuous improvement (e.g. Org13, Org14, Org15) while showing strong leadership were lacking in enabling the workforce to contribute to problem solving and to working in teams, especially cross functional teams. There was an overall lack of emphasis on the potential value of workforce generated ideas. Figure 3 shows a plot of the innovation scores against the continuous improvement scores ( R = 0.99, p < 0.05). The graph infers that those organisations which scored highly on innovation also scored highly on continuous improvement and vice-versa. These data would appear to support the inferred hypothesis in the earlier discussion, namely that continuous improvement can act as a solid foundation on which to build an innovative organisation. While Figure 3 does not enable a causal relationship to be confirmed it gives grounds for further investigation using the qualitative data.

developed between the management teams and the employees. This in turn led to the management team having a limited view of what employees could contribute to innovation, e.g. “the chance for employees taking part in innovation is minimal”, “all they want at the end of the week is a pay cheque” and “very few have any thoughts regarding improvement”. Management spoke of innovation in terms of customers and new machines but almost never in terms of employee creativity. These organisations also had no underlying approach to continuous improvement through which increased innovation could flourish. This was reflected in the quotes “there is no history of improvement in this organisation”, “there is no process to bring forward new ideas”. Ideas on reward and recognition were restricted to bonus systems for extra production rather than for new ideas. Employee responses were typical in these organisations. For example, “we’re working hard as it is without new ideas”. Measurement, as in monthly production figures, was seen as a way to remove “slackers”. Despite this negative approach a number of the employees recognised that their potential was under utilised reflected in comments such as “employees could contribute more if allowed to do so”. Overall, the qualitative data on low scoring organisations revealed a top down, threat based culture, in which employees were not expected to contribute beyond their job skills. Management prejudices about the limited innovative ability of employees was reinforced by a lack of continuous improvement process on which more innovative practice could be established. The qualitative data relating to those organisations which scored highly in both innovation and continuous improvement revealed some consistent factors. All of these organisations had a history of continuous improvement on which they were now “building larger innovation”. For example training associated with continuous improvement has led in many cases to increased employee knowledge of customers, competitors and markets which in turn has led to employee generated innovative product related ideas. No longer is this information seen as limited to the “capable few” at the top of the organisation. The language was one of openness and trust rather than adversarial. For example, “employees ideas are encouraged”, “I appreciate this culture as I came from an adversarial

Qualitative results Qualitative data were obtained from the interviews with the Managing Director, management team and employee focus groups, ethnographic observations and from the feedback discussions, for each organisation. These data were taped, transcribed and analysed using grounded theory and content analysis (Easterby-Smith et al., 1995). The qualitative analysis revealed a number of common issues relating to those organisations which had scored low on both innovation and continuous improvement. In these organisations there were frequent “directives” for employees to “become more innovation conscious”. Such directives were usually followed up by threats such as “someone else will take your job”. Rather than creating an innovation culture a “them and us” style of language Figure 3 Correlation of innovation and continuous improvement ratings for the organisations Continuous Improvement Score 5 4.5 4 3.5 3 2.5 2.5 3 Innovation Score

3.5

4

4.5

5

145

Investigation of the relationship between total quality and innovation

European Journal of Innovation Management

Rodney McAdam, Gren Armstrong and Brigitta Kelly

Volume 1 · Number 3 · 1998 · 139–147

type of company”, “there is open dialogue”. The increased trust generated by continuous improvement approaches, although it “takes time”, “started with slight changes that led to greater innovative ideas”. Rather than be restricted to problem-solving groups, employees contribute to innovation in new product development in an atmosphere of increased openness, trust and communication. Instead of seeing their future success as limited to new markets and technology, there are statements from the management team such as “our future success is staff development” and “new technology will be built on people development”. The management teams considered the innovation leading on from continuous improvement to be highly successful for business quoting new product development, increased quality and reduced costs as some resultant outcomes. Overall, organisations scoring highly on innovation and continuous improvement were found to have established a culture of continuous improvement on which increasingly innovative practice could be established on people development, openness and trust. Thus it is concluded that the relationship between innovation and continuous improvement shown in Figure 3 is in fact a causal one where the introduction of continuous improvement over a period of time has led to increased innovation.

build more innovative practice. Further related research and analysis is currently under way.

Conclusions and planned research programme

References Collins, P. (1994), “Approaches to quality”, TQM Magazine, Vol. 6 No. 3, pp. 39-43. DTI, CBI & National Manufacturing Council (1993), “Innovation – the best practice: the report”. Easterby-Smith, M., Thorpe, R. and Lowe, A. (1995), Management Research, Sage, London. EFQM Publication (1997), Self-Assessment and the European Quality Award Model. European Commission (1995), “Green paper on innovation”, Commission Publication. Heraculeous, L. and Langham, B. (1996), “Strategic change and organizational culture at Hay Management Consultants”, Long Range Planning, Vol. 29 No. 4, pp. 485-94. Hill, S. (1994), “From quality circles to total quality management”, in Wilkinson, A. and Willmott, H. (Eds), Making Quality Critical, London. Hutchins, D. (1992), Achieving Total Quality, Institute of Director Books. Ireland, R. and Hitt, M. (1997), “Performance strategies for high-growth entrepreneurial firms”, Frontiers of Entrepreneurial Research Conference, Babson College. Kanji, G. (1996), “Can total quality management help innovation?”, Journal of Total Quality Management, Vol. 7 No. 1, pp. 3-9. Kanji, G. and Asher (1993), “Total quality management process: a systematic approach”, Journal of Total Quality Management, Vol. 4, Suppl. Luthans, F. and Kessler, D. (1993), “Meeting the new paradigm challenges through total quality management”, Management Quarterly, Spring 1993, Vol. 34 No. 1, p. 2 (12). Mole, V. and Elliot, D. (1987), Enterprising Innovation: An Alternative Approach, Frances Pinter, London.

The study has enabled innovation to be defined as an achievable step beyond that of total quality/continuous improvement in the current context, whilst recognising there are branches of TQ and innovation that fall outside these categories. Comparison and contrast of an existing innovation type and an existing TQ type model revealed that the innovation model was mostly people, culture and leadership focused, whilst the TQ model was more process based. The research study showed that organisations that scored highly on innovation and TQ tended to have built an innovative culture on an established TQ programme of continuous improvement. Those organisations which scored lowly on innovation were frustrated by the lack of an existing TQ basis on which to

Samaha, H.R. (1997), “Overcoming the TQM barrier to innovation”, HRMagazine, Vol. 41, June, pp. 144-49 (IIL). Taylor, B. (1995), “The new strategic leadership – driving change, getting results”, Long Range Planning, Vol. 28 No. 5, pp. 71-81. Teal, T. (1996), “The human side of management”, Harvard Business Review, November-December. Zairi, M. (1994a), “Innovation or innovativeness? Results of a benchmarking study”, Total Quality Management, Vol. 5 No. 3. Zairi, M. (1994b), “TQM: what is wrong with the terminology?”, TQM Magazine, Vol. 6 No. 4, pp. 6-8.

Appendix 1. Description of the European Quality model (enablers only) The model consists of nine criterion parts. Five of these are called “enablers” and four are referred to as “results”. As explained in

146

Investigation of the relationship between total quality and innovation

European Journal of Innovation Management

Rodney McAdam, Gren Armstrong and Brigitta Kelly

Volume 1 · Number 3 · 1998 · 139–147

5. Processes 5a. How processes key to the success of the business are identified. 5b. How processes are systematically managed. 5c. How processes are reviewed and targets are set for improvement. 5d. How processes are improved using innovation and creativity. 5e. How processes are changed and the benefits evaluated.

the paper only the “enablers” are relevant in the current context. Each criterion and criterion part can be used in association with a self-assessment based scoring system (EFQM, 1997). The five enabler criteria are: 1. Leadership 1a. How leaders visibly demonstrate commitment to a culture of TQ. 1b. How leaders support improvement and involvement by providing appropriate resources and assistance. 1c. How leaders are involved with customers, suppliers and other external organisations. 1d. How leaders recognise and appreciate people’s efforts and achievements.

Appendix 2. Centrim innovation model (summarised) 1. Directing a creative business 1.1. MD support for new ideas. 1.2. Business plan showing when changes are needed. 1.3. Speed of change when superior methods are available.

2. Policy and strategy 2a. How policy and strategy are based on information which is relevant and comprehensive. 2b. How policy and strategy are developed. 2c. How policy and strategy are communicated and implemented. 2d. How policy and strategy are regularly updated and improved.

2. Developing creative capability 2.1. Individuals with creative ideas. 2.2. Capabilities needed for success. 2.3. Change efficiency.

3. People management 3a. How people resources are planned and improved. 3b. How people capabilities are sustained and developed. 3c. How people agree targets and continuously review performance. 3d. How people are involved, empowered and recognised. 3e. How people and the organisation have an effective dialogue. 3f. How people are cared for. 4. Resources 4a. How financial resources are managed. 4b. How information resources are managed. 4c. How supplier relationships and materials are managed. 4d. How buildings, equipment and other assets are managed. 4e. How technology and intellectual property are managed.

3. Building a creative culture 3.1. Encouraging staff to take initiative. 3.2. Objectives for new ideas. 3.3. Mutual support for new ideas. 4. Managing learning for new ideas 4.1. External access for new idea sources. 4.2. Availability of experienced people. 4.3. Staff up-dating with best practice learning. 5. Organising for creativity 5.1. New product introduction efficiency. 5.2. Support for new ideas from the top. 5.3. Organisational structure to support creativity. 6. Taking wise decisions 6.1. Resources to develop ideas. 6.2. Consideration of ideas before decisions are made. 6.3. Plan for development.

147

Introduction

Activity-based management and the product development process Paolo Maccarrone

The author Paolo Maccarrone is a Research Fellow at Politecnico di Milano University, Milan, Italy. Abstract Most studies on activity-based management (ABM) focus on applications in manufacturing environments. Instead, little attention has been given to the potentialities of ABM for support units, although these are widely considered to be one of the most relevant sources of inefficiencies, especially in large firms. The purpose of the paper is to illustrate how the ABM methodology can be applied to R&D activities, with particular regard to the product development process. As a matter of fact, when implementing an ABM system for R&D operations, some relevant theoretical problems arise, essentially due to the high percentage of non-routine, hardly standardisable activities. However, if adequately adapted to the characteristics of this function, ABM can be of great help in a number of issues, such as: improvement of the efficiency of the activities that constitute the process; evaluation of the economic benefits that can be gained through a redesign of processes; improvement of the effectiveness of links between product development activities; evaluation of product life-cycle costs and budgeting and control of product development activities.

European Journal of Innovation Management Volume 1 · Number 3 · 1998 · pp. 148–156 © MCB University Press · ISSN 1460-1060

According to a widely accepted definition, activity-based management (ABM) is “a discipline that focuses on the management of activities as the route to improving the value received by the customer and the profit achieved by providing this value. The discipline includes cost drive analysis, activity analysis, and performance measurement. Activity-based management draws on activitybased costing (ABC) as its major source of information” (Raffish and Turney, 1991). The fundamental distinctive element of ABM and ABC approaches is constituted by the introduction of the concept of “activity”, as the key element to understand the consumption of resources inside the organisation, as well as to improve efficiency and effectiveness of internal processes. The main steps of an ABM project can be summarised as follows: (1) definition and sharing of the objectives of the project, in strict collaboration with firm’s top management, whose commitment is of fundamental importance for the success of the project. In this phase two important decisions must be taken, concerning respectively: • the choice between an integrated vs. a parallel system: the implementation of an integrated system implies the substitution of the traditional internal accounting system of the firm, while a parallel ABM system has no interferences with the existing accounting system, which usually remains unchanged; • the choice between an incremental vs. a global approach, i.e. between the application of the methodology to the whole organisation, or to a selected part of it (the so-called “pilot project”); (2) activity accounting (or activity analysis): this phase includes all the analyses aimed at identifying activities and resources, tracing costs to activities (through appropriate resource drivers), as well as determining cost drivers for each activity; (3) process mapping: in this phase, the activities are linked in processes, according to an input-output logic; Financial support of MURST research fund “Sistema produttivo nazionale di fronte alla transizione: risposte strategiche, tecnologiche ed organizzative” is gratefully acknowledged.

148

Activity-based management

European Journal of Innovation Management

Paolo Maccarrone

Volume 1 · Number 3 · 1998 · 148–156

(4) diagnostic analysis: its aim is to define some important characteristics of the activities, which will be of great help in the subsequent steps of the analysis. In particular, activities are classified as value/not-value added. Hence, the strategic importance of the activity is analysed. Moreover, the fundamental performance measures of each activity are identified; (5) the previous phases are mostly independent from the specific objective of the ABM project. All the information generated in the previous steps can be used for many purposes, such as: • cost reduction, acting on the efficiency of single activities; • redesign and simplification of processes; • improvement of performances of processes; • support for decision making (for example, in the evaluation of benefits of AMT investments).

by Ray and Schlie (1993), although several articles discuss the limits of conventional accounting systems in R&D environments, poor results have been achieved to solve these problems, and there is little evidence of studies focusing on the use of ABM concepts for the budgeting and control of these activities. Throughout the paper, particular emphasis is given to the linkages of ABM principles with other important and well-known methodologies, both of product development theory (concurrent engineering, design for manufacturing), and of management accounting (budgeting, life-cycle costing).

The use of ABM for the improvement of the product development process

According to several authors (Brimson, 1991; Player and Keys, 1995; Turney, 1992), in the long term, the implementation of a stable, integrated ABM system should support continuous improvement, as well as contribute to a change in the strategic management of the firm. So far, ABM has been applied especially in manufacturing environments: literature shows many examples of ABM applications (most as pilot projects), aimed at improving logistic or production activities. Instead, little emphasis has been given to the potentialities of this approach for the analysis of support activities. This work discusses the use of ABM for the product development process: all companies (but especially those dealing with high technologies) face an environment characterised by accelerated technological change and shortened life-cycle. In this context, many technical and organisational solutions have been developed to improve the performance of the product development process: concurrent engineering, design for manufacturing, computer-aided design, quality functional deployment are all labels that have been introduced in the last years to characterise these new methodologies. In contrast, little attention has been given to this strategically relevant issue by management accounting researchers; as reported also

As for every kind of process, it is important to determine the key performance indicators, which can generally be classified in the three macro-categories of cost, time and quality measures. In particular, Rosenau (1989) identifies four fundamental outcome factors, namely: (1) achievement of specified performance features; (2) attainment of specific factory cost objectives; (3) meeting of development schedule; (4) staying within the project budget. As will be illustrated in the following paragraphs, ABM can be used for the improvement of all the above listed performance features. In particular, the identification of performance indexes for each activity is extremely useful for the achievement of product performance specifications, while the joint use of activity-based principles and lifecycle costing theory can facilitate the identification of design solutions that allow the attainment and respect of factory target costs. With respect to time performance, we will see how ABM can improve cross-functional integration, at the same time enabling costbenefit analyses of different architectural and organisational solutions. Finally, it is quite obvious that ABM, which is born as a management accounting methodology, has great potentialities in the field of budgeting and control of product development costs: in particular, the activity-based logic can be used for different purposes, from the design of ad hoc programs aimed at reducing the costs of

149

Activity-based management

European Journal of Innovation Management

Paolo Maccarrone

Volume 1 · Number 3 · 1998 · 148–156

the process, to the re-design of the product development budgeting and control systems.

The use of ABM for the improvement of efficiency The first action that is usually performed using the outcomes of the activity accounting analysis consists in the improvement of the efficiency of the process. This objective can be pursued through: (1) the elimination of activities; (2) the improvement of the efficiency of single activities. Whatever the objective, the first step consists in dividing the activities constituting the process into two categories: value or not-value added. This represents a very delicate task: from the one hand, this classification rises a number of conceptual problems, from the other hand it is of fundamental importance for a correct design of the interventions on the process. In general, an activity is considered as value-adding if it generates value for the customer, i.e. if the customer is willing to reward it, thus increasing the value of the firm. The problems arise in the operationalisation of this concept, with regard to: • identifying the customer (firm’s customers, the clients of the process, the users of the activity?); and • understanding the real value added. With respect to R&D, and, in particular, to product development activities, the identification of customers of each activity is simpler, given its inherent process orientation. But limiting the analysis to the specific process can be dangerous: most product development activities may apparently not add value, if we look only at the product development process, but they can nonetheless be of fundamental importance to reach market target performances. Hence, the analysis of the value added must be conducted with respect to all the stages of the life-cycle of a product. Once this classification has been completed, the attention will initially focus on not-value added activities, which should be eliminated, as far as possible. But in the great majority of cases this turns out to be a hardly possible task: as a matter of fact, at a deeper analysis not-value added activities can be grouped into three main categories:

(1) compulsory activities: their existence is linked to exogenous variables, i.e. factors that are not under control of the actors involved in the analysed process; (2) not compulsory, but not eliminable activities: these are activities that, although not imposed by specific laws, nevertheless cannot be cancelled, at least in the short term, because of consolidated procedures, whose modification implies significant investments and long implementation times, which generally exceed the objetives of the project; (3) not compulsory, totally discretional (or redundant) activities: these are the only ones that can be immediately reduced as far as possible, at least in the short term. With regard to the product development process, it can be noticed that the percentage of compulsory activities is usually not relevant, with few exceptions (as, for example, the pharmaceutical and chemical sector). On the contrary, the product development process is generally characterised by a number of both formal and informal standardised procedures, which have been developed during several years, and are deeply rooted in the organisation. Hence, the modification of these procedures implies an articulated and co-ordinated action on the organisation, aimed at removing cultural barriers, as well as at overcoming organisational inertia. The second step of the analysis consists in the improvement of the efficiency of elementary activities (both value and non-eliminable, non-value added) that came through the previous phase. Hence, the focus of the analysis shifts to accounting information: in particular, the two key measures consist in the cost of each activity, and the cost per unit of cost driver. The first is important to identify the most relevant activities (through a Pareto curve, for example); the second is fundamental to identify the areas of inefficiency, as well as to measure the effects of the implementation of programs of rationalisation. In particular, the identification of inefficient activities should be conducted comparing the actual values of cost per unit with the target values. With regard to this point, it must be underlined that the identification of target levels of efficiency for each activity is usually considered a critical issue: in this perspective, the analysis of historical data is certainly important, but the recourse to internal/external

150

Activity-based management

European Journal of Innovation Management

Paolo Maccarrone

Volume 1 · Number 3 · 1998 · 148–156

benchmarking is of fundamental importance. Moreover, the definition of “standard” values may be differentiated, according to the specific characteristics of the process: indeed, different kinds of projects may require different efforts on some phases of the development process. Once the relevant areas of inefficiency have been identified, the analysis continues with the design of the alternatives of intervention. Actions can be classified into two main streams: (1) those aimed at improving the level of output, with the same amount of resources; (2) those aimed at reducing the amount of resources (i.e. the inputs), without effects on the output of the activity. The choice between these two alternatives depends heavily on two factors: (1) the degree of saturation of the activity: if the inefficient activity is also a bottleneck of the process, the modifications should be aimed at increasing the output, thus balancing the capacity of the various elements of the process; (2) the rigidity of resources: if the improvement of efficiency leads to a reduction in the consumption of a rigid resource (i.e. a resource whose total amount cannot be modified, at least in the short term), there would be no effect on costs. In the case of activities characterised by a high percentage of fixed resources, the only way to gain immediate benefits from the application of an ABM approach consists in the redeployment of the free resources at other activities/processes inside the organisation, which, on the contrary, are under-dimensioned, and would otherwise call for new resources from outside. Another alternative consists in an accurate redesign of processes, aimed at joining the use of the same activity by different processes, in order to optimise the saturation rate of rigid resources. Example:

engineer: since this kind of resource cannot be split (and is paid on a fixed base), the time reduction does not lead to a cost reduction. If the activity would be joint by the two processes, the total time reduction would correspond to one resource unit (one of the four engineers): this resource could then be released, and then redeployed or fired.

The interrelationships between ABM, life cycle costing and the product development process Life cycle costing is an accounting methodology that became very popular in the 1980s, when the increasing competitive pressure on most markets forced firms to improve their capability to control and reduce manufacturing costs. Studying how (and why) costs were generated in operational environments, it soon became evident that many important linkages can be identified between the product development process and production activities: in particular, what emerges from these studies is that most of the production costs of a new product – up to 85 per cent (McNair et al., 1988) – are committed before the product is ever produced. This is due to the fact that the choices made in the early phases of the development process dramatically impact on the characteristics of the production process, and, hence, on the performance dimensions of the manufacturing system. This effect is well illustrated by the so-called life-cycle curve, reported in Figure 1 (Berliner and Brimson, 1988): the graph actually includes two lines, the first representing the product costs cumulated along its life-cycle (the accrued costs), the second the committed costs. Of course, the total cumulated cost is the same for both curves: what changes is the timing, i.e. the position along the x-axis. In the first case, the costs are summed up when they are incurred, while in the second curve they are

Figure 1 The two life-cycle cost curves

Suppose, for example there are two product development departments in a diversified company. The activities carried out are rather different, except for one of them. Two engineers are full-time devoted to these steps of the process, in each department. The results of a benchmarking analysis based on activity-based logics show that 25 percent of the average time could be saved with few modifications. This corresponds to the 50 percent of the time of one

151

CUMULATED COSTS

COMMITTED COSTS

ACCRUED COSTS Exploratory Product research development

Production TIME and sales (stages of the life cycle)

Activity-based management

European Journal of Innovation Management

Paolo Maccarrone

Volume 1 · Number 3 · 1998 · 148–156

traced back to that phase of the product lifecycle in which they are implicitly determined. This methodology can be considered as the equivalent, in the management accounting theory, of those product development techniques that are generally grouped under the term of “design for manufacturability” (or “design for assembly, according to the specific characteristics of the production process). Although the unquestionable importance of the concepts underlying the life-cycle cost theory, its popularity among firms is not as high as that of corresponding technical tools: this is to be ascribed mainly to its implementation problems, such as: • the low measurability of data: this is true both in the planning phase (the most important, because the life-cycle cost curve is used to quantify the effects of alternative design options on the whole life-cycle cost of a product), and in the ex post phase (when the budgeted curve of incurred costs is compared with the real curve, to check for variances, as well as for learning objectives); • the timeliness of the tool: as a matter of fact, this technique requires a huge amount of time to be correctly and carefully implemented, which may not be compatible with time constraints, especially in environments characterised by time-based competition.

level, while fixed costs remain unchanged (or are slightly increased), except for some particular entries, which are examined separately (for example, new depreciation costs, as a consequence of new equipment). It is quite evident that, using such kind of accounting systems, it is not easy to estimate the change in production costs caused by change in product specifications, even if the item is already being produced, unless its effect is limited to direct costs (raw materials or components, typically). For example, it is generally assumed that a reduction in the number of parts of a product leads to a reduction in material handling operations, and, hence, in production costs. The problem consists in estimating this economic benefit: as a matter of fact, most production centres need material handling, but its cost is joint with that of many other activities that are performed inside each cost centre. Instead, using an activity-based management system in production departments we can easily measure the effect of a reduction of material handling on resource drivers, and, hence, on the cost of each resource. Moreover, the adoption of an activitybased approach can enhance the quality of life-cycle costing estimates for a new product: as a matter of fact, in literature on life-cycle costing three methodologies are generally suggested for the estimation of the costs of a new product (Blanchard, 1979): (1) by analogy with other products, produced by the firm itself or by competitors (in the latter case it is necessary to have recourse to benchmarking, of course); (2) through parametric models, which relate the cost of the product to its main design parameters (e.g. weight, performance targets, dimensions); (3) through industrial engineering.

The ABM methodology can be of great help to overcome these implementation problems, thus enabling a better exploitation of the potentialities of life-cycle costing. As a matter of fact, the introduction of an activity-based approach in accounting systems allows firm managers and controllers to better understand the effects on costs of the alternative product and process design options; in traditional firm accounting systems, indirect costs are generally added up in some cost pools (usually represented by responsibility centres, like production departments, job-shops, etc.), and then allocated among products using some base of allocation (one for each pool), which is generally linked (proportionally) to production volumes. Poor attention is given to indirect costs also in the budgeting process: for their estimation, indirect costs are divided into two main categories, including respectively fixed and variable entries. Variable costs are then calculated according to the budgeted production

By splitting the production process into its elementary activities, ABM enables a more accurate analysis of differences between the new and the existing products. In particular, if the production process of the new product is very similar to that of an existing one, or, anyway, if it can be seen as a combination of activities that are already carried out in the production system, it is possible to recourse to a parametric approach, where the parametrisation is made with respect to the most important production activities.

152

Activity-based management

European Journal of Innovation Management

Paolo Maccarrone

Volume 1 · Number 3 · 1998 · 148–156

So far, we have seen how the use of ABM in production environments can facilitate the implementation of life-cycle costing principles in the phase of design and engineering of a new product. But in the previous paragraphs we have seen some issues concerning the application of an ABM approach for product development activities. As a matter of fact, the larger the extension of ABM methodology, the greater will be the benefits for processoriented techniques. In particular, benefits essentially consist in: • a higher degree of precision in the estimation of costs along the whole life-cycle of the product; • a deeper comprehension of trade-off links between related activities in different stages of the life-cycle.

manager can immediately identify its possible internal sources. This can lead to anticipated and quicker problem-solving cycles, with positive effects on the overall duration of the product development process; • enhance the economic-financial evaluation of different architectural solutions for cross-functional integration. Each organisational solution that can be designed to improve cross-functional communication is characterised by a cost, which should be more than offset by economic benefits in subsequent phases of the product development process, or even in the whole lifecycle of the product (for example, one of the most important benefits of concurrent engineering consists in the reduction of time to market, which is considered a key success factor in many competitive environments, since it leads to important advantages in terms of market share and premium prices in the first stages of the market life-cycle). The implementation of any organisational tool or procedure should be then preceded by an economic evaluation: the ABM methodology can be very useful in estimating the changes in the costs of the product development process deriving from this kind of process reengineering interventions.

ABM and cross-functional integration One of the most important issues in the product development management theory is represented by cross-functional integration, which is considered as essential to achieve outstanding results in the development process, with regard to time, cost, and quality performances (Clark and Wheelwright, 1993). Unfortunately, cross-functionality is often considered just a problem of communication between different functions; on the contrary, it implies important changes in the way detailed work is done in each function. Crossfunctional integration occurs when design engineers, process engineers and marketers work together to solve joint problems in the development of a new product. With respect to this issue, ABM can: • enhance the management of crossfunctional linkages: the identification of elementary activities, of their performance measures, as well as of their internal/external clients, facilitates the links between upstream and downstream groups. First, the individuation of internal clients of each activity, and of characteristics of target output, implicitly forces the suppliers to communicate with the users, in order to understand their needs. Hence, upstream managers become knowledgeable of downstream constraints and capabilities, thus enhancing their ability to predict the consequences of their choices. Second, each time a problem occurs in some downstream activity, through the analysis of flowdiagrams and activity owners the project

The use of activity-based budgeting for the control of product development activities Great benefits for the planning and control of R&D activities can derive from the application of activity-based principles. The budgeting of support activities has always been disregarded in literature on management accounting: traditionally, two main approaches are proposed, namely: (1) the incremental approach: the budget of a specific function (like R&D or engineering, for example) is determined starting from actual costs incurred in the last year, and correcting them through a coefficient to include the effect of inflation, as well as of significant changes in the level of activities of the function (for example, in a phase of fast growth, the coefficient should be over the unity); (2) the zero based budget approach: according to this approach, the budget allocated to a function is not a function of the

153

Activity-based management

European Journal of Innovation Management

Paolo Maccarrone

Volume 1 · Number 3 · 1998 · 148–156

• by spending categories; • by project/processes.

historical levels of cost, but is determined as follows: • first, the minimum amount of resources (those necessary for the surviving of the function) is fixed; • then, a variable number of “slots” of resources is added, each of them corresponding to a specific project/program. The choice of which programs to include, as well as of the total budget allocated to the function, is up to functional and corporate management, with the latter usually holding the final decision.

This enables a more accurate budgeting of product development projects, as well as of the organisational units that are involved in the various phases of product development. The most critical aspect in the implementation of an ABB approach is represented by the availability of the budgeted/standard cost per unit of activity. This issue raises two basic questions, concerning respectively: (1) the standardization of R&D activities; (2) the computation of standard values.

The ZBB is characterised by a higher precision, since it eliminates some important problems of the incremental approach (first of all, the amplification of errors, which occurs each time the actual level of expenses of a function include costs due to unforeseen, not repetitive, events which occurred during the year). The main limit of this methodology consists in its onerousness, for the long time required, and the difficulties in the phase of data gathering. But, even more important, the ZBB is still characterised by a low measurability of data: indeed, the main problem controllers face in the implementation of this technique is the quantification of the budget to be allocated to each specific program. This due to the characteristics of accounting systems, which hamper a detailed analysis of how resources are used for different purposes. A great step forward can be made by adopting an activity-based budgeting approach, which represents the natural extension of activity-based concepts in the field of budgeting. Hence, activity-based budgeting focuses on cost of activities that are necessary to produce a target output. The three key steps in activity-based budgeting are: (1) to determine which activities are necessary on the base of target outputs, as well as the demand for each activity; (2) to determine the cost of each activity, on the base of budgeted (or “standard”) cost per unit of activity; (3) to compute the budgeted cost of each function, by aggregating the cost of activities that relate to the same functional area. The great advantage of using ABB is that costs can be easily re-aggregated: • by functional areas;

With respect to the first point, it is generally assumed that R&D activities can hardly be codified and standardised: while this can be true for basic, exploratory research, product innovation literature is rich with exhaustive listings and flow diagrams of activities that make up the product development process. For the estimation of budget unitary values, the only solution consists in building a complete database wherein to store the information coming from the application of the ABM principles to the product development process. The database should essentially include one record for each activity, and the associated cost per unit of output (the cost per unit of cost driver, according to the terminology used in this work). The up-dating of the database should focus on two aspects: (1) the modification of cost per unit of activities that have already been registered; (2) the insertion of new activities, related to new kind of products, or new organisational schemes. This has important consequences on the way in which ABM should be implemented in a firm: as illustrated in paragraph 1, in the literature three ways of applying ABM are usually illustrated, according to the specific objectives pursued. If the aim is to extend the activity-based approach to the phase of budgeting, a permanent accounting system should be developed: otherwise, the definition of accurate standard values would be extremely difficult, if not impossible. But the benefits of the ABB approach clearly emerge also in the phase of control: as a matter of fact, literature on management control systems is rather “vague” also on this issue. The traditional way of controlling discretional cost pools consists in checking the respect of allocated budgets: in some cases,

154

Activity-based management

European Journal of Innovation Management

Paolo Maccarrone

Volume 1 · Number 3 · 1998 · 148–156

some non-financial, quantitative output measures are identified for each unit, which are then related to the financial input (the allocated budget, or the actual expenses), so to determine a sort of “productivity” ratio. The ABB approach enables a much more careful evaluation of the efficiency of organisational units involved in the development process. In particular, the typical concepts that guide the traditional variance analysis can be resumed also for non-manufacturing activities. The traditional analysis of cost variances is based on a multi-level structure: at the first level, two basic variances are identified, namely the volume variance and the efficiency variance. To do this, a particular budget is introduced, intermediate between the original budget and the actual cost, which is called flexible budget, determined on the base of the actual volume of production and the standard efficiency level. In non-manufacturing environments (and especially in some support units), this kind of approach generally cannot be applied, due to the particular nature of the activities performed, since it is very difficult to establish a relationship between the inputs and the output of these units. The ABB methodology can yield a great contribution to overcome this limit: by destructuring the process in elementary activities, and identifying a cost driver for each activity, it is possible to determine a flexible budget, where the flexible parameter is represented by the cost driver. Hence, through this flexible budget, built using the standard values of cost per unit and the actual level of cost driver, we can determine once again a volume and an efficiency variance. But the analysis can go further: as a matter of fact, the standard values of cost per unit can be related to the target performances. In other terms, the cost per unit of cost driver can be directly linked to the values of performance indexes that have been identified for each activity. Hence, the efficiency variance can be split into two addenda: a performance variance, and a yield variance. The intermediate budget is determined by modifying the standard costs per unit, according to the actual value of performance measures: the performance variance includes the possible changes of required performance levels, while the yield variance is a measure of the real efficiency in the use of resources for that activity. For example:

Consider the phase of detailed design of a new component of a gear system. One of the main outputs of this phase consists in a variable number of CAD files, each of them containing morphological details of the new item. The ABB system allows the estimation of costs for this particular activity (elaboration of graphic files). In particular, the budgeted costs were the following: • Standard unitary cost (cost/file): $5,000 /document. • Number of scheduled documents (for that specific kind of project): 20. • Budgeted cost: $100,000 . Actual cost was $135,000, and 24 documents were elaborated. Flexible budget can be calculated as follows: 5,000 × 24 = $120,000 . The unit responsible for this activity seems to have been inefficient, also considering the different number of documents elaborated. But at a deeper analysis, what emerges is that the required specifications for each file have been changed in the early phases of the development process, as a consequence of an innovation introduced in the engineering department. The analysis of activities shows that the standard cost per unit rises to $6,000 /document. The current flexible budget is then: 6,000 × 24 = $144,000 , and the yield variance is: 135,000 – 144,000 = –$9,000 (favourable). Hence, the unit has been more efficient than expected.

The use of the activity-based logic can be very useful also in some typical decision-making issues of the budgeting phase: in particular, it can help managers in cost-benefits analysis in the phase of allocation of resources to the various organisational units involved in product development activities. As a matter of fact, one of the main outcomes of the activity analysis consists in the identification of performance measures for each activity, according to an approach which is also typical of business process reengineering projects (Hammer and Champy, 1993): these can be related to the specific output of the activity, as well as to the overall output of the product development process. Hence, once the fundamental performance indicators for the whole process have been defined, it is possible to determine: (1) the extent to what each activity contributes to the achievement of each dimension of performance (i.e. how much each activity influences each performance measure); (2) the positioning of each activity, with respect to those “local” performance indicators that are related to the global target performances of the process.

155

Activity-based management

European Journal of Innovation Management

Paolo Maccarrone

Volume 1 · Number 3 · 1998 · 148–156

This model is of a great help to understand: (1) which are the “critical” activities, i.e. those characterised by a bad positioning, with respect to the whole set of performance measures; (2) which are the activities to focus on, in order to improve a specific performance of the product development process. Once the activities on which to act have been identified, the next step consists in the identification of the possible interventions. Each of them consists in a number of changes in the way the activity is performed: the activity accounting enables controllers to estimate the impact of these actions on the use of resources, and, then, to translate it in economic-financial terms (i.e. investments needed to implement the modifications, or changes in operating costs of the organisational units involved). Hence, in the phase of budgeting, when the choice of the actions to be performed have been done, it will be easier to include the related expenses (investments/costs) in the functional budgets, with positive effects on the precision of the resource allocation process. Moreover, the joint use of activity accounting and of performance measures can assist managers in the cost-benefit analysis of alternative programmes. As already explained in the previous paragraphs, the different interventions on the activities constituting the product development process should be aimed at improving one or more dimensions of its performance. A further step of the analysis consists then in the evaluation the economic effects of these improvements: from a theoretical point of view, the need of improving a particular performance must derive from the perception of a weakness in the internal configuration of the firm, which has (or may have) consequences on its competitiveness. Hence, in the ultimate analysis the improvement of a performance of the product development process should be converted into an economic benefit, basically in terms of reduction of costs, or increased revenues. Of course, this is generally not a simple task, since we can have both short- and long-term effects, and many of them are

“intangible” in nature, or hardly quantifiable. Anyway, in the last years some important methodologies have been developed to deal with quali/quantitative data in decision making (Walls, 1996); hence, the recourse to these tools should allow this problem to be overcome. Once the economic benefits of a given action have been estimated, it is possible for managers to proceed to a cost-benefit analysis of different alternative programs, which impact on different activities, and/or on different performances, thus enhancing the quality and the effectiveness of the resource allocation process.

References Berliner, C. and Brimson, J.A. (Eds) (1988), Cost Management for Today’s Advanced Manufacturing, Harvard Business School Press, Boston. Blanchard, B.S. (1979), Design and Manage Life Cycle Cost, M/A Press, Portland. Brimson, J.A. (1991), Activity Accounting: An ActivityBased Cost Approach, John Wiley & Sons, New York, NY. Clark, K.B. and Wheelwright, S.C. (1993), Managing New Product and Process Development, The Free Press, New York, NY. Hammer, M. and Champy, J. (1993), Reengineering the Corporation, HarperCollins Publishers, London. McNair, C.J., Mosconi, W. and Norris, T. (1988), Meeting the Technology Challenge: Cost Accounting in a JIT Environment, Montvale, National Association of Accountants, NJ. Player, R.S. and Keys, D.E. (1995), “Lessons from the ABM battlefield: moving from pilot to mainstream”, Journal of Cost Management, Fall. Raffish, N. and Turney, P.B.B. (Eds) (1991), “Glossary of activity-based management”, Journal of Cost Management, Fall. Ray, M.R. and Schlie, T.W. (1993), “Activity-based management of innovation and R&D operations”, The Journal of Cost Management, Winter. Rosenau, M.D. Jr (1989), “Schedule emphasis of new product development personnel”, Journal of Product Innovation Management, December. Turney, P.B. (1992), “ABM puts ABC information to work”, Management Accounting, January. Walls, M.R. (1996), “Integrating business strategy and capital allocation: an application of multi-objective decision making”, The Engineering Economist, Vol. 40 No. 3.

156