CIMA Publishing is an imprint of Elsevier Linacre House, Jordan Hill, Oxford OX2 8DP, UK 30 Corporate Drive, Suite 400, Burlington, MA 01803, USA First edition 2009 Copyright © 2009 Elizabeth Daniel, Andrew Myers and Keith Dixon. Published by Elsevier Ltd. All rights reserved. The right of Elizabeth Daniel, Andrew Myers and Keith Dixon to be identified as the authors of this work has been asserted in accordance with the Copyright, Designs and Patents Act 1988 No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means electronic, mechanical, photocopying, recording or otherwise without the prior written permission of the publisher Permissions may be sought directly from Elsevier’s Science & Technology Rights Department in Oxford, UK: phone (144) (0) 1865 843830; fax (144) (0) 1865 853333; email:
[email protected]. Alternatively you can submit your request online by visiting the Elsevier web site at http://elsevier.com/locate/permissions, and selecting Obtaining permission to use Elsevier material Notice No responsibility is assumed by the publisher for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions or ideas contained in the material herein. British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloguing in Publication Data A catalog record for this book is available from the Library of Congress ISBN–13: 978-1-85617-682-8
For information on all CIMA Publishing publications visit our web site at books.elsevier.com Typeset by Thomson digital, Noida, India Printed and bound in Great Britain 07 08 09 10 10 9 8 7 6 5 4 3 2 1
I think everything has a finite lifetime because if you keep things the same for too long, people stop seeing what you are trying to do. The fact that you have shaken things up gives you an opportunity to communicate in a fresh way. It gives you an opportunity to train people in a fresh way, to design your incentive schemes in a new way… . So whilst I wouldn’t recommend chucking everything out every 12 months, I think sticking to the same thing almost as a religion – in my experience – causes things to fall by the wayside [Case study interviewee]
Researchers' Contact Details Elizabeth Daniel Open University Business School, United Kingdom, Email:
[email protected] Andrew Myers Open University Business School, United Kingdom
Researchers Contact Details
Keith Dixon
viii
University of Canterbury, New Zealand
Acknowledgements The authors would like to thank the Research Foundation at the Chartered Institute of Management Accountants (CIMA) for funding the research described in this report. We would also like to thank the managers within the case study organisations who gave their time and shared their experiences of adopting and implementing managerial innovations and the reviewers who provided valuable feedback on an earlier draft of this report.
Acknowledgements ix
Don’t Blame the Tools The Adoption and Implementation of Managerial Innovations
1.
Executive Summary
The people governing and managing organisations are often criticised for being ‘faddish’ in their use of new management ideas or innovations. Many have been seen to easily fall into the trap of adopting the latest new idea or concept that is ‘flavour of the month’. It is not always obvious, however, that this new idea will be of value to the organisation. This is particularly the case when many of the new ideas often involve considerable time and capital investment for implementation. Many ideas seem to be just a re-labelling of old ideas that were tried in the past and were often seen to fail to deliver. What is perhaps more troubling is that the development of new management ideas seems to be on the increase and that these newer ideas are often aimed organisation-wide, making them potentially even more costly and time consuming for implementation. A concern over the issue of management fashions or fads led CIMA to launch a research initiative in 2005 entitled ‘The Organisation and Adoption of Management Ideas, Tools and Practices’. This report presents the findings of one of the three projects funded under this initiative. The project commenced in April 2006 and completed in July 2007. The project sought to look at the link between the decision to adopt a new idea or tool and its subsequent implementation. To our knowledge this link has not been studied previously and hence our project was exploratory in nature. In particular, it explored factors that could potentially influence the decision to adopt and the impact these would have on implementation. It also sought to identify practices that could help the newer organisation-wide innovations be successfully embedded in an organisation.
2
1.1.
The Adoption and Implementation of Managerial Innovations
Research Approach
The well-established balanced scorecard (BSC) was selected as one idea or innovation to study. To provide a comparison, the relatively newer idea of a programme management office (PMO) was also selected. As organisations carry out more, and increasingly complex, projects, such offices are being established to coordinate and centrally manage these projects. The exploratory nature of the study suggested a case study approach, and four in-depth case studies of organisations were undertaken. Three organisations adopted a BSC and one organisation had adopted a PMO. The three organisations that had decided to adopt the scorecard provided an in-depth range of experiences: one organisation was still in the process of implementing the scorecard, one had implemented it and was using it, whilst the third organisation had implemented the scorecard, used it for a number of years and had ceased using it and moved on to a new idea.
1.2.
Findings and Recommendations
■ Managerial innovations are a natural consequence of a dynamic business environment. The findings of this research show that rather than simply following fashions, the adoption of managerial innovations is a response to an external or internal trigger. The adoption of such innovations should not therefore be seen as a weakness to be addressed, but as a natural consequence of a dynamic business environment. However, it is also recognised that the introduction of organisationwide innovations, in particular, can be very costly and disruptive; and hence, organisations should select innovations carefully and test their usefulness before adoption, as discussed later. ■ The need for the innovation should be separated from the chosen solution.The people governing and managing organisations should seek to separate the event or problem that triggers the need, or perceived need, for the adoption of an innovation and the actual adoption decision. Whilst there is likely to be strong support for addressing the trigger, there may not be such strong support for the solution. ■ A champion is vital. This research underlines the importance of an influential individual or group in supporting implementation and following it through. However, it also finds that in the cases studied, hiring an individual from outside to champion the innovation may be less effective than finding a senior level champion from within the organisation. ■ A robust business case is needed and the innovation should be tested before adoption. Whilst the influence of a significant individual or group is important, preparing a robust business case for the innovation, including forecast benefits and
Don’t Blame the Tools
3
costs and pilot testing the innovation are equally important. Preparing a business case that is open to review will encourage the evaluation of alternative innovations and will also prevent the adoption appearing like a ‘pet project’ of an individual or a knee-jerk reaction to the recommendation of management consultants or other groups. Accounting specialists have an important role to play in preparing and scrutinising such business cases, alongside business managers and other specialist staff. ■ If possible, include the innovation in a wider programme of change. Linking the introduction of the innovation to a wider programme of change was also shown to be effective. The rationale for such wider programmes may be more easily understood and supported by staff, for example a ‘customer focus programme’, can be more easily understood and adopted by staff than the less obvious ‘need to improve performance management’. ■ Actively manage the end game. The discontinuation of innovations should be actively managed, rather than simply letting them dissipate or fall into disuse. The transition in moving away from an innovation that was used in the past to the adoption of a new management innovation should be communicated throughout the organisation, highlighting that the organisation faces new events or challenges and that the previous innovation was no longer suitable to meet these challenges. The process identifying a true champion of the next innovation, developing a robust business case and piloting or testing that next innovation should then be undertaken. Other approaches that appear to aid implementation include ■ adapting the innovation to fit the organisation, ■ training tailored to the needs of individuals or groups. Approaches that slow implementation include ■ allowing a phased approach to be slowed by different factions within the organisation, ■ introducing an innovation where existing ways of working are adequate, ■ providing benefits to the organisation but not to the individuals that must use the innovation. Whilst the case study method employed has provided a rich picture of the adoption and implementation process within the organisations studied, the findings are recognised as exploratory in nature. Further studies that can confirm and extend the study findings are suggested.
4
The Adoption and Implementation of Managerial Innovations
2.
Introduction
. . .management fashions are not cosmetic and trivial . . .[they] shape the management techniques that thousands of managers look to in order to cope with extremely important and complex managerial problems. (Abrahamson, 1996, p. 279) Implementation and use of most management tools is expensive, sometimes costing firms tens of millions of dollars to implement. (Rigby, 2001, p. 139)
2.1.
Managerial Innovations
A prominent responsibility of managers in all types of organisations is to seek continually to improve the profitability, service or other performance criteria of their own organisations, or the parts of organisations in which they are particularly active or at least being seen to be doing so. New challenges such as decreasing product life-cycles, global competition, customers becoming more demanding and greater technological progress will present new issues for such managers and hence lead them to try to find innovative managerial approaches to address these issues. All areas of management practice have seen a wide range of these innovative managerial approaches, including management accounting. Dugdale et al. (2006, p. 3) describe how, following Johnson and Kaplan’s (1987) observation that management accounting had lost its relevance for management decision-making, ‘there was considerable interest in new management accounting theories and practices’. They cite as examples of these innovations, the rise in interest and adoption of activity-based costing and management, economic value added and throughput analysis. However, they also note that many of these so-called innovations do not seem to be new at all, but are re-work of earlier approaches. For example, activity-based costing (ABC) could be seen as a re-working and re-labelling of functional costing, which had appeared in the 1950s. Such experiences give the impression of volatile fluctuations in the popularity of managerial innovations. The results are often described collectively as management fashions or fads, invariably in a pejorative sense, and their individual importance being downplayed. The description usually implies that implementing, or trying to implement, these new management approaches absorbs much time of employees, diverting them from more important, well-established matters of importance, implicit in which is a recognition of the considerable investment costs that innovations entail. Much conventional extant research suggests that the rate of production and adoption of managerial innovations is increasing and the life-cycle of each innovation is decreasing (Carson et al., 2000). Moreover, recent innovations are tending to be broadly based, often organisation-wide, making them increasingly costly and
Don’t Blame the Tools
5
disruptive to implement (Rigby, 2001). As managers are faced with an increasing choice of innovative tools and techniques, it is important that they can select those that best suit their needs within the organisational context. Equally importantly, as managers gradually identify a new approach relevant to their needs, they need to implement it using ways that increase how effective the approach is and with as little disruption as possible to those operations of the organisation that are to be retained.
2.2.
Purpose of the Research
This report presents the findings of a study undertaken for the Chartered Institute of Management Accountants (CIMA) as part of their 2005 research initiative – ‘The Organisation and Adoption of Management Ideas, Tools and Practices’. The objective of this initiative was to ‘consider the origins of management ideas and practices, motives for their adoption, longevity, process of legitimacy, recognition and success’. Pursuant to this CIMA objective, the literature review undertaken for this study identified that, whilst earlier studies characterised a life-cycle for management innovations, no work seems to have been undertaken to explore how factors or influences at one stage of the life-cycle could affect later stages. Thus, this study addresses this gap in the literature, focusing on the adoption and implementation stages of the innovation life-cycle. Since there are no previous studies that have sought to explore this relationship between life-cycle stages, this study is exploratory in nature. The study, in particular, seeks to identify the crucial influences in the decision processes of organisations that lead to adopting a management innovation; and how these influences are related to the implementation of that innovation. It also seeks to identify how the adoption and assimilation of managerial innovations by organisations might be improved, that is, it seeks to identify good practice in this area. However, it is noted that what is considered good practice is likely to be contingent on many factors including: what the organisation is seeking to achieve from the innovation, what resources and skills it has at its disposal and what other innovations it has adopted previously.
2.3.
Report Structure
The following sections provide a brief review of existing literature relating to the lifecycle of managerial innovations and includes a consideration of the rationales for their adoption. The research aim and questions guiding the study are then presented in Section 4, and the method adopted in the study is described in Section 5. A summary of the cases is then provided in Section 6 and the findings of the study are presented in Section 7. The findings are summarised in Section 8 and conclusions are presented in Section 9. The implications of the research for practising managers and accountants and suggested future research directions are presented in the final Sections to the report.
6
The Adoption and Implementation of Managerial Innovations
2.4.
Intended Audience
There are three intended audiences for this work: ■ Managerial and professional staff who are involved in the adoption, use or ongoing evaluation of managerial innovations, including senior management that may be required to approve and sponsor such innovations. Some of these are likely to be accountants and some are likely to comprise the finance function of their organisation. ■ The academic community involved in researching the areas addressed by the report. ■ Staff of, and management consultants to, organisations involved in promoting and disseminating managerial innovations. This includes people associated with professional bodies and academic institutions; and those who provide other forms of learning, such as continuing professional development (CPD) and professional qualifications and training.
3.
Adoption and Implementation of Managerial Innovations
The following section provides a brief introduction to the existing literature that considers managerial innovations. It describes the idea of a life-cycle of innovations from the invention to their eventual decline. It also discusses the rationales leading to the adoption on such innovations which form a significant starting point for this study.
3.1.
The Life-Cycle of Managerial Innovations
Extant studies have suggested that managerial innovations, like other innovations, follow a life-cycle model. Carson et al. (1999) suggest the following stages in this life-cycle: ■ Invention – marginal awareness of innovation; emergence of literature on the innovation ■ Acceptance – implantation of innovation; case study support; increased number of publications ■ Disenchantment – evaluation and potential adaptation of innovation; literature highlighting limitations, disadvantages; number of publications plateaus
Don’t Blame the Tools
7
■ Decline – abandonment of innovation; literature disclaiming effectiveness; decline in number of publications on the innovation. They note that organisations are more likely to adopt an innovation that is in the ‘invention’ or ‘acceptance’ stages and abandon innovations in the latter two stages. However, individual organisations and their individual staff are likely to pass through this life-cycle at different times and different rates, resulting in ‘lagging’ organisations adopting an idea, whilst ‘leading’ organisations are already abandoning it. Etorre (1997) agrees with the life-cycle model proposed by Carson et al. (1999), but adds a fifth dimension ‘hard core’. This is where true believers remain loyal. Whilst a useful model for the stages that occur within an organisation, the Carson et al. model does not include the necessary precursor stage; the emergence or origination of the idea or innovation (Abrahamson, 1996).
3.2.
The Development of Managerial Innovations
Abrahamson (1996) proposes the model shown in Figure 1 to help understand the generation and adoption of management fashions or innovations. It is based on the notion that there are fashion setters who seek to produce and shape innovations and fashion followers who use those fashions. He identifies management gurus, business schools, mass media and management consultant firms as typical fashion setters. He also asserts that rather than ideas for improved practices arising spontaneously, they are ‘deliberately produced by fashion setters in order to be marketed to fashion followers’ (Abrahamson, 1996, p. 263). Williams (2004, p. 775) also identifies management consultants as being involved in ‘the production and engineering of management fashions, fads, buzzwords, buzz phrases and acronyms’. They are also involved in the selling of ideas and their application. However, he notes that they do not necessarily have a free rein; rather they are likely to operate within an institutional context that is subject to constraints. The rate of assimilation and retention of ideas initiated by outside management consultants inside an organisation can largely depend on the client managers (or similar) themselves who can act as a filter for ideas.
3.3.
The Adoption and Implementation of Managerial Innovations
As shown in Figure 2, Swanson (2001) suggests that the decision to adopt a particular innovation follows a stage of ‘comprehension’, when the organisation first learns about the innovation, and importantly, how it will both address the needs and fit their organisation. The decision to adopt the innovation is followed by two further
8
The Adoption and Implementation of Managerial Innovations
Figure 1 Management Fashion or Innovation Setting Process. (Source: Abrahamson, 1996).
Figure 2 Managerial Life-Cycle Stages. (Source: Swanson, 2001).
Don’t Blame the Tools
9
stages: implementation and assimilation. Implementation is regarded as undertaking the project, making it happen and bringing the innovation to life for its users, whereas assimilation is about making the innovation a part of routine, everyday practice.
3.4.
Rationales for the Adoption of Managerial Innovations
An important consideration in the adoption of innovations is the rationale underlying that adoption. Sturdy (2004) identifies six such rationales, as summarised below and described in detail in Table 1: ■ ■ ■ ■ ■
Rational – organisational effectiveness Psychodamic – relieving anxiety and securing identity Dramaturgical – successful rhetoric Political – furthering career, function, status or control
Cultural – cultural resonance or meaning ■ Institutional – securing organisational legitimacy Sturdy (2004, p. 168) notes that the rationales are not always mutually exclusive and the adoption of ideas is often multi-dimensional. In addition, he suggests a contingent view, whereby ‘different accounts of adoption are seen as appropriate depending upon the circumstances’. For example, prior research suggests that the rational view is more appropriate for situations of low uncertainty and that early adopters may be more rational than the ‘herds of followers’. Similarly, adoption rationale may vary with levels of management, with managers in a position where they will be held accountable more inclined to attempt rationale evaluation.
3.5.
Summary of Literature Review
The preceding review of literature shows that previous authors have considered various individual parts or stages of the innovation life-cycle. For example, Abrahamson (1996) considers the origination of such innovations and Sturdy (2004) considers the rationales affecting the adoption decision. Other authors have proposed theoretical models that describe the complete life-cycle or parts of it (Ehigie and McAndrew, 2005; Swanson, 2001). However, no studies have sought to explicitly link the factors at one stage of the innovation life-cycle, with factors at a subsequent stage. An understanding of how factors influencing early stages of the innovation lifecycle may affect later stages will be useful to organisations, since they can therefore address such factors as they make their adoption decisions and commence implementation, in ways that will improve the effectiveness of their chosen innovation.
Rationales for the Adoption of Managerial Ideas and Practices
10
Table 1
Limitations /Counter Views
1
Rational view – organisational effectiveness
Associated with individuals and based on a cognitive approach using models or proven techniques of (rational, economic) decision-making. Also associated with finding a solution to match a perceived problem or crisis, the search for ‘proven’ techniques and causal links between practice and performance and being able to ‘get beyond the hype’. The basis of most management accounting techniques (e.g. cost–volume–profit analysis, discounted cash flow methods)
Management rationality recognised as bounded – or even unobtainable. Difficulties in specifying problem and isolating impact of intervention on performance. Others view rationality as a way of justifying political and emotional decisions and hiding chaos, intuition etc. – and often applied as post-rationalisation of a decision taken for other reasons, a psychological ‘comfort blanket’. Widely criticised in management accounting research of the institutional and critical schools.
2
Psychodynamic view – relieving anxiety and securing identity
Associated with emotionally informed views and can be based on an impulsive decision to adopt ideas that may or may not be of benefit to the organisation. A competing psychosocial process that involves both autonomy and belonging so paradoxically, managers want to be seen to be using some new technique both before and at the same time as everyone else. Psychodynamics often used to explain the transience of management ideas.
Highly influential because of its counter view to the rational perspective. However, criticised because it views managers as so under pressure and anxious that they will adopt anything, particularly if others are doing so. This misses out that many managers are active in the production and transformation of ideas.
The Adoption and Implementation of Managerial Innovations
Description
Dramaturgical view (rhetoric) – successful rhetoric
This view focuses more on the supply side of the relationship, the persuasive influence of management gurus, management consultants, academics, etc. and their presentation techniques. Overall, impression management is vital, not necessarily the content. Three rhetoric elements are identified: source credibility, emotional appeal and logical proof.
Like the limitations to the psychodynamic view, the rhetorical view tends to give the view of managers as passive ‘victims of the clever tricks of organisational witchdoctors’ (p. 161). It also presents a top–down view of the knowledge diffusion process rather than one based on interaction and transformation.
4
Political view – furthering career, function, status or control
Broadly concerned with ‘the instrumental use of ideas to secure power and/or with their content in terms of their material and/or discursive power effects’ (p. 161) – basically ‘which ideas and practices are diffused depends in part on who has control of the means of dissemination’ and ‘ideas flow from the powerful’ (p. 162) – leading to the current domination of US and neo-liberal ideas. This also includes a distancing perspective – high status organisations or individuals will seek to move on when fashions become too popular.
Political perspectives tend to generalise the process of adoption – focussing on interests and outcomes rather than context or relevance. It therefore fails to explain adequately why some ideas spread whilst others do not or do so at a different time or in a different sequence.
Don’t Blame the Tools
3
11
Table 1
(Continued) Limitations /Counter Views
Cultural view – cultural resonance or meaning
Highlights that the diffusion of ideas can be spread across cultures, for example, through globalisation, mergers/joint ventures. Local knowledge can act as a ‘bridge or barrier’ to transfer. However, it also stresses that culture can affect national, sectoral and organisational diffusion of ideas.
In certain contexts, ideas and practices are attractive precisely because they are alien or a challenge to existing arrangements (e.g. Japanese or North American) – however, this ‘dominance’ effect is often overlooked.
6
Institutional view – securing organisational legitimacy
This view has become dominant in organisational studies, including much management accounting research. It considers various societal or social influences on organisational practices – their ‘institutional embeddedness’. Its premise is that organisations operating in a similar environment will employ similar practices (isomorphism). Practices tend to seek peer and shareholder legitimacy at the possible expense of efficiency or control outcomes – and hence this view presents a challenge to the rational view. Variability in institutional conditions is seen to account for different levels and patterns of adoption – and can result in ‘ceremonial’ adoption.
Despite its dominance, institutional theory has been criticised, including by critical accounting researchers. Power dynamics are underplayed or taken for granted and it tends to ignore nations or organisations that do not fit the pattern of the region or sector.
Source: After Sturdy, 2004.
The Adoption and Implementation of Managerial Innovations
5
12
Description
Don’t Blame the Tools
4.
13
Research Aim and Questions
This study seeks to address the identified gap in linking the stages of the innovation life-cycle evident in extant literature. It would appear most useful to focus the study on the early stage of the life-cycle, as these may be found to affect later stages. The study therefore seeks to study the decision to adopt an innovation and seeks to determine the impact that this may have on subsequent implementation of the innovation. In particular, Sturdy’s (2004) identified rationales for adoption will be used to identify different adoption rationales and to determine the effect these rationales appear to have on later stages of the life-cycle, particularly implementation. The aim of the research can therefore be summarised as: To identify the significant influences on the decisions made by organisations to adopt a management innovation and how these are related to the subsequent implementation of that innovation. This aim is pursued by exploring the following questions: 1. How is the adoption of a management innovation by an organisation triggered? 2. How do the adoption decisions studied relate to the adoption rationales identified by Sturdy (2004)? 3. How do these influences on the adoption decision, including the rationales identified by Sturdy, relate to the subsequent implementation of the innovation? 4. How can the assimilation of organisation-wide innovations be improved, that is, what appears to be good practice given the organisational context?
5.
Research Method
The previous section established the aim of the research to identify the significant influences on the decisions made by organisations to adopt a management innovation, and how these are related to the subsequent implementation of that innovation. Qualitative case study approaches are well-accepted methods when addressing ‘how’ questions in the management and accounting research domain (Chua, 1988; Eisen\hardt, 1989; Mintzberg, 1979; Yin, 1989), and such a method was adopted for this study, involving multiple cases to allow comparisons between organisations.
5.1.
Identifying Cases and Interviewees
Extant literature suggests that more recent managerial innovations tend to be organisation-wide and that it is these types of innovations that managers are finding most difficult to implement. The study therefore focuses on such organisation-wide
14
The Adoption and Implementation of Managerial Innovations
innovations. Two distinct managerial innovations were chosen to meet this requirement, the BSC and the programme management office (PMO). Four organisations were included in the study. Three organisations had adopted and implemented the BSC and one organisation had adopted and implemented a programme management office. One of the organisations that had implemented the BSC had, after using the innovation for a few years, discontinued its use, allowing us to study the decline and discontinuance of an innovation. Undertaking the cases in different organisations that are, or had used the same innovation, allowed issues that were common to the innovation, and also allowed issues that appeared to vary across organisations, to be identified. Studying two distinct innovations also allowed comparison of issues that appeared common to both as well as those that were specific to one. Having identified the type of cases to be studied, a convenience approach to sampling was adopted (Yin, 1989). Two of the researchers involved in the project approached organisations that they knew fitted the study requirements. These were located in Scotland and England. Although this proved to be a lengthy process, which is often the case in case study research, agreement was achieved from four organisations. For the purposes of this report, the organisations have been anonymised and will be referred to as indicated as follows: ■ Hospital Trust (HT) implemented and used a BSC ■ County Council (CC) implementing a BSC ■ Retail Bank (RB) has implemented and used a BSC and now discontinued its use ■ Police Force (PF) implemented and used a PMO Within each case study organisation, interviews were carried out as shown in Table 2. Interviews were undertaken with a range of relevant staff including; senior managers responsible for the decision to adopt and managers responsible for leading the implementation, some of whom were either from finance functions or from a financial background. It is recognised that our interviewees did not include middle mangers, other staff in the organisation and other stakeholders. The findings of this study therefore reflect the views of senior managers. As discussed in the suggestions for further research, the inclusion of a wider range in interviewees may provide a wider range of perspectives on the perceived rationales for adoption and the challenges of implementation and use. Consistent with other case study based research, interviews were guided by an interview schedule derived from the research questions (Yin, 1989). The questions were semi-structured in nature. All interviews were recorded and transcribed. Summaries of the case studies were passed back to interviewees to ensure accuracy and to ensure their agreement for the data collected to be included in the study
Don’t Blame the Tools
Table 2
15
Case Study Interviewees
Organisation
Number Interviewed
Roles
Hospital trust (HT)
5
1. Chairperson 2. Director of Finance and Performance 3. Chief Operating Officer 4. Director of Human Resources 5. Deputy Director of Finance
County council (CC)
2
1. Head of Performance Improvement 2. Performance Improvement Manager
Retail bank (RB)
4
1. Group HR Director 2. Head of Mortgages 3. Head of Retail Employee Relations 4. Head of Distribution Risk and Sales Quality
Police Force (PF)
4
1. Head of PMO (Police Inspector) 2. PMO team member (Police Sergeant) 3. Manager, Performance Improvement Unit 4. Project Manager (Police Inspector)
findings. Cross-case analysis was undertaken by identifying themes within the cases and then grouping or contrasting the cases according to those themes. Tabular layouts of the case data according to the identified themes were used to aid the cross-case analysis.
6.
Summary of Case Studies
This section presents a brief overview of each of the four case studies. Further information from the case studies is presented in Section 7, where the detailed findings of the case studies are discussed.
16
6.1.
The Adoption and Implementation of Managerial Innovations
Case 1: Hospital Trust (HT)
HT is a hospital trust in the English National Health Service (NHS). Around 2002, HT attracted significant, high profile negative publicity relating to poor clinical practice. HT was subsequently identified as being one of the worst performing trusts in the country by the Audit Commission. Following this, an independent assessment by the regional health authority identified a number of issues at HT, including poor governance and senior management. Coinciding with this was the arrival of a new chairperson, whose primary objective was that HT would improve its performance rating (from zero) by changing various structures, processes and staff. However, in seeking to improve HT’s performance, its expenditure increased, triggering a financial crisis. In 2005, HT brought in external management consultants in an advisory capacity. As part of addressing the situation, they recommended that HT should use a BSC in order to improve governance and performance management. Development and implementation of the BSC was accomplished within six months. HT is now one of the top 40 hospitals in the NHS and is currently applying for foundation status.
6.2.
Case 2: County Council (CC)
CC is a local authority serving a county of England. Like other organisations in the public sector (including HT and PF), CC is under constant performance review by central government. CC has been assessed as performing consistently well over the years. Despite this good level of performance, in 2001, senior managers had a strong desire to ensure continuous improvement. Total quality management (TQM) was considered as one approach that CC could adopt, but officials of the strategic management board (SMB) identified the BSC as more appropriate and useful. Adoption commenced in 2002, and someone with experience of the BSC elsewhere was recruited from outside the organisation as head of performance improvement to lead implementation. Management consultants were brought in to assist initially. The process of implementing the BSC organisation-wide has been much more protracted than in RB or HT. A phased approach to implementation commenced with use by the SMB, followed by plans to roll out adoption in each of the several service divisions of CC. It was expected by those interviewed in 2006 that the BSC implementation across CC would be complete in 2008.
6.3.
Case 3: Retail Bank
RB is the retail arm of a large bank based in the UK. In 1993, a new chief executive was appointed. He thought that RB was too focused on short-term financial performance,
Don’t Blame the Tools
17
and so was excluding other issues, such as innovation, customer service and employee development. He was keen to develop a new performance management system. RB identified a BSC as fitting its needs, and after testing extensively, the BSC went live across the organisation on 1 January 1996. Rather than the more usual strategic BSC (see Kaplan and Norton, 1996a, 1996b), it was decided to develop an operational scorecard. The measures from this could be ‘cascaded up’ the organisation. This BSC was included as part of a wider culture change programme called the Customer Focus Programme. RB used its BSC for about three years. Then, there was a change of chief executive and of focus. The incoming chief executive had a successful background in retailing and believed the crucial factor for the success of RB was the sale of products. In 1999, RB replaced the BSC with a sales management system.
6.4.
Case 4: Police Force
PF is a police force based in Scotland. In 2003, a corporate risk register drawn up by PF identified ’an inability to deliver major projects to time, cost and quality’. At about the same time, Her Majesty’s Inspectorate of Constabulary (HMIC) also identified a lack of coherent business processes within PF. In 2005, PF established new governance mechanisms to cover all projects in order to address these issues. A PMO was established as part of these mechanisms. This occurred rapidly, in just ten months. The role of the PMO was to promote the use of standard project management methods and tools across all projects, monitor the progress of ongoing projects and support the new governance structure. In 2007, an evaluation of project management within PF was conducted as part of an audit by the City Council that oversees PF, resulting in a positive audit report. However it is recognised that within the force not all projects used the suggested approaches or used them consistently.
7.
Discussions of the Findings
The previous section summarised the case studies. The following section presents an interpretation of the case studies and the cross-case analysis that was undertaken. Following Patton (1990), in order to focus this interpretation, the following six themes or topics relating to the adoption decision phase and the implementation phase of the innovation life-cycle were considered across the four case studies: ■ The problem or event that triggered the need for an innovative way of working ■ Influences on the decision to adopt the particular innovation chosen and how these relate to decision-making approaches identified by Sturdy (2004)
18
The Adoption and Implementation of Managerial Innovations
■ When in the life-cycle of the innovation adoption commenced ■ How the innovation was embedded in the organisation ■ The length of time it took to implement the innovation after making the decision to adopt ■ How long the use of the innovation lasted.
7.1.
Triggers of Innovation Adoption
All four cases exhibited a clear trigger, such as an event or an ongoing problem that caused each organisation to consider the adoption of the managerial innovation studied. In the case of HT, the public criticism of certain activities in the hospital led to questions about management and control at HT, in particular the level of governance being exercised by the board. In trying to improve the areas of poor performance identified in the criticisms, HT generated a significant financial deficit, culminating in a financial crisis. The board needed to address this deficit, but did not want service provision to suffer, since they were planning to apply for foundation status. As part of an overall review, a management consulting firm recommended the BSC as an approach that would allow the financial challenges to be addressed, whilst also allowing the board to gain, and also maintain, control over all other aspects of the HT’s operations. A mixture of both external and internal triggers also occurred in the PF case and led to the creation of the PMO and the adoption of standard project management methods. A report by HMIC identified a lack of consistent processes within PF, whilst their own corporate risk register identified a lack of ability to deliver major projects to time, cost and quality. Both RB and CC appear to have had largely internal based triggers for the adoption of scorecards. The appointment of a new chief executive at RB – and the challenging task of integrating and developing consistent operations for two distinct organisations following a merger – led to the adoption of the BSC. The adoption of the scorecard at CC, whilst influenced by a comprehensive performance assessment undertaken by the Audit Commission, was driven by the CC’s recognition of its limited use of management information and performance measures to manage the diverse operations of the organisation and its desire to achieve continuous improvement. A striking feature is that all triggers derived from powerful institutional stakeholders and at senior levels in the organisations. Grassroots clamour for these innovations was not evident in the cases, although that may be because informants were content to play this down.
Don’t Blame the Tools
7.2.
19
Influences on the Adoption Decision
In considering the descriptions given by interviewees relating to the rationale and influences on the adoption of the innovations considered, a number of differences and similarities emerged. A striking similarity across the four cases was the lack of formal evaluation of alternative ways of addressing the events or problems that had triggered the requirement for a managerial innovation. CC described considering the use of TQM rather than the BSC, but an extensive rational comparison does not appear to have been carried out. However, even this limited exploration of alternatives is more than what the other organisations appear to have undertaken. Two aspects seem to emerge from the descriptions of the decision to adopt and vary across the cases; the degree to which the decision to adopt was influenced by a significant individual or group, and the degree to which the organisation either looked externally to other organisations to ratify their adoption, or relied on their own internal sense making and known capabilities for justification. Adoption of the BSC at RB was clearly highly influenced by the then relatively new chief executive who was determined to improve performance management in the organisation and who was also very keen to ‘reduce the amount of paper being used to measure performance’. Whilst the original idea of the BSC came from outside of the organisation, descriptions given by interviewees of the rationale to adopt, did not refer to wishing to emulate other organisations. Given RB were a relatively early adopter of the BSC, as discussed further below, there were few other organisations to either learn from or emulate. Rather, a few staff had learnt about the scorecard from their studies at leading business schools and were confident in their own judgement that this would address the wish to improve performance management and that they could make it work in their organisation. It would appear that the new chief executive wanted to make a public mark of his arrival and intent to change, and the BSC rhetoric was a way of doing this. It also represented the enthusiasm and new learning brought back by those who had been away on courses. In contrast, there was no individual or group that clearly led the decision to adopt the BSC at CC. As may be expected in a public sector body, with a high degree of influence by elected members, the decision to adopt the scorecard appeared to have been a decision by a group or inner circle of a formal committee. Indeed, in complete antithesis to what was observed at RB, the group decision at CC left the scorecard as a solution looking for a champion: initially there was no individual strongly driving the use of the balanced scorecard [CC] The decision to adopt the BSC was also strongly influenced by external parties, in particular, the use of the scorecard by other organisations. Whilst there were likely
20
The Adoption and Implementation of Managerial Innovations
to be more widespread use of the scorecard when it was adopted by CC, since it was a much later adopter than RB, it was noted that this use and apparent popularity influenced the decision to adopt: at the time, the balanced scorecard was seen as a vogue new thing and a different thing to try, that could be of benefit to the organisation [CC] The adoption of the scorecard by HT was also influenced by the legitimacy gained from its widespread use, particularly amongst similar healthcare organisations and other parts of the public sector. However, in contrast to a decision to adopt that arose from group interaction, HT’s adoption was directly influenced by the recommendations of the external management consultants that the board had commissioned and was strongly championed by the new chairperson. The decision to form a PMO within PF appears similar to the group or ‘corporate’ decision-making witnessed in CC. Again, this shared decision-making, as opposed to a more autocratic approach, is consistent with PF being part of the public sector. However, as witnessed by the case of HT, public sector organisations can be influenced by groups or even individuals. Once again, the group decision to form the PMO resulted in the search for a sponsor or champion for this initiative. However, in this case a champion was found relatively easily from within the organisation in the form of a senior officer. Like RB, PF did not appear to be highly influenced by other organisations adopting this way of working. Again, this may be due to them being early adopters of a PMO, particularly across police forces, and hence had few organisations to emulate. However, the ideas were known in other organisations, yet PF did not cite their influence when discussing adoption. Instead, like RB, they appeared confident that their choice of innovation would address the issues identified in the corporate risk register and address the concerns raised by HMIC. Combing the two criteria: the degree to which the decision to adopt was influenced by a significant individual or group; and the degree to which the organisation either looked externally to other organisations to ratify their adoption, or relied on their own internal sense making and known capabilities for justification, form the basis of the 2×2 matrix shown in Figure 3. The four case studies are shown in this matrix according to the preceding discussion. It can be seen that each of the cases occupies a distinct quadrant of the matrix. This model assumes a certain amount of reification of formal organisations, giving rise to a distinction between outside and inside said organisations. Some might argue that some of the externals are only so, provided one distinguishes between organisational arms of the same government (e.g. HMIC, the Audit Commission, the NHS, including HT, and police forces like PF could all be considered as branches of one organisation). Conversely, within the division or units of the case study organisations may see themselves as distinct units to a greater extent than part of CC, HT or PF).
Don’t Blame the Tools
Figure 3
7.3.
21
Influences on Adoption Decision Matrix.
Link to Sturdy’s Adoption Rationales
The adoption decision in each of the four cases was also considered according to Sturdy’s (2004) six rationales for adoption. Sturdy (2004) indicates that the six rationales he describes are not exclusive, and the adoption of any particular innovation may exhibit more than one type of rationale. In particular, his observation of the dominance of the rational view of adoption and its use to address the scepticism that often meets the introduction of new ideas means that a rational reason for the adoption of most innovations will be given when discussing the adoption, even if this was not the sole or the primary rationale for adoption. In all cases studied, a description of the event or problem that triggered the adoption of a managerial innovation was given as a rational reason for the adoption. That is, those interviewed were keen to justify the adoption by stating problems they were facing at the time, rather than giving reasons for adopting the particular solution or innovation in preference over other courses of action that could have been available. However, in analysing their responses in more depth, rationales for the particular approaches adopted could be identified. In each case, one rationale tended to be more significant than others. At RB, the pivotal role of the new chief executive in the adoption of the scorecard suggests what Sturdy (2004) refers to as a political rationale for adoption. In such cases, ‘which ideas and practices are diffused depends in part on who has control of the means of dissemination’ and ‘ideas flow from the powerful’ (Sturdy, 2004, p. 161 and p. 162). The chief executive was keen on a new performance management system to change the focus of the organisation from financial process and paper procedures
22
The Adoption and Implementation of Managerial Innovations
to measure performance and used the rhetoric of innovation, customer service and employee development and leadership to justify a new system, a bill that the BSC fitted. A coincidental factor was that some RB staff had recently learnt about the scorecard and their knowledge came together well with the chief executive’s need for an innovative solution. The institutional rationale for the adoption of innovations appears most closely to fit observations at CC. Derived from organisational theory, one premise of this view is that organisations will seek to emulate similar organisations by adopting similar practices (isomorphism). Indeed, one interviewee at CC described the choice of using the BSC, in part, being influenced by the fact that the scorecard seemed to be ‘a vogue’ amongst other organisations at the time. Public sector organisations like CC (including HT and PF) have been encouraged to pursue performance measurement and performance management rhetorics as part of new public management (NPM). They have been under pressure from protagonists of NPM, often within central government departments, to manage (or appear to manage) like the allegedly more efficient and superior private sector, and so the BSC fitted this need to be seen in a particular way. Ideas of multi-dimensional, quantitative and qualitative performance measurement have a long history in local authorities due to the need to provide an integrated approach to a disparate collection of services within one democratic-bureaucratic organisation and have included other managerial approaches, which were innovations in their time, for example, management by objectives and corporate planning. What Sturdy (2004) terms the psychodynamic rationale appears to most closely describe the formation of the PMO by PF. Such a rationale is based on the need to relieve the anxiety about an issue or problem, most often doing this in a way that is well accepted by others. PF were keen to address and be seen to address the issues about large projects raised in their corporate risk register and the lack of standard processes identified by the HMIC report. Introduction of formal project management methods was likely to take some time and be difficult to demonstrate easily. In contrast, the formation of a physical PMO could be relatively swift and the co-location of a number of dedicated officers and support staff in a single office would provide a clear symbol that the issues identified were being addressed. Adoption of the BSC by HT was clearly ‘influenced’ by the firm of management consultants. The influence or persuasiveness of the promoters or suppliers of innovations, suggests Sturdy’s (2004) dramaturgical or rhetoric view played a major role in the adoption in this case. The new chairperson was a champion of the scorecard, suggesting that like RB, the adoption by HT also had a ‘political’ element to it, hence confirming Sturdy’s (2004) observation that adoption decisions can have multiple rationales. In Figure 4, the four distinct adoption rationales identified in the cases studied are overlaid on the matrix developed in the previous section (see Figure 3). The different rationales identified by Sturdy (2004) are clearly relevant to our four cases, with each
Don’t Blame the Tools
Figure 4
23
Sturdy’s (2004) Adoption Rationales.
falling into a separate quadrant. This suggests that the characteristics used as axes in the matrix can usefully differentiate between cases and appear to be consistent with Sturdy’s (2004) typologies. As described above, the decision to adopt by both RB and HT was strongly influenced by either an influential individual or a group, hence their appearance in the top row of the matrix. As noted by Sturdy (2004), in the case of the dramaturgical rationale, this is fuelled by rhetoric and represents the supply or external provision of advice or innovations, as in the case of HT. Whilst it is conceivable that the influence of a powerful individual inherent in the political rationale for adoption could be from an external individual, it is most likely that anyone who could wield the necessary operational knowledge and formal authority would be internal to the organisation, as in the case of RB. Both PF and CC demonstrated a group or consensual approach to decision-making, hence they appear in the bottom row of the matrix. However, they differed markedly in the degree to which they were influenced by the use of the innovation by other organisations in making their adoption decision. Although the problem to which PF sought a solution was externally signalled, the solution was its own choice, whereas CC followed other organisations in the innovation it chose.
7.4.
Time of Adoption in the Innovation Life-Cycle
Graphical plots of the number of bibliographic citations of a given innovation are often used to illustrate its life-cycle. A plot of the number of citations in the literature
24
The Adoption and Implementation of Managerial Innovations
of the term ‘balanced scorecard’, suggests that, if citations are closely related to adoption, then growth in adoption first began around 1997. Maturity, as signalled by a plateauing in the frequency of citations, occurred around 2002. Of the three organisations studied that adopted a BSC, RB was the first to do so. It began developing and testing its scorecard in 1995 and the scorecard was implemented and went live across the business on 1 January 1996. Comparison with the citations count suggests RB could be considered an innovator or early adopter according to Rogers’ (1995) innovator types. CC began its development of its scorecard in 2002, with the scorecard going live for the SMB in 2003. Roll out to other parts of the organisation is still ongoing. Although having an earlier ‘traffic light’ reporting system for key targets, HT did not develop and implement its BSC until 2005. Comparison with the plot of citations, suggests that these organisations could be described as majority adopters (Rogers, 1995). Whilst the use of project management tools and methodologies has shown growth over the last decade, the establishment of PMOs and similar groups is not yet widespread amongst organisations. PF established its PMO in early 2004 and can therefore, like RB, be considered as an innovator or early adopter, although it remains to be seen when maturity of this type of innovation will occur. Both organisations identified as early adopters described undertaking activities to share their knowledge and experience of use of the relevant innovation with peers and other organisations. Staff at RB who were responsible for the development and implementation of the scorecard: participated in various conferences promoting their use of the scorecard and also participated in developing a teaching case study based upon their use for MBA and management diploma students [RB] Similarly, staff from PF have: been active in introducing the benefits management approach promoted by staff in the PMO to other police forces, and to that end have organised workshops for staff from other forces [PF] These staff recognised that such activity helps their organisation to be perceived as forward looking and innovative, not just to those who receive the promotional activities, but also to staff within their own organisation. However, these activities are also undertaken to help and support other organisations considering adoption and, hence, they actively contribute to the institutional rationale of adoption identified in the CC case.
Don’t Blame the Tools
7.5.
25
Relative Speed of Implementation
In considering the speed of implementation of the innovations studied, the elapse of time from an initial decision to adopt to the point at which an innovation is as widely deployed in the organisation as it is intended to be, is identified as a variable of interest. Hence, in a similar manner to the consideration of the time of adoption, consideration of the speed of implementation suggests the use of a diffusion curve. However, rather than a single curve derived from an external measure that can be applied to each of the organisations adopting the same innovation, the speed of implementation indicates the need for a diffusion curve for each organisation. Maturity of diffusion can be taken when the innovation was available to as many individuals or users as the organisation has determined as necessary or useful. The most rapid implementation amongst the organisations studied is that of HT’s adoption of the BSC. HT decided to adopt a scorecard following the recommendations of a management-consulting firm in May 2005, and the scorecard was available to the board of HT later that year, approximately six months later. A number of factors contributed to this rapid implementation: HT was already using traffic light reporting of its vital targets, which could be used as a basis for the scorecard. Due to being a relatively late adopter, HT and their advisors, were able to learn from similar hospitals that had developed their own scorecards. HT also adopted a ‘technologylight’ approach. Rather than relying on a large data warehouse to collect data for the scorecard or commercial software to produce it, a member of staff pastes data into a tailored spreadsheet. Finally, the intention was limited to developing a scorecard to report information to the board of HT, rather than to develop scorecards for the many areas that constitute the hospital. Whilst many HT staff were aware of the use of the scorecard by the board (e.g. they were required to provide information that is included in the scorecard), this has removed the need to determine significant measures and goals for each area of the organisation and communicate and train the many hundreds of staff within HT. RB was not only an early adopter of the BSC, it also carried out implementation relatively rapidly, undertaking all development, testing, communication and roll out in less than one year. Rather than develop the traditional strategic scorecard, RB used their scorecard to improve operational performance management. Scorecards were therefore developed for the retail network, which at the time consisted of over a hundred branches and thousands of staff. All staff were trained in the interpretation of the scorecard and had a scorecard relevant to their role available on a PC within their branch or regional office. Interestingly, RB’s implementation was not only relatively fast, the lifetime of the scorecard in the organisation was only three years, as the scorecard was discontinued when a new chief executive joined the organisation and replaced it with a sales management system. This three-year lifetime was less than the time some of the other organisations studied have taken to implement or roll out their scorecards.
26
The Adoption and Implementation of Managerial Innovations
PF was relatively swift in establishing its PMO, launching this just ten months after the idea was first identified following the risks associated with large projects. However, the diffusion of the use of the project management tools and frameworks that the PMO was seeking to encourage has been proving to be somewhat slower. There are some examples of the use of the tools, and their use in planning the policing of the high-profile conference of world leaders that helped spread awareness. However, use of the tools is still patchy in most projects. One group seen to be significant to encourage adoption is project managers’ superiors who have ultimate responsibility for the projects in question (termed senior responsible owners, SROs). Buy-in from them has been highly variable. It is expected that it will take a number of years before the use of the project management tools is widespread within the organisation. Finally, CC represents the slowest rate of implementation of the cases studied. A decision to adopt the scorecard was made in 2002, and whilst the scorecard went live for the SMB in 2003, the organisation has always intended and is still intending to develop scorecards for each of the divisions or services within CC, which would link, via ‘golden threads’ to the SMB card. In December 2006, the Environment Division had made the most progress, having had a working scorecard for about a year. A scorecard for the Corporate Services Division was in the process of being built during the middle of 2007 and was due to be populated soon after. Scorecards for Adult Care Services and Children, Schools and Families were due to be in use by 2008. Progress at CC has been slower than they expected, as described by the Head of Performance Management: if you had said to me in 2002, you’ll still be battling away with this at the end of 2006, I would have thought that sounds like a very long way away [CC] This has been caused by a number of factors. Each division undertakes very different activities and therefore has different major performance measures. Hence, an identification of these measures and how they relate to those in the SMB scorecard must be undertaken. The existing levels of performance management varied significantly across divisions. So, in some divisions it was necessary to educate staff about performance management ab initio, whilst in others it was necessary to encourage them to change from their existing approaches to the BSC. A significant issue is that many existing measures are determined by external reporting and monitoring requirements, and these need to continue being generated and used. Finally, CC was keen to use technology to support its BSC and has used its SAP ERP system as a platform. This has caused some delays as CC was required first to ensure that its SAP system was robust and used appropriately, and that staff responsible for the scorecards was able to build and maintain them using this system.
Don’t Blame the Tools
7.6.
27
Good Practice in Embedding Innovations
What we have called good practice (see section 2.2 for a discussion of our use of this term) was identified by looking across all cases and finding those practices that seemed to be linked to more rapid and effective implementation. These included: ■ Adapting the innovation to fit the organisation. All four organisations studied adapted the innovation in question to meet the contingencies of their organisation. For example, all three organisations that adopted the BSC sought to identify four quadrants that reflected their organisational needs and sought to populate these with appropriate measures. In the case of RB, rather than use the scorecard as a strategic tool, as suggested by Kaplan and Norton (1996a, 1996b), they adapted it to become an operational tool. In the case of PF, an example of adaptation to best suit its local needs was the development of a light touch project management approach. Rather than adopting the standard PRINCE2 methodology, which is renowned as being highly detailed, a lighter version was developed from this by the PMO staff and was shared with other staff within the organisation. ■ Training to use the innovation including tailored training. Again all four organisations studied sought to ensure that the relevant staff, through training, understood how to use the respective innovation and their role in its use or dissemination. They also recognised that different users or user groups would require different approaches for this training, either with respect to the content of the training, or how it was delivered. For example, in the case of PF, instructing project managers about all of the stages involved in project management required them to provide a lot of information, and when the staff ‘were given all of this information in one go, they didn’t come back!’ The staff in the PMO, therefore, set about delivering training on a ‘need to know basis’, informing staff about just enough of the process so that they could work effectively on the next stage of their project without getting overwhelmed by what might come later. ■ Powerful champion within the organisation. Consistent with the rationale for adoption being driven by a significant individual within the organisation, the adoption of the BSC at RB, and the sales management system that followed this, were driven by a high profile and powerful champion in each case. Although in both cases the individual was new to the organisation, they had already joined the organisation and then had championed the innovation in question. In contrast, CC made, what would appear to be a committee decision to adopt the BSC, influenced strongly by the adoption by other public sector organisations and councils. It then sought to recruit an expert who could champion its development. Previous authors (e.g. Open University, 2004) have compared the role and success of internal
28
The Adoption and Implementation of Managerial Innovations
champions to external champions. They find that whilst internal champions have certain strengths that can aid change, such as a knowledge of internal systems and processes, external champions also have strengths that may be important in certain cases, such as, the ability to use their expert and independent status to solicit and influence views in the organisation. ■ Pilot, stress testing and a business case. Staff interviewed at RB were keen to stress that their investment in new ways of working entailed in the BSC were subject to the same testing and scrutiny as other organisational investments. The scorecard, and innovations they have developed since, such as their current performance management system, were piloted with a limited group of staff and then stress-tested with a larger group. Only after this testing had been carried out, was a business case submitted to win approval for wider implementation of the innovation. Lessons learnt from the testing, such as how to implement the innovation and the degree or size of the benefits that might be expected, could then be included in the business case. Thus, although testing was part of legitimising the innovation, and so had socio-political significance, it also had an economic, rational meaning. In contrast, such rigorous testing and quantification of the expected benefits were not apparent in the HT, PF and CC cases, nor did this absence seem unusual. Consistent with the institutional adoption rationale in the case of CC, this organisation seemed to believe that adoption by other organisations demonstrated that effective implementation could be achieved and that benefits would be realised. For HT, it seemed good enough that outside professionals vouched for the innovation. Moreover, both perceived that it would provide legitimacy in terms of NPM, performance management and emulating the private sector in how to manage. Part of the reason for an absence of formal business cases may lie in the fact the adoptions were not seen as generating costs, or at least not ones requiring distinct expenditures. However, these case studies have shown that innovations are associated with the engagement of consultants, the acquisition of systems or the establishment of new posts and organisational units. Approval for such inputs or resources may often be approached separately and piecemeal, masking the true cost. ■ Link to wider initiatives that are easily understood. The adoption and use of the BSC at RB was linked to the wider ‘customer focus programme’. Inclusion within this programme was seen as a major contributor to the success of the BSC:
staff understood the ‘customer focus programme’ and having a new measurement system was seen as part of the programme – so we got it in under the radar a bit. [RB]
Don’t Blame the Tools
29
Similarly, introduction of the BSC helped the success of the ‘customer focus programme’ as it emphasised the importance of the customer perspective in addition to the sale of products. PF also used other changes and events to increase awareness and acceptance of the PMO. New governance arrangements for projects and programmes were introduced at the same time as the PMO was formed. The PMO also supported planning for the very high profile conference of world leaders held in their region, exposing many more staff in the organisation to the working of the PMO and the project management tools it was trying to introduce into the organisation. ■ Sustainability is aided by an established performance management environment that has been cascaded down the organisation. Here we use the term sustainability to refer to an innovation that continues in use, with appropriate levels of support expected for ongoing operation, rather than the significant level of support and resources necessary for the implementation of that innovation. We do not consider that to be sustainable an innovation must demonstrate indefinite use. This would be in conflict with our finding that the adoption of innovations is a natural consequence of a dynamic business environment, and hence any innovation cannot be expected to be in use indefinitely. The case studies, in particular RB and HT, suggest sustainability. That is, the innovations were used routinely and effectively in the organisation with a moderate level of support. In contrast, the drawn out implementations demonstrated by both CC and PF have meant that these have not yet reached a level of sustainability and still both require considerable ongoing efforts and resources to make sure they are adopted more widely and used routinely. The cases HT and particularly RB, both established a strong performance management environment in addition to the scorecard itself, which was communicated and cascaded down the organisation. In contrast, the following appeared to slow implementation, and so might be described as contrary to good practice: ■ Allow a phased approach to become slowed by divisions with diverse operational areas and belief systems. A phased approach is often a sensible way of managing workloads within a project and a means of managing disruption and risk. However, it may also signal that there are different timescales for different parts of the organisation. It would appear that the CC introduction of the BSC has some elements of this, with different service divisions within CC moving at very different speeds towards development and the use of a scorecard for their particular areas. Some of this may also be due to the various parts of an organisation not only working in quite different operational environments but also having different
30
The Adoption and Implementation of Managerial Innovations
divisional ‘cultures’, or systems of beliefs and values, making them more sceptical of alien ideas and so unwilling and unable to change. ■ Introduce an innovation where existing ways of working are adequate. In addition to what has just been said, part of the challenge in the case of CC’s introduction of the scorecard was that a number of the service divisions already had performance measurement and management approaches that were working adequately, met their needs and satisfied the needs of knowledgeable and authoritative external agencies along the policy system chain. PF faced a similar challenge, in that many of their SROs, who were seen to be pivotal in encouraging the use of the project management tools, were comfortable with the range of current approaches in use and thought that these should be acceptable. In both cases, whilst the local approaches were adequate within divisions or individual projects, both organisations were seeking to change local behaviour in at least the name of making organisation-wide improvements. In the case of CC, this was to develop a consistent performance management approach across the organisation based on a scorecard supported by a number of linked divisional scorecards. In the case of PF, the intention was to develop explicit and consistent processes across the organisation to address the risks relating to large-scale projects. It seems likely that the divisions affected saw these as attempts at control of their divisions by persons in the case of CC who, although formally were higher up in authority terms, probably lacked professional legitimacy to be entitled to control. The same would apply at HT. At PF, one would have thought such professional legitimacy existed, with the chief constable seemingly prominent among those attempting to control lesser ranks of the service, but it may be that is not how it was seen and perhaps the unwillingness to comply with central wishes were part of the various officers who were SROs signalled a desire to maintain some autonomy. ■ Ensure there are benefits for both the organisation and for individuals. One interpretation of the challenge faced by both CC and PF is that whilst both innovations sought to deliver benefits at the organisational level, there appeared to be few benefits for the divisional units and individuals involved. Without the driving force of a powerful champion and the organisational momentum that such an individual can generate, it is necessary to identify the benefits for both individuals and the organisation, if the divisional units and individuals in question are to engage actively in the adoption, implementation and usage processes. Assurance that the innovation will not be discontinued before it can deliver improvement is also likely to help. The public sector, in particular, is seen to have suffered from whims, political expediency and impatience at the top manifesting themselves at the middle as inconsistent changes whose advantages for service recipients and professional providers are difficult to observe and among whose disadvantages are uncertainty, costs and diversions from core activities.
Don’t Blame the Tools
31
Figure 5 Matrix Combining Influences on Adoption, Time of Adoption and Speed of Implementation.
7.7.
Combining the Influences on Adoption Decision, Time of Adoption and Speed of Implementation
Adapting the matrix in Figure 3 to show the time and speed of adoption, as discussed for each of the cases above, gives rise to Figure 5. This pictorial analysis indicates a link between three phenomena: the influences on the adoption decision identified by Sturdy (2004), the time in the innovation life-cycle that the innovation was adopted and the speed of implementation. It can be seen that, as may be expected, adoptions that were influenced by adoption of the same innovation by similar organisations (CC and HT), are associated with later or majority adopters. In contrast, the rationales of the two early adopters (PF and RB) were other than institutional in nature, being political and dramaturgical (Sturdy, 2004). Again, the association of a political rationale with early adoption confirms suggested theory, since the political rationale is often associated with wishing to be associated with new and relatively exclusive ways of working, with such organisations ‘moving on when a fashion becomes too popular’ (Sturdy, 2004, p. 162). The influence of a significant individual or group appears to be associated with more rapid implementation. If an organisation, or the knowledgeable and authoritative individuals within that organisation, are likely to wish to move on once an innovation has become widely adopted, it follows that they should seek to implement it quickly, such that they are positioned to move on when opportune. Although this is a logical
32
The Adoption and Implementation of Managerial Innovations
explanation after the event, it is not obvious that organisations will recognise their rationale for adoption as political a priori. It is also unlikely that they will consciously recognise that the consequence of such decision-making is the likelihood of a relatively short life for the innovation. Indeed, one respondent from RB stressed that before they could be approved, all of its adoptions of managerial innovations were required to provide information on expected lifetime and return on investment, in a similar way to that required for all other investments made by the organisation. He was clear the adoption of a major innovation, such as the BSC, was undertaken with the intention of using the facility or system for some time. If we do spend money to put something in, we would want it to have some life, not just put it in for six months – it takes more than that amount of time to role it out to 40,000 staff. [RB]
7.8.
Level of Achievement
Measuring the success of any initiative is recognised as very difficult because of the multi-dimensional nature of success. This research did not seek explicitly to explore success. However, a number of issues relating to the level of achievement were observed and these are noted here. HT was described in previous sections as implementing its BSC relatively rapidly, particularly compared to the implementation by CC. However, achieving this rapid implementation was a consequence of HT limiting use of the scorecard to the trust board. Hence, whilst that board is satisfied with the implementation, and the scorecard in question meets the identified challenge of maintaining high-level control of the organisation, the ambition was less than that of other organisations implementing scorecards, including the other organisations studied, namely RB (which spread it throughout its staff and branches) and CC (which has tried to spread it across the various service divisions but with varying results). Both PF and CC have been concerned about the length of time implementation has taken, or is likely to take. As noted before, the formation of the PMO by PF was relatively rapid. However, the adoption and consistent use of the project management tools advocated by the PMO is progressing more slowly and likely to take years. The slow rate of progress at CC resulted from, and contributed to, a feeling of a lack of urgency or importance about the scorecard, as described by one interviewee: there was a feeling that if we had let it slip, it would have gone away [CC] The drawn out implementations in both cases appears to have resulted in a gentle lowering of expectations. In the case of PF, there was no requirement for all projects to
Don’t Blame the Tools
33
use the recommended methodology or for project managers to draw on the expertise of the staff in the PMO, which might be regarded as the ideal outcome. Rather, staff were allowed to continue operating their projects as they had done previously, with the hope that they might change their approach and behaviours in the future through realising that there was benefit in doing so. Similarly in CC, whilst there have been some attempts to implement according to an agreed timescale, the separate divisions within CC have been allowed to determine their own timescales and degree of involvement with the scorecard. Already this has resulted in a number of years elapsing without a scorecard being used throughout the organisation.
7.9.
How Long the Innovation Lasted
Only one organisation studied, RB, had ceased using the innovation considered. The political nature of RB’s adoption decision was described above. Whilst those in RB gave a rational reason why they discontinued their use of the BSC (i.e. that is the organisation had lost focus on what was stated to be the most important driver, that of retail sales), the cessation of the scorecard coincided with the appointment of a new chief executive. Hence, it would appear that a political rationale led not only to the adoption of the BSC at RB, but also to it being discontinued. We cannot discuss here how long the other three innovations studied will be used for, since they are still in use and there are no plans in any of the organisations to cease their use. However, it would appear logical, that where innovations are implemented more slowly, the organisations in question are likely to be in use for a relatively longer time; since they are likely to wish to make full use of the innovation once implemented before they decide to discontinue it and move on to another way of working. The case of CC demonstrates how a slow implementation can make it appear that the innovation has been in use for a relatively longer period compared to other organisations. However, it should be noted that this use, although protracted by the slow implementation, is likely to involve only parts of the organisation, as in the case of some divisions within CC; or involve only part of the innovation, as in the case of the project management tools in PF.
8.
Summary of Findings
The key findings from the study are summarised in Figure 6. We suggest that managerial innovations will tend to occur early and be undertaken relatively rapidly when they are associated with an influential individual or group, who in turn are influenced by internal needs and capabilities. Adoption of an innovation should be treated like any other investment decision made by the organisation, provided there is an established
34
The Adoption and Implementation of Managerial Innovations
Figure 6
Impact of Influences on Adoption Decision on Outcome of Adoption.
method of doing this. For example, if it is normal that an investment is preceded by a careful pilot evaluation that feeds into the development of a robust business case that is reviewed and accepted or modified by others within the organisation, then that should be applied. If there are alternative but equally valid ways of evaluating an investment (e.g. by democratic process or consensual professional process), then those should be followed. Early adopters driven by a political rationale for adoption may well discontinue the innovation after a relatively short time in accordance with the ‘distancing perspective’ associated with this rationale by Sturdy (2004) and supported by the RB case study. Perhaps rather counter intuitively then, those organisations that implement innovations well, and can therefore move onto new innovations, appear most ‘faddish’, a term that is often associated with organisations switching rapidly between innovations, as they are either not succeeding with them or not meeting their needs. Where the adoption of an innovation is associated with an influential individual or group but is influenced by external factors (Figure 6, top right quadrant), most commonly the use of the innovation by similar organisations, implementation is also relatively rapid. However, this is likely to take place later, once the innovation has
Don’t Blame the Tools
35
gained legitimacy. In our study, the case in this quadrant of the matrix shown in Figure 6 appeared to demonstrate limited ambition. That is, the staff of the organisation did what they did so straight forwardly, and relatively swiftly, that perhaps they could have achieved more. Whilst further studies will be needed to demonstrate if this is a finding that can be generalised, an underlying explanation may lie in the fact that whilst external influence of the adoption and implementation by other organisations provides lessons that speed up implementation, they also limit exploration or creativity of how the innovation can be adapted to the organisation. Indeed, whilst it is hard to measure degrees of adaptation, HT, which is the case located in this quadrant, showed one of the lowest levels of adaptation of the innovation studied. The organisation did seek to identify four quadrants for its BSC, and measures within each of these that met its needs. However, such level of adaptation is relatively routine in the adoption of a scorecard. The lack of influence by a significant individual or group was associated with relatively drawn out implementations. This is consistent with the literature identifying the importance of a sponsor or champion for any change programme (e.g. Hope Hailey and Balogun, 2003). The study illustrates how important it is to have someone, with influence, who genuinely champions the project, rather than simply identifying or recruiting someone to fill the gap. Whilst such individuals may seek to undertake all necessary tasks, it would appear that they cannot emulate the influence, conviction, or perseverance of a significant individual or group that genuinely champions the innovation. In our study, the organisation associated with a limited influence by a significant individual or group, but with an internal origin or rationale for adoption is PF (bottom left quadrant). This organisation was also linked to a slightly lower level of achievement from its use of the innovation studied, particularly related to the slow up-take of the project management methods that the PMO was seeking to introduce. It would appear that without the driving force of a genuine and influential champion or influential group, particularly when no inducements for use or sanctions for non-use are put in place, up-take and, hence, overall levels of what is achieved are lower than when compared to what could be achieved with complete, or at least more extensive, up-take. The lack of a driving individual or group, and a predominantly external based rationale for adoption appears to result in a later and drawn out adoption and implementation. By definition, such cases will require a champion to be identified, and given the external influence, it appears likely that an external candidate will appear to be attractive, for example due to having gained experience with the innovation elsewhere. However to treat finding a champion as filling a role, rather than incorporating someone with desirable personal characteristics, leads to the same challenges as discussed above. An additional challenge for an outside candidate is that they are unlikely to have the experience of how the organisation works in order to actually get things done.
36
The Adoption and Implementation of Managerial Innovations
The lack of an influential individual or group and a predominantly external influence on adoption is also associated with lower performance than could be achieved, primarily due to the drawn out implementation. The CC case might also be interpreted as indicating that drawn out implementation is associated with a long life of the innovation, even if it is not being used by all intended staff. Hence, in contrast to the early and fast implementers, the later and slower implementers may appear less faddish, and hence by association, the innovation, more successful on grounds of endurance.
9.
Conclusions
The research set out to identify the main influences on the decision to adopt a management innovation and to find how these influences are related to the subsequent implementation, including in the case of organisation-wide innovations, embedding of the innovation throughout the organisation. A proposed relationship, drawn from Figures 3–6, is shown in Figure 7. An event or problem triggers the recognition that the organisation needs to employ a managerial innovation to address the event or problem. A number of influences are then brought to bear on the choice of how the identified event or problem will be addressed. The case studies seem to suggest two particular influences that highlighted both similarities and differences across the cases studied. These are: the presence of an influential individual or group and the degree to which the organisation sought to draw from external parties or from internal capabilities. Use of these characteristics allowed the cases to be differentiated according to Sturdy’s (2004) adoption rationale, suggesting their robustness and utility. Figure 7 also suggests that there is a relationship between the influences on adoption and when the innovation is adopted within its life-cycle and how long implementation takes. Our case studies suggest, as would be expected, that there seem to be some approaches to implementation that are associated with shortening the time for implementation. This is represented in Figure 7, which shows that there is a link between the influences on the adoption decision and when the innovation is adopted and how long implementation takes, with the latter being mediated by how the innovation is implemented. The foregoing discussion of the findings also suggested that how long implementation takes has an influence on the level of achievement. Those whose implementation is prolonged are more likely to realise a lower level of achievement than is possible or they were expecting. However, very rapid implementation can also be linked to limited ambition. The influences on adoption are also seen to provide a link to how long an innovation lasts. In particular, the adoption of innovations that are driven by a political rationale are likely to be short lived, as the powerful figures involved will be keen to move
Don’t Blame the Tools
Proposed Relationship between Adoption and Implementation Characteristics.
37
Figure 7
38
The Adoption and Implementation of Managerial Innovations
on to the next new initiative, and next career move. How long an initiative lasts is also determined to some degree by how long implementation takes, with slower implementations appearing more enduring. The overall conclusion from this study is that the adoption of managerial innovations, sometimes termed management fads and fashions, is a natural consequence of the dynamic environment in which organisations operate. As shown by this study, such innovations are introduced to address an event, problem or challenge facing the organisation. Such events and problems are likely to change in nature over time, and hence, will require new ways of being addressed. ■ As observed in the study, organisations introducing innovations will be keen to find a rational explanation in order to counter potential scepticism that often accompanies such introductions. This study suggests that organisations may use a statement of the problem they are facing as a rational justification. However, there is a subtle but important difference between the problem or trigger and the solution the organisation chooses. Without a clear differentiation of these two distinct stages, organisations are unlikely to evaluate clearly what alternative approaches are available and what their implications are, and so cannot enhance the chances of addressing the problem in the most effective way. ■ This study suggests that two key dimensions in the choice and adoption of managerial innovations are: - influence of a significant individuals or group - origin of influence on the adoption decision (internal or external). ■ Differences along these dimensions are supported by different adoption rationales identified by Sturdy (2004). This identification of different adoption rationales, despite each organisation stating the problem or trigger as a rational reason for adoption, further supports the difference between the trigger event and the influences on the selected solution or approach. ■ The influences on adoption (i.e. significant individual or group, and external or internal influence) also appear to influence when the innovation is adopted and, as mediated by certain approaches to adoption, how long implementation takes. ■ Approaches to adoption that appear to be linked to more rapid implementation identified were: - adapting the innovation to fit the organisation - training of staff, including tailored training to meet their particular needs - pilot, stress testing and a business case that includes how the innovation will be implemented in addition to the benefits that are expected. - link to wider initiatives that are easily understood by staff. ■ Approaches that slow implementation include: - allowing a phased approach to be slowed by groups with varied beliefs, values and methods of operating
Don’t Blame the Tools
39
- introducing an innovation where existing ways of working are adequate - providing benefits to the organisation but not to the divisions or individuals that must use the innovation ■ The key dimensions identified also appeared to have influence on how long an innovation is used for, and, how long implementation took and on the level of achievement and ambition for the innovation. One case study provided insight into a complete life-cycle for an innovation. This suggested that progression to a subsequent innovation is aided by: ■ Complete implementation of the previous innovation ■ Effective use of this innovation for an appropriate period ■ A new event or problem that triggers the need for a new way of working.
10. 10.1.
Implications of the Research Implications for Practitioners
As with much management research, it is impossible to generate a single recipe or set of rules for organisations to follow when adopting and implementing managerial innovations. The research, however, does throw some light on what is good practice: ■ Organisations should recognise that introducing managerial innovations, using them and then discontinuing their use is normal and not a problem to be avoided or sign of failure. New events, problems or challenges will face organisations and they will be required to find new ways of addressing these. Indeed, this research suggests that more rapid and effective implementation is likely to be associated with more rapid change or turnover in ways of working than slower, less effective implementation. However, the significant cost, particularly of organisation-wide innovations, mean organisations must approach their adoption with care. ■ Organisations should seek to separate the event or problem that triggers the need, or perceived need, for the adoption of an innovation and the actual adoption decision. Whilst there is likely to be strong support for addressing the trigger, there may not be such strong support for the means to address the solution. This research shows the importance of an influential individual or group in supporting if implementation is not to be drawn out. ■ Whilst the influence of a significant individual or group has been shown to be important in effective implementation, preparing a robust business case for the
40
The Adoption and Implementation of Managerial Innovations
innovation, including forecast benefits and costs, plus pilot testing the innovation can be important, if in keeping with the organisation’s normal way of working. Preparing a business case that is open to review will encourage the evaluation of alternative innovations and will also prevent the adoption appearing like a ‘pet project’ of an individual or a ‘knee-jerk’ reaction to the recommendation of management consultants or other groups. ■ Linking the introduction of the innovation to a wider programme of change is also shown to be effective. The rationale for such wider programmes may be more easily understood and supported by staff, for example a ‘customer focus programme’, than the less obvious ‘need to improve performance management’. ■ Other approaches to implementation that appear to aid implementation include: ◦ adapting the innovation to fit the organisation ◦ tailoring training to the needs of individuals or groups. ■ The discontinuation of innovations should be actively managed, rather than simply letting them dissipate or fall into disuse. The utility to the organisation of the innovation to be discontinued should be communicated at the relevant time, as should the new event or challenge that is suggesting an alternative innovation is necessary. This approach, which was adopted by the one case, RB, in which an innovation was discontinued, is quite distinct from simply moving on to the next innovation, leaving the paraphernalia of the earlier one to continue along as part of the organisation. Staff should be made aware that the previous innovation was not a mistake, rather it was appropriate for the circumstances at the time, but as those circumstances change, then there is a need for a new approach. The process of a true champion of the next innovation emerging or being identified, and developing a robust business case and piloting or testing the new innovation should then be undertaken.
10.2.
Implications for Management Accountants
Professional accountants should play an important role within their organisation in determining the need for new managerial approaches, in their evaluation and implementation. In addition to illustrating how senior managers within organisations approach the adoption and implementation of such innovations, the foregoing findings and recommendations suggest the following particular roles for accountants: ■ Developing a robust business case. The findings suggest that developing a robust business case to justify and guide the adoption and implementation of a managerial innovation is important. Management accountants and other financial specialists will have a vital role to play in developing and scrutinising such cases, aided by
Don’t Blame the Tools
41
business managers and other specialist colleagues. Such business cases should include estimates of both the expected costs and benefits, both of which can be refined by undertaking the suggested pilot testing. As regards providing relevant information at this stage and the stages below, accounting systems need improving in order that the costs of innovation are not just anecdotal and are not hidden. They also need to signal diversions of resources from existing activities or increases in activities (productivity) that arise from the innovation. ■ Continuous monitoring and assessment. Given the considerable financial and other resource implications of the use of managerial innovations, their implementation and assimilation should be followed by monitoring and assessment, to ensure they remain appropriate to the organisation’s need and remain effective. Finance professionals can have an important role to play in designing and undertaking such monitoring and review. They will also need to establish when it is appropriate to discontinue the use of an innovation. ■ Support from Technology. Getting the best out of a management innovation also depends on how information or data is captured and used. Sometimes this can be quite sophisticated, for example the use of SAP by CC or in some cases, such as HT, fairly basic (use of Excel spreadsheets). Management accountants can play an important role in assessing and developing data systems that will be to the organisation’s benefit.
10.3.
Implications for Academics
We believe that this study contributes to the field of managerial innovations in a number of ways that will provide additional understanding for other academics and also provide the basis for further studies. ■ We believe that this study is the first empirical study of the adoption rationales posited by Sturdy (2004). The findings of this study support his rationales and suggest that they are robust and could form the basis of future studies in this domain. ■ We also believe that this is the first study that seeks to link the stages of adoption of managerial innovation, that is, it seeks to determine the influence of one stage in the process with subsequent stages. In addition to repeating this study in a larger sample to improve reliability, this suggests that studies could be undertaken to identify the influences between other stages of the adoption and implementation process. Such studies would provide a richer and more complete view of the adoption of managerial innovations, rather than the simplistic staged and linear models currently in use. Such studies may allow non-linear or contingent models that have been suggested for new product development to be developed and tested in the field of managerial innovations.
42
The Adoption and Implementation of Managerial Innovations
■ The findings of this study should also be of interest to academics in related fields such as change and project management and performance management. Such domains tend to ignore the rationales, particularly the non-rational ones, that may result in the adoption of a change initiative, project or new performance regime. This study highlights that range of rationales, and the multiple rationales that may lead to such activities and importantly may determine the subsequent approach to adoption, implementation and the level of achievement.
11.
Limitation of the Study and Suggestions for Further Research
Case studies can allow a rich picture of the organisations studied to be obtained, including an understanding of the interrelationships between factors and the unique context of the organisation. However, they tend to be exploratory in nature (Hoskisson, 1999). That is, they can produce propositions or hypothesis for further testing to reach generalisable theories by confirmatory methods such as large-scale surveys. Our findings from this study should therefore be viewed as exploratory in nature and offered as the basis for further research through testing. Particular further studies that would be of value to both academics and practitioners in the domain of managerial innovations include: ■ Inclusion of a wider number of case studies. The current research design, in which we sought to focus particularly on the differences between cases (i.e. the use of the 2×2 matrix), resulted in only a single instance in each quadrant. A study that incorporates multiple cases of each type would not only allow more differences or similarities to be identified, but would also improve reliability of factors within types to be established. ■ Due to those interviewed, our study reflects the views and experiences of senior managers. Further studies of the adoption of managerial innovations could usefully seek to include the view of a wider range of staff and other stakeholders, such as customers, suppliers and external regulatory agencies. Such individuals may have quite different views and experiences of the innovations adopted from senior managers, who may be unaware of the difficulties inherent in the adoption or implementation. ■ Our sample contained three public sector organisations and one for-profit private sector organisation. It may be that differences between influences on adoption and approaches to implementation exist between these two groups, and between them and third sector organisations. For example, in public sector organisations it may be more difficult and undesirable for a significant individual or a group to influence
Don’t Blame the Tools
43
the adoption decision resulting in no obvious champion. Separate studies could usefully be undertaken within these sectors, and the results compared. ■ Finally, for those inclined towards philosophies that encompass generalisability, a larger-scale quantitative study could be undertaken in order to generalise the findings of this study. In particular, a study that seeks to confirm the link between influences on adoption and the time, the speed of implementation and the life-time of innovations as illustrated in Figure 7 could be explored.
References Abrahamson E. 1996. Management fashion. Academy of Management Review 21(1):254–85. Carson P, Lanier P, Carson K, Birkenmeier B. 1999. A historical perspective on fad adoption and abandonment. Journal of Management History 5(6):320–33. Carson P, Lanier P, Carson K, Guidry B. 2000. Clearing a path through the management fashion jungle: some preliminary trailblazing. Academy of Management Journal 43(6):1143–58. Chua WF. 1988. Interpretive Sociology and Management Accounting Research: A Critical Review. Accounting, Auditing and Accountability Journal 1(2):59–79. Dugdale D, Jones TC, Green S. 2006. Contemporary Management Accounting Practices in UK Manufacturing. London: CIMA Publishing. Ehigie B, McAndrew E. 2005. Innovation, diffusion and adoption of total quality management (TQM). Management Decision 43(6):925–40. Eisenhardt KM. 1989. Building theories from case study research. Academy of Management Review 14:532–50. Etorre B. 1997. What’s the next business buzzword? Management Review, September, 33–5. Hope Hailey V, Balogun J. 2003. Exploring Strategic Change, 2nd ed. London: Prentice Hall/FT Publications. Hoskisson RE, Hitt MA, Wan WP, Yiu D. 1999. Theory and research in strategic management: swings of a pendulum. Journal of Management 25(3):417–46. Johnson HJ, Kaplan RS. 1987. Relevance Lost: The Rise and Fall of Management Accounting. Boston, MA: Havard Business School Press. Kaplan RS, Norton DP. 1996a. The BSC: Translating Strategy into Action. Boston, MA: Harvard Business School Press. Kaplan RS, Norton DP. 1996 Jan–Feb. Using the BSC as a strategic management system. Harvard Business Review 75–85. Mintzberg H. 1979. An emerging strategy of “direct” research. Administrative Science Quarterly 24:582–9. Open University. 2004. Managing Projects and Change (B700 Block 3). UK: Open Universtiy. Patton MQ. 1990. Qualitative Evaluation and Research Methods, 2nd ed. Newbury Park, CA: Sage. Rigby D. 2001. Management tools and techniques: a survey. California Management Review 43(2):139–60. Rogers EM. 1995. Diffusion of Innovations, 4th ed. New York: Free Press. Swanson, E.B. (2001), Diffusing software and product and process innovations, IFIP WG8.6 Working Conference, Banff, Alberta, April 7–10.
44
The Adoption and Implementation of Managerial Innovations
Sturdy A. 2004. The adoption of management ideas and practices: theoretical perspectives and possibilities. Management Learning 35(2):155–79. Williams R. 2004. Management fashions and fads: understanding the role of consultants and managers in the evolution of ideas. Management Decision 42(6):769–80. Yin RK. 1989. Case Study Research: Design and Methods. London: Sage.
Further reading Abrahamson E, Fairchild G. 1999. Management fashion: life-cycles, triggers and collective learning processes. Administrative Society Quarterly 44:708–40. Cerdin J. 2003. International diffusion of HRM practices: the role of expatriates. Beta Scandinavian Journal of Business Research 17:48–5. Department of Health, (2007), Background to NHS Foundation Trusts [Online]. http://site320.theclubuk.com/en/Policyandguidance/Organisationpolicy/Secondarycare/ NHSfoundationtrust/DH_4062852. Fenton O’Creevy M. 2003. The diffusion of HR practices in the multinational firm; toward a research agenda. Beta Scandinavian Journal of Business Research 17(1):36–47. Kostova T, Roth K. 2002. Adoption of an organisational practice by subsidiaries of multinational corporations: institutional and relational effects. Academy of Management Journal 45(1):215–33. Ponzi L, Koenig M. 2002. Knowledge management: another management fad? Information Research 8(1):145–60. Spell CS. 1999. Where do management fashions come from, and how long do they Stay? Journal of Management History 5(6):334–48.
Index
Adoption decision 22, 31, 35 Adoption decision matrix 22 Adoption decision phase 18 Adoption implementation phase 18 Adoption rationales 24 Adoption time 32
Management accountants 3, 41 Management ideas 1 Management innovation implementation 3 Managerial ideas 10 Managerial innovations 2 Managerial life-cycle stages 8
Balanced scorecard (BSC) 2
National Health Service (NHS) 16 New public management (NPM) 23
Chartered Institute of Management Accountants (CIMA) 5 Chief executive 22 Commercial software 26 Continuing professional development (CPD) 6 Corporate risk register 21, 23 Corporate services division 27 County council (CC) 17 External management consultants 21
Performance management system 3, 22, 27 Performance measurement 30 Police force 18 Political expediency 31 Political rationale 32 Programme management office (PMO) 2 Project management tools 25, 33 Project managers 33 Public sector 21, 31 RB’s adoption decision 34 Retail bank 17
Influential individual/group 2 Innovation adoption 35 Innovation life-cycle 18 Institutional adoption rationale 29 Intended audience 6 Interviews 15
SAP system 27 Senior responsible owners (SROs) 27 Staff interview 28 Staff responsibility 27 Strategic management board Sturdy’s adoption rationales
47
r
Healthcare organizations 21 HMIC 21 Hospital Trust (HT) 16
Index
Group decision 20
Organisational investments 29