Article Find and Summary

ISD.pdf

Home >Computer Science homework help >Article Find and Summary

Information Systems Research Vol. 20, No. 3, September 2009, pp. 329–354 issn 1047-7047 � eissn 1526-5536 � 09 � 2003 � 0329

informs ® doi 10.1287/isre.1090.0236

Agility from First Principles: Reconstructing the Concept of Agility in Information Systems Development

Kieran Conboy National University of Ireland, Galway, Ireland, kieran.conboy@nuigalway.ie

Awareness and use of agile methods has grown rapidly among the information systems development (ISD)community in recent years. Like most previous methods, the development and promotion of these methods have been almost entirely driven by practitioners and consultants, with little participation from the research community during the early stages of evolution. While these methods are now the focus of more and more research efforts, most studies are still based on XP, Scrum, and other industry-driven foundations, with little or no conceptual studies of ISD agility in existence. As a result, this study proposes that there are a number of significant conceptual shortcomings with agile methods and the associated literature in its current state, includ- ing a lack of clarity, theoretical glue, parsimony, limited applicability, and naivety regarding the evolution of the concept of agility in fields outside systems development. Furthermore, this has significant implications for practitioners, rendering agile method comparison and many other activities very difficult, especially in instances such as distributed development and large teams that are not conducive to many of the commercial agile meth- ods. This study develops a definition and formative taxonomy of agility in an ISD context, based on a structured literature review of agility across a number of disciplines, including manufacturing and management where the concept originated, matured, and has been applied and tested thoroughly over time. The application of the texonomy in practice is then demonstrated through a series of thought trials conducted in a large multinational organization. The intention is that the definition and taxonomy can then be used as a starting point to study ISD method agility regardless of whether the method is XP or Scrum, agile or traditional, complete or fragmented, out-of-the-box or in-house, used as is or tailored to suit the project context.

Key words: agile; systems development; conceptual research; agile manufacturing; agile management History: Sandra Slaughter, Senior Editor and associate Editor. This paper was received on June 1, 2007, and

was with the authors 9 12 months for 3 revisions. Published online in Articles in Advance August 31, 2009.

The Emergence of Agile Methods The last 10 years or so have seen the emer- gence of a number of information systems develop- ment (ISD) methods,1 which have collectively been labeled as agile. Some of the most popular include

1 The term ISD method is often vaguely defined, and extensive research has attempted to provide some clarity, distinguishing ISD method from methodology (e.g., Avison and Fitzgerald 2003, Checkland 1981, Oliga 1988, Vonk 1990, Welke 1983, Brinkkemper 1996), project management method (e.g., Brinkkemper 1990), process (e.g., Connors 1992, Fitzgerald et al. 2002), and practices (e.g., Iivari et al. 2000, Wynekoop and Russo 1995). The interpretation of method in this study is an inclusive one, amalgamating the definitions of Hirschheim et al. (1995), Brinkkemper (1990), and Wynekoop and Russo (1995):

An ISD method encompasses the complete range of practices involved in the process of designing, building, implementing, and maintaining an information system, how these activities are accomplished and managed, the sequence and frequency of these activities, as well as the values and goals of all of the above.

eXtreme Programming �XP� (Beck 1999), the Dynamic Systems Development Method �DSDM� (Stapleton 1997), Scrum (Schwaber and Beedle 2002), Crystal (Cockburn 2001b), Agile Modeling (Ambler 2002), Feature Driven Design �FDD� (Coad et al. 1999), and Lean Software Development �LSD� (Poppendieck 2001), along with variants of each, e.g., XP-Lite (Aveling 2004).2 These

2 Methods are often distributed and communicated in different ways, namely through manuals, research papers, consulting, men- toring, etc. For consistency this study will refer to the version of each method in the associated reference above. It is important to distinguish between the “documented” method and the way it may be used in a particular situation (commonly referred to as “method- in-action” (Fitzgerald et al. 2002)). Iivari and Maansaari (1998) sug- gest that a method may be interpreted not as a set of “rules,” but as “an ideal in the sense that it is not expected to be followed literally.” In more recent years the notion of method as an idea and enactment, as opposed to a rigorous prescription, has become widely accepted in ISD thinking and is therefore the interpretation adopted in this study.

329

methods have been well received by those in ISD, and there is strong anecdotal evidence to suggest that awareness and indeed use of these methods is highly prevalent across the community.

Problem Statement As is often the case with new and emerging phenom- ena in ISD, agile method practice has led research, with the creation, promotion, and dissemination of these methods almost completely due to the efforts of practitioners and consultants. Now agile method research is beginning to gain momentum, as is evi- dent from the increasing number of dedicated jour- nal special issues, conferences, conference tracks, and workshops. As a result of the dominating role industry played in the early years, however, almost all existing agile method research is now built on these industry-driven methods. Although many dif- ferent studies have attempted to answer many dif- ferent questions, the overwhelming majority base their study on XP, Scrum, or an individual practice within these methods. In many ways, this is very positive and represents a welcome departure from much IS research, which is all too often maligned for being pretentious and lacking relevance to practice (Benbasat and Zmud 1999, Galliers 1994). While there are merits to adopting such a practice-

oriented focus, little if any research effort has focused on the conceptual development of agility in the ISD field. Consequently, it can be argued that the current body of agile method knowledge now suffers from a number of conceptual problems as follows. Lack of Clarity. One of the most fundamental at-

tributes of a concept is that it is clearly communi- cated and understandable (Metcalfe 2004; Dubin 1976, 1978; Weick 1989). While early promotional material such as the agile manifesto (Fowler and Highsmith 2001) portrays agile ISD as a clear, simple, and cohe- sive concept, the reality is much different. Firstly, hardly any two agile method texts or papers adopt the same definition of agility or agile method. Agility as a concept is highly multifaceted and has been used by many different people to refer to very different phenomena. To state that a particular method is or is not agile is almost meaningless given the lack of consensus on what the term “agile” refers to. Sec- ondly, many different agile methods, variants, and

derivatives exist, and yet it is not so much the num- ber of methods that causes the problem but the fact that these are so disparate. Some represent prescrip- tive operational instructions for developers (i.e., XP), others bear closer resemblance to project manage- ment methods rather than ISD methods per se (i.e., Scrum), and yet more can best be described as a set of philosophical principles (i.e., Poppendieck’s Lean Software Development). In extreme cases these meth- ods are even contradictory (e.g., XP requires collec- tive code ownership, FDD demands individual code ownership). It is logical and perhaps inevitable that different method creators will have different ideas on how agility can be achieved, and providing a diverse range of options for ISD teams must be beneficial for them. However, it could prove very challenging and confusing for ISD teams who wish to be agile when they are given completely conflicting, polar opposite advice. Such inconsistency and contradiction evokes similar emotive descriptions to those leveled at IS research in general, such as “fragmented adhocracy” (Banville and Landry 1989), and “crucial but con- fused” (Checkland and Holwell 1998). Lack of “Theoretical Glue.” Behind any good con-

cept or theory there should be a strong underly- ing logic and rationale. Whetten (1989) refers to this as a “theoretical glue” that should bind all the fac- tors together. However, an analysis of the practices within XP, Scrum, and other commercially labeled agile methods reveals that while some have clear links to the concept of agility in a traditional sense, the con- necting logic behind many others is not as clear. Sys- tem metaphor and pair programming again come to mind; while short iterations, close links with the cus- tomer and co-location have related concepts in agile manufacturing and agile management; where in agile manufacturing or management have we ever heard of agility being increased by people working in pairs? Lack of Cumulative Tradition. A good concept or

theory should cumulatively build on existing research (Dubin 1978), yet this is something that IS researchers have not done particularly well (Benbasat and Zmud 1999; Keen 1991, 1980). Keen (1991) notes that most concepts and areas of concern in IS research are not as “new” as often claimed, and often “turn out to have long roots.” This trend seems to continue where agile method research is concerned. Firstly,

some (such as Boehm and Turner 2003) have noted that agile methods are incorrectly perceived as rev- olutionary and that there is a false dichotomic split between agile and every other method that went before, where commentators fail to realize that most agile practices have origins in much older methods. Secondly, while many researchers are studying agile methods and learning many valuable lessons, there are few that compare agile method projects with those using traditional approaches, which one would expect when a new range of methods emerges and claims to be superior. Thirdly, one would expect that stud- ies of ISD agility would have drawn on the exist- ing bodies of knowledge regarding agility, flexibility, and leanness in other disciplines such as manage- ment and manufacturing, where these concepts orig- inated, matured, and have been applied and tested thoroughly over time. While there are occasional ref- erences, there are hardly any ISD studies that embrace and reflect on this tradition, with most empirical studies of agile methods using XP, Scrum, or some other commercial agile method or individual prac- tice as a point of departure from the literature. Fur- thermore, there is evidence to suggest that this lack of cumulative reflection on past research is continu- ing. For example, in the new body of agile method research, there are many studies documenting how agile methods have been tailored, fragmented, and even combined (e.g., Fitzgerald et al. 2006, Aydin et al. 2004, Bowers et al. 2002, Cao et al. 2004, Fenwick 2003). Yet rather than taking the new method variant forward, subsequent researchers inevitably return to “base camp” and relay yet another account of how the original textbook version was tailored or customized. While such studies still have merit, it might be ben- eficial to see a variant from one study taken forward by multiple subsequent researchers and subjected to further implementation, stress-testing, validation, and refinement. Rather than having countless variants of a commercial method, a survival-of-the-fittest approach such as this might then see the emergence of a new, well-tested, robust method variant. Lack of Parsimony. Authoritative works on con-

cept development usually advocate a parsimonious approach, removing any factors that provide little additional value to our understanding (Whetten 1989). However, if we were to compile a list of all

ISD methods, practices, processes, tools, and artifacts that are commercially labeled as agile, then we would surely find redundancy and duplication. Even within individual agile methods there is a lot of nonvalue- adding content. For example, XP’s system metaphor practice is rarely used despite the popularity of the parent method (Succi and Marchesi 2001, Khaled et al. 2004, Fowler 2001). If so, then one would log- ically expect that method developers either provide more substantive advice on how the barriers affect- ing adoption of that practice can be overcome, or else issue a revised version of the method description with the redundant practice removed. Even the most recent studies involving XP still include this practice. A cohesive research effort would continuously reflect on such anomalies and avoid wasting time on them in future. Limited Applicability. The range of applications

of a concept is a key criterion for judging its qual- ity (Metcalfe 2004; Dubin 1976, 1978; Weick 1989), so agile methods should be applicable in a wide variety of ISD contexts. Irrespective of justification on purely conceptual grounds, much research in the ISD com- munity itself has highlighted the importance of broad ISD method applicability and have called on method developers to consider ways in which their meth- ods can be “transferred” from concept to as wide a diaspora of commercial development environments as possible (e.g., Lings and Lundell 2004, Lundell et al. 2004). Despite this, some argue commercial agile methods are largely restricted to small, co-located development teams, noncritical system development, on-demand access to developers, and many other such constraints (e.g., Stephens and Rosenberg 2003). It must be noted that some vehemently argue that agile methods are applicable in broader contexts, but the sheer volume of research attempting to rebuild and tailor these methods to various environments (e.g., Lindvall et al. 2004, Kahkonen 2004, Bowers et al. 2002, Cao et al. 2004, Crispin and House 2003, Stotts et al. 2003, Sarker 2009, Cummings et al. 2009) suggests this might not be so simple.

Implications for Practice One may be tempted to argue that the issues described above are not that significant. After all, it is already a well-established fact that the IS literature is eclectic

and inconsistent (Checkland and Holwell 1998, Miller 1993, Backhouse et al. 1991)—universally accepted meanings of even such basic terms as information sys- tem, information system development, method, and prac- tice have proven elusive. In fact, almost every field and discipline is saturated with ambiguous terms and inconsistent interpretations, and this is not just symp- tomatic of the softer sciences, as even the history of biology, chemistry, and physics is strewn with com- peting and conflicting definitions, theories, and mod- els (Bryson 2004). Secondly, it is widely accepted that IS researchers are often preoccupied with theory and rigor to such an extent that relevance to practice is often forgotten (Benbasat and Zmud 1999, Galliers 1994), and one could contend that zealously chasing an overarching definition and concept of agility is yet another illustration of researchers valuing rigour at the expense of relevance. While there is certainly merit to these arguments, there is always some level of justification that can be gleaned from the oft-quoted statement “there is nothing as practical as a good the- ory.” However, in this case there is more substan- tial justification available, because the lack of concep- tual foundation regarding agility in ISD does actu- ally present significant problems for practice. Such an absence means that assessing agility is achieved by measuring compliance to commercial agile methods such as XP and Scrum. While this might yield interest- ing results and insights, and has underpinned many previous agile method studies, there are a number of limitations associated with such measurement as follows. Encouragement of Method Improvement. Much

research and practitioner attention has focused on the importance of continuous reflection on methods and method use, and the encouragement of ongo- ing process and method improvement (e.g., Paulk et al. 1995, Hunter and Thayer 2001). The position taken in this study is that, in contrast, compliance- based assessment of agility encourages a more dog- matic adherence to a method and might discourage improvement. Some agile methods such as XP are highly prescriptive and operational. An over-arching concept-centric view of agility would allow practi- tioners and researchers to critically reflect on methods such as XP, and constantly find ways of extending or tailoring the method to become “more agile.” A clear

symptom of this problem is that, while many studies describe cases where agile methods have been prob- lematic and consequently tailored, there is no con- solidated body of knowledge bringing these studies together and proposing how these methods can be improved upon—the textbook version of XP, Scrum, and other agile methods are still the only point of ref- erence for developers when evaluating and choosing agile methods. Method Comparison. As discussed earlier, com-

mercial agile methods differ greatly in terms of abstraction, goals, practices, tools, and many other features. Therefore, method comparison is akin to comparing apples and oranges; it is extremely diffi- cult for developers to decide which method is supe- rior, or which should be chosen as the “baseline” method against which others are assessed. Given the number of existing agile methods, as well as the need to consider hybrids and variants, comparison of agility across methods is surely important to develop- ers and project managers who wish to choose between methods in a structured, intelligent manner. Assessment of Traditional or In-House Methods.

Research has shown that developers rarely adhere to commercial methods, and that the vast majority use tailored variants or in-house creations. In fact, some studies show that rigorous adherence to meth- ods among the ISD community could be as low as 5% (e.g., Hardy et al. 1995; Fitzgerald 1997, 1998). In some instances the variant or in-house method might be similar to a commercial agile method such as XP or Scrum. However, given the complex and diverse nature of systems development, the practices of an in- house method might bear little resemblance to those of any commercial agile methods, and in such cases, assessing agility by compliance might be difficult. This does not mean, however, that the method is auto- matically nonagile. The in-house method might fulfill all the same goals and values as XP, Scrum, and other agile methods but might do so by radically different means. Assessment of Environments Unsuitable for Com-

mercial Agile Methods. Rewarding compliance to a commercial agile method might provide an undesir- able incentive for developers to adopt a “square peg in a round hole approach,” forcing these methods into situations to which they are not suited. An analysis

of the literature bears this out, with the vast majority of existing agile method research focusing not on small co-located teams, as was originally intended, but rather on nonstandard implementations, i.e., large teams (Lindvall et al. 2004, Kahkonen 2004, Bowers et al. 2002, Cao et al. 2004, Crispin and House 2003), start-ups (Auer and Miller 2002), distributed develop- ment environments (Kircher et al. 2001, Stotts et al. 2003), greenfield sites (Rasmusson 2003), educational environments (Fenwick 2003, Johnson and Caristi 2003, McDowell et al. 2003, Melnik and Mauer 2003, Wainer 2003), open source development (Kircher and Levine 2001), outsourcing (Kussmaul et al. 2004), and systems maintenance (Poole and Huisman 2001). Facilitation of a Staged Conversion to Agility.

Through a compliance-based view of agile method adoption, the literature identifies two ways to gradu- ally introduce an agile method into an organization, either through “testing the water” by starting with less significant projects (e.g., Nielsen and McMunn 2005), or through the introduction of a few practices at a time (e.g., Fenwick 2003). By moving beyond a com- pliance model, gradual introduction can be achieved via a third alternative, namely by adopting a mildly agile practice and then replacing it with more adven- turous ones in the future. So, for example, rather than adopting XP’s pair programming practice, which has proved to be highly problematic and disruptive in some circumstances (Cockburn and Williams 2001, Williams and Kessler 2000, Stotts et al. 2003, Miller 2002), pairs of developers could start by just review- ing each others code after it is written, or just pair- ing on certain parts of code, thus introducing the practice on a gradual basis. While much research has focused on the importance of gradual or incremen- tal method adoption (e.g., Lundell et al. 2004, Moody 2002), the issue has not received any attention in agile method research, and as far as this researcher is aware, no agile method lists optional practices with varying degrees of extremism. While the Crystal family (Cockburn 2001b) contains methods with vary- ing degrees of extremism, these are intended to be used in different environments and not to gradually replace each other as stepping stones to agility. Assessment of Adherence to Overarching Values

and Goals. As defined on p. 329, a “method” con- tains not just a set of practices or techniques but also

the values and goals that bind those practices and techniques together. It is important to look at the method level and not just the practice level to ensure that practices across a method are not conflicting, but instead are all aligned to a common set of goals and values (Brinkkemper et al. 1999, Ågerfalk and Wis- trand 2003, Ågerfalk and Fitzgerald 2005). In contrast, however, measuring compliance to practices or tech- niques of a commercial method such as XP focuses at an operational level and does not consider the con- vergent or divergent nature of the higher level goals behind those practices. An over-arching conceptual statement of agility specifying these values would allow such an assessment to take place. The aim of this research is to provide a start toward

overcoming the issues outlined above by providing a rich definition and conceptualization of agility in an ISD context, based on a literature review of agility across a number of disciplines including manufactur- ing and management where the concept originated, matured, and has been applied and tested thoroughly over time.

Research Approach A methodological review of past literature is a “cru- cial endeavour” (Webster and Watson 2002) for any academic research, and it is vital that this is done in a rigorous and comprehensive manner (Webster and Watson 2002, Walsham 2006, Levy 2006). The litera- ture analysis strategy and approach is rarely given much attention, with most papers elaborating more on the empirical data collection and analysis phases of the research. Because this study is based purely on a review of existing literature, however, the approach to this is now discussed in detail. Multidisciplinary Sources of Literature and Exclu-

sion of ISD Research. Agility is not a concept unique to ISD, first appearing in the mainstream business lit- erature circa 1991, e.g. Goldman et al. (1991). Since then the term has become widely used across many fields and disciplines within the business domain. Research efforts within some of these fields have already focused on agility as a concept and have used conceptual and empirical means to eliminate ambiguity and to move toward a single representa- tion of agility through the development and refine- ment of constructs and dimensions (e.g., Sharafi and

Zhang 1999, Vokurka and Fliedner 1998, Sharafi et al. 2001, Katayama and Bennet 1999). The purpose of this study is to contribute to ISD research by reviewing this literature and bringing it into the ISD domain. Because of this objective, and because there is a dearth of similar literature in ISD research in any case, the literature analysis strategy explicitly excluded ISD lit- erature, and focused only on management, manu- facturing, organizational behavior, and other relevant business research. Analysis of Concepts Underpinning Agility.

While it is clear that the concept of agility has been subjected to more rigorous analysis in other disci- plines and that ISD can learn from these efforts, one would be misguided to suggest that agility in other disciplines is fully and clearly understood. It would be similarly erroneous to think that all the limita- tions behind the way we view agility in ISD can be overcome by simply sourcing a suitable and uni- versally accepted concept from elsewhere. Concep- tual research across disciplines has shown agility to be highly polymorphous and not amenable to sim- ple definition (Sharafi and Zhang 1999, Towill and Christopher 2002, Vokurka and Fliedner 1998, Sharafi et al. 2001, Katayama and Bennet 1999, Dove 1995). In fact, agility in manufacturing has been described as “a ragbag collection of tools and practices look- ing for a purpose in order to become a concept” (Sharafi et al. 2001). This suggests that the search for a definitive, all-encompassing concept of agility might not be found simply through an examination of agility in other fields; rather, it might be found through a “first-principles” approach, examining the underlying concepts of agility, namely flexibility and leanness (Sharafi and Zhang 1999, Towill and Christo- pher 2002) that have much earlier origins. For exam- ple, lean thinking can be traced back to the Toyota Production System in the 1950s, with its focus on the reduction and elimination of waste (Ohno 1988), the production of the Spitfire airplane in World War 2 (Childerhouse et al. 2000), and even as far back as the automotive industry in 1915 (Drucker 1995). To reflect this past, this study examines the literature on flexi- bility and leanness and compares and contrasts both to agility before arriving at a final conceptualized def- inition of agility itself.

Figure 1 Concept Development

Flexibility

Flexibility vs. agility

Leanness vs. agility

Definition of ISD agility

Leanness

Concept-Centric Approach. An author-centric lit- erature review usually fails to adequately synthe- size the literature and allow critical, constructive concept development and so should normally be concept-centric where possible (Levy 2006, Webster and Watson 2002). This was particularly pertinent in this study (see Figure 1), where the objective was to critically examine the concept of agility. A concept matrix was used in this study (see Figure 1), as rec- ommended by Webster and Watson (2002) and Sali- pante et al. (1982), whereby the main concepts (agility, flexibility, and leanness) and their underlying subcon- cepts were mapped against all the literature reviewed (see excerpt in Table 1). This matrix was then used to identify the most important concepts and logical ways to group and present them. The structure for the remainder of the paper and the development of the definition and conceptualization of agility were then based on this logical grouping. Incremental Concept Development. One potential

drawback of this study is that it criticizes the cur- rent body of knowledge for having a plethora of

Table 1 Excerpt from the Concept Matrix Adopted in This Study∗

Concept

Flexibility Leanness Article Creation Proaction Reaction � � � � � � � � � � � �

Gerwin (1993) X X X � � � � � � � � � � � � De Groote (1994) X X � � � � � � � � � � � � Upton (1994) X X � � � � � � � � � � � � Golden and Powell (2000) X X � � � � � � � � � � � � Eppink (1978) X � � � � � � � � � � � � Gustavsson (1984) X � � � � � � � � � � � � Piore (1989) X � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �

∗The complete matrix contained 3 concepts, 13 subconcepts against a total of 195 references.

conflicting definitions of agility, yet then paradoxi- cally proceeds to add another one to the list. A key differentiator and contribution of this study, how- ever, is that the definition of agility is developed incrementally, through a total of 16 iterations, unlike most studies of agility, which lead with a “take it or leave it” definition. The arguments for and against the inclusion of each conceptual component are dis- cussed separately, along with the impact of each on the definition, thus allowing future researchers to use the parts of the definition they agree without having to take the remainder. Consistent Abstraction. When developing any def-

inition or concept, it is always difficult to decide what level of granularity should be used. Every researcher faces a trade-off between focus and multi- dimensionality and between comprehensiveness and memorability (DiMaggio 1995). In assessing the volume of literature reviewed in this study, the researcher may have been tempted to include many different concepts, philosophies, methods, tools, and practices. However, bearing in mind that the objec- tive of this study is to present a consistent, clear def- inition, and conceptualization, adding some structure to agility in ISD, the researcher erred on the side of focus and memorability, adopting what Sutton and Staw (1995) call “strategic reductionism.” All the arti- facts relating to agility, regardless of level of abstrac- tion were coded and grouped into a series of high level “intellectual bins” (Miles and Huberman 1999), with the definitions and conceptual development then based on these.

Toward a Conceptualized Definition of Agility Flexibility Merriam-Webster’s dictionary3 highlights the com- plexity of the adjective flexible, defining it as “a ready capability for modification or change, by plasticity, pli- ancy, variability and often by consequent adaptability to new situations.” This complexity could be exacer- bated further if one tried to understand flexibility fur- ther by drilling down on plasticity, pliancy, and each of the other terms within this definition, which by

3 Merriam-Webster’s Collegiate Dictionary, 11th ed.

themselves are inherently unwieldy and difficult to define. To reach a purposeful definition of ISD method flexibility, however, this research started with one of the most simple and popular interpretations of the term, namely, “the ability to adapt to change.” The first working definition for this study therefore reads simply as “the ability of an ISD method to adapt to change.” The words “proactively” and “reactively” were then

inserted. Flexibility is traditionally considered solely as a response to environmental uncertainty. Gerwin (1993), however, introduced the notion that flexibility might be used in a proactive manner, recognising the fact that an entity is not helpless while waiting for change to occur and that steps can be taken in advance of change as well as in response to it. This is sup- ported by other conceptual studies of flexibility (De Groote 1994, Upton 1994), although this is sometimes referred to as offensive versus defensive (Golden and Powell 2000) flexibility, or initiative versus response (Goldman et al. 1995, Upton 1994). The words “proac- tively” and “reactively” were therefore inserted, and after making this adjustment the revised definition becomes “the ability of an ISD method to proactively or reactively adapt to change.” The next adjustment was to insert the word “inher-

ently” into the definition of ISD method flexibility. Robustness (Hashimoto 1980), resilience (Hashimoto et al. 1982), or mobility (Upton 1994) are included as components of flexibility, referring to the ability to endure all transitions caused by change, or the degree of change tolerated before deterioration in per- formance occurs without any corrective action. This concept indicates that a truly flexible entity might be capable of absorbing some change without having to take any action at all. The revised definition now appears as “the ability of an ISD method to proac- tively, reactively, or inherently adapt to change.” The term “adapt to change” was then replaced with

“embraces change.” When Golden and Powell (2000) made the distinction between defensive and offensive strategies discussed earlier, the rationale involved more than just the difference between taking steps before and after change. In addition, they noted that when change occurs, not only can an entity attempt to return to its original state, but it can take advantage of the change to place itself in a better position. The

term “adapt to” implies that an entity is homeostatic and that its only objective in the face of change will be to return to its original state. “Embrace,” on the other hand, implies that the entity not only might try to return to its original state but might capital- ize on the incoming change and improve its position. After making this adjustment the revised definition becomes “the ability of an ISD method to proactively, reactively, or inherently embrace change.” The next adjustment was to insert the term “create

change” into the definition. While some researchers indicate that a flexible entity can take steps to ben- efit from an incoming change, Gerwin (1993) and Piore (1989) extend this notion even further. As well as identifying a change, and attempting to benefit from it, flexibility can also refer to the ability “not just anticipate change, but create it” (Piore 1989), driving new change that would never have occurred were it not for the entity’s actions, e.g., creating more uncertainties for rivals, thus establishing a power- ful competitive advantage (Gerwin 1993). As well as justifying the introduction of the term “create change,” this argument of Gerwin and Piore also pro- vides further justification for the earlier replacement of “adapt to” with “embrace.” The words “adapt to” implies that change is the driving force and the entity’s actions are as a result of that force. “Embrace” signifies a two-way process where the entity not only reacts to change but can also influence it. The defini- tion of ISD method flexibility now becomes “the abil- ity of an ISD method to create change, or proactively, reactively, or inherently embrace it.” The literature also highlights a distinction between

flexibility that is inside and that which is outside the boundaries of the entity in question (Golden and Powell 2000). Van Oyen et al. (2001) illustrate this concept through labour flexibility, referring to inter- nal flexibility as the ability of an organization to vary employee’s duties, working hours, or salaries; while external flexibility refers to the ability of an organization to draw resources through subcontrac- tors, short-term contracts, or temp agencies. In the context of an ISD method as defined earlier in this study, this means that a method’s flexibility refers not just to how its practices, proceses, values, and goals govern the project team but also how they facilitate and impact external people, entities, processes, and

structures. Therefore, the words “through its internal components and relationships with its external envi- ronment” are inserted to reflect the fact that an entity might not be a closed system. Rather, it might interact with other systems in its environment and might be able to use these interactions to handle change. After incorporating these changes, the revised definition of flexibility now reads as “the ability of an ISD method to create change, or proactively, reactively, or inher- ently embrace it, through its internal components4

and relationships with its environment.” Much of the literature proposes speed of response as

a key attribute of flexibility (Eppink 1978, Gustavsson 1984, Goudswaard and de Nanteuil 2000). Golden and Powell (2000) describe the temporal dimension of flex- ibility as the “length of time it takes for an organiza- tion to respond to environmental change” or to “adapt within a given time frame.” However, careful word- ing is required, as references must not include abso- lute speed as a measure of success. Volberda (1998) compares time taken to adapt to change against the variety of that change, acknowledging the fact that rapid response to familiar change is not necessarily better than a slightly slower response to large, strategic change. Therefore, the words “in a timely manner” are included in the definition to illustrate that it is impor- tant to consider speed of response, but that it is relative to the nature of the change being made.

Final Definition of Flexibility The final definition of flexibility is “the ability of an ISD method to create change, or proactively, reac- tively, or inherently embrace change in a timely man- ner, through its internal components and relationships with its environment.”

Flexibility vs. Agility In many ways, the terms flexibility and agility are very similar and have often been used interchangeably throughout the literature (Agarwal et al. 2006). An analysis of the relevant conceptual literature reveals

4 The term “internal components” might invoke many connota- tions. For the purposes of this study an ISD method’s components are those incorporated into the earlier definition of an ISD method on p. 329, i.e. practices, activities, sequence or frequency of activi- ties, values, and goals.

that the term agile has come to mean “proactive” (Nay- lor et al. 1999), “reactive” (Stratton and Warburton 2003), “change-embracing” (Goldman et al. 1995), and all the other terms that exist in the definition of flexi- bility proposed above. One component of the definition that does per-

haps require justification is that of change creation, as this, at least in the researcher’s mind, would not have been automatically assumed to be part of agility. However, from an analysis of the agility literature, this component clearly should be included. Agility, according to Gunasekaran and Yusuf (2002), focuses on “new ways of running businesses” and “cast- ing off old ways of doing things.” Agility requires “distinctly aspirational tendencies” (Stratton and Warburton 2003) and involves “exploration” (Yusuf et al. 1999), “opportunity exploitation” (Christopher and Towill 2000), “acquiring new competencies, devel- oping new product lines, and opening up new mar- kets” (Vonderembse et al. 2006). Gunasekaran and Yusuf (2002) cite numerous other examples in the agility literature to support this notion (e.g., Kidd 1994, Gould 1997, Hong et al. 1996, James-Moore 1996). Such literature should dispel any concerns regarding the validity of incorporating change cre- ation within an overall theory of agility. Although there are many conceptual similarities

between flexibility and agility, there are also some key differences however, and the definition of flexibility must be changed and modified to reflect these before it can be applied to agility. Firstly, Lindbergh (1990) and Sharafi and Zhang (1999) indicate that agility is made up of two components. The first is flexibil- ity, but it shares equal prominence with the second, which is speed. Essentially, an organization must be able to “respond flexibly” and “respond speedily” (Breu et al. 2001). Terms such as “speed” (Tan 1998), “quick” (Kusiak and He 1997, De Vor and Mills 1995, Gunasekaran et al. 2002, Yusuf et al. 1999), “rapid” (Hong et al. 1996), and “fast” (Zain et al. 2002) occur in most definitions of agility. This reference to speed was discussed within the context of flexibility and was the justification behind inserting the words “in a timely manner” in the earlier definition. However, because research on the definition of agility has placed such emphasis on rapidity, it merits an adjustment to the definition before it can be applied to the term

agile. Therefore, the words “in a timely manner” are removed and replaced by the word “rapidly.” Also, the term is inserted at a more prominent point in the definition to highlight its importance. It is coupled with the word “inherently” as an extreme example of to do something rapidly is to do something inher- ently, i.e., where time taken equals zero. The defini- tion now becomes “the ability of an ISD method to rapidly or inherently, create change, or to proactively or reactively embrace change through its internal com- ponents and relationships with its environment.” Another distinction between agility and flexibility

is that where agility is concerned, there exists the assumption that change is continuous, and embrac- ing it is an ongoing activity. This assumption was laid down in the key contribution of Goldman et al. (1995), where they described agility in general terms as “a continual readiness to change.” The flexibility literature, and therefore the definition as it stands, makes no reference to continual change as opposed to a once-off change. Therefore, the definition is mod- ified further, replacing “the ability of an entity” with “the continual readiness of an entity.” The definition of ISD method agility now becomes “the continual readiness of an ISD method to rapidly or inherently create change or to proactively or reactively embrace change through its internal components and relation- ships with its environment.” Another concept emphasized heavily in the agility

literature is the inclusion of knowledge and learning, and specifically, knowledge and learning from change. Jin-Hai et al. (2003) point to various literature (e.g., Prahalad and Hamel 1990, Senge 1998), which dis- cuss how the “learning barrier” renders “an enter- prise unable to keep up with environmental change.” Jin-Hai et al. (2003) suggest that various learning activities should be used within an agile enterprise, such as R&D, procession learning or “learning by doing.” Many have tried to develop a set of con- cepts and characteristics of an agile entity. Across these studies, characteristics proposed include “the learn- ing employee” (Towill and Christopher 2002), “knowl- edge workers” (Yusuf and Adeleye 2002) “information enrichment” (Christopher and Towill 2000), “knowl- edge sharing” (Breu et al. 2001), the “knowledge- driven enterprise” (Yusuf et al. 1999), and the “learn- ing organization” (Gunasekaran and Yusuf 2002). With

so many constructs evident throughout the various models of agility, the inclusion of learning in the definition is merited. Therefore the definition now becomes “the continual readiness of an ISD method to rapidly or inherently, create change, proactively or reactively embrace change, and learn from change, through its internal components and relationships with its environment.” For some, agile means to apply the concepts of

flexibility throughout different parts of the organiza- tion and not to a specific part such as manufactur- ing or production processes (Katayama and Bennet 1999). This has led to the coining of terms such as “agile supply chains” (Christopher 2000), “agile decision support systems” (Huang 1999), and “agile workforce” (Van Oyen et al. 2001). Therefore, it is not appropriate to state that agility is flexibility applied to different parts of the organization. How- ever, some suggest that agility is flexibility with an “organizational orientation” (Christopher 2000), in that it is applied collectively throughout the enter- prise (Goldman et al. 1995, Preiss et al. 1996). This notion would be in line with Goldman and Nagel’s (1993) “agile enterprise,” Nagel and Dove’s (1991) opinion that agility must be viewed in a “business- wide context,” and that of Gunasekaran et al. (2002), which states that agility is “not a series of prac- tices but a fundamental management philosophy.” To reflect these, the definition is amended, chang- ing the words “through its internal components” to “through its collective components.” The definition now appears as

Definition of Agility Part 1 � � � the continual readiness of an ISD method to rapidly or inherently create change, proactively or reactively embrace change, and learn from change, through its collective components and relationships with its environment.

Leanness In comparison to the concept of flexibility, the notion of leanness is somewhat more straightforward. Most literature traces lean thinking back to the Toyota Pro- duction System in the 1950s (Ohno 1988), although as stated earlier, some have traced leanness back to World War 2 (Childerhouse et al. 2000) and even as far back as 1915 (Drucker 1995). Early lean thinking

at Toyota was initially applied to car engine manufac- turing but extended to vehicle assembly in the 1960s and the wider Toyota supply chain in the 1970s. It was only at the very late stages of this evolution that formal manuals were developed and the “secrets” of this lean approach were shared with companies out- side Toyota for the first time (Hines et al. 2004). In fact, these manuals were written in Japanese, and it took almost another decade before the first English literature was available (Ohno 1988, Monden 1983, Shingo 1981, Schonberger 1982). Even then, it did not make a significant impact in the mainstream literature until MIT’s five-year study of the automotive industry identified huge productivity differences between the United States and Japan, attributing this difference to a lean approach (Womack et al. 1990). For many years, leanness has focused on the prin-

ciple of economy, although this is usually referred to as cost reduction (Ohno 1988), “the elimination of waste” (Womack et al. 1990, Naylor et al. 1999, Ohno 1988), or “doing more with less” (Towill and Christopher 2002). Utilization of all resources is max- imized, and no unnecessary resources are maintained (Ohno 1988). Just-in-time production, zero inventory (Womack et al. 1990), kanban production (Shingo 1981, Monden 1983, Ohno 1988), and maintaining minimum reasonable inventory (MRI) (Grunwald and Fortuin 1992) are all practices and approaches that illustrate the principle of economy. Therefore, the starting definition of leanness adopted in this study is: “contribution to economy.” Leanness also incorporates many references to qual-

ity, none more so than total quality management (TQM), a philosophy central to lean thinking (Hines et al. 2004). As the concept emerged, lean manufac- turing started to encompass continuous improvement and process re-engineering to increase the quality of the products being developed. A “zero defect policy” was implemented, along with constant training and promotion of employees within the lean organization (Scholtes and Hacquebord 1988, Womack et al. 1990). The focus shifted to ensuring quality was built into the product at an early stage, “upstream” as opposed to “downstream” (Swiss 1992). There was also a move from quality as determined by the organization to quality as perceived by the customer (Hines et al. 2004). Adjusting the definition of leanness in order to

Table 2 Evolution of Lean Thinking

Mid-1990s to 2000 quality, Phases 1980s to 1990s Awareness 1990–Mid-1990s quality cost, and delivery 2000 + Value systems Literature theme Dissemination of

shop-floor practices Best practice movement,

benchmarking leading to emulation

Value stream thinking, lean enterprise, collaboration in the supply chain

Capability at system level

Focus JIT practices, cost Cost, training, and promotion; TQM, process reengineering

Cost, process-based to support flow Value and cost, tactical to strategic, integrated to supply chain

Key business processes

Manufacturing, shop-floor only

Manufacturing and materials management

Order fulfilment Integrated processes, order fulfilment, and new business development

Industry sector Automotive—vehicle assembly

Automotive—vehicle and component assembly

General manufacturing-often focused on repetitive manufacturing

High and low volume manufacturing, extension into service sectors

Key authors (Shingo 1981) (Schonberger 1982) (Monden 1983) (Ohno 1988) (Mather 1988)

(Womack et al. 1990) (Hammer 1990) (Harrison 1992)

(Womack and Jones 1994) (Womack and Jones 1996) (Lamming 1993)

(Hines and Taylor 2000) (Hines et al. 2002) (Holweg and Pil 2001)

Adapted from Hines et al. (2004).

reflect the influence of quality, it now reads as “contri- bution to economy and quality.” Simplicity forms a key tenet of lean thinking. In fact

much of the Toyota Production Systems is based on simple practices and approaches. In terms of work- flows, cells of employees work on a product to get it through in one sweeping flow (Monden 1983). Measurement of quality is straightforward, as illus- trated by Shingo’s “poka-yoke” system (Shingo 1986), which uses simple statistics and mistake-proofing whereby one employee checks the work of the previ- ous employee. Scheduling is based on simple mech- anisms such as Kanban, where a card is placed in a pile of stock to trigger a reorder when a minimum level is reached (Monden 1983). Lean organizations are usually simple structures, as exemplified by the Toyota’s “focused factories” (Monden 1983, Shingo 1981). Given the fact that simplicity is a key tenet of leanness, the definition is once again adjusted and now reads as “contribution to economy, quality, and simplicity.” As illustrated in Table 2, while cost reduction

and quality were and still are fundamental concepts within lean thinking, there was a shift in focus from cost reduction, and a “shop-floor-focus” (Womack and Jones 1996) on waste, to an approach that strove to increase overall value. Until the mid-1990s, value was viewed as being equal to cost reduction, which Hines et al. (2004) describes as a “common yet crit- ical” misconception. Instead, this new perspective

on lean thinking aimed to increase customer value through additional products and services, as well as achieving value through the traditional route of internal cost reduction (Womack and Jones 1996, Hines et al. 2004). “Value” is therefore inserted into the definition and is awarded a prominent position, illustrating the fact that value supersedes all other components “contribution to value through economy, quality, and simplicity.” Value is a subjective concept, however. In the context

of leanness, existing conceptual research suggests that value should be measured from the customer’s view- point, as opposed to that of the producing organization (Holweg and Pil 2001, Hines et al. 2004, Lamming 1993). As Hines et al. (2004) state, “regardless of whether an activity appeared to be wasteful from a shop-floor point of view or be costly, it is the customer that ultimately decides what constitutes muda (the elimi- nation of waste), and what does not.” Therefore, the definition is modified to include “perceived customer value” reflecting this narrower interpretation.

Final Definition of Leanness. The final definition of leanness is the “contribution to perceived customer value through economy, quality, and simplicity.”

Leanness vs. Agility Traditionally, any conceptual comparison of agile and lean would ultimately focus on the absence of value in lean thinking and the predominance of the same in the theory of agile. One of Katayama and Bennett’s

(1999) four central tenets of agility is “delivering value to the customer.” Burgess’s (2002) model of agility discusses the customer-focused “core value stream” and its underlying “project value stream,” the latter being directly relevant to this study of ISD projects. Industry specific studies of agility also con- tain many references to value. For example, prod- ucts based on agile dynamics have moved from being standardized to being “customer-perceived value- driven” (Sharp et al. 1999). However, as can be seen from Table 2 earlier, value has recently emerged as a key component of leanness, and so has closed the gap to some extent between lean and agile. While this might no longer represent a divide between agile and lean thinking, there are still a number of other distinc- tions cited in the literature, and these are discussed in turn. The first and most distinguishing factor between

lean and agile is that the former cannot cope well with variability, a fundamental requirement of the latter. Maskell (1996) stated that, “to be lean is to be good at things you can control while to be agile is to be good at things you cannot.” Christopher and Tow- ill (2000) describe leanness as “containing little fat,” whereas agile requires little or no fat in order to be “nimble.” This is illustrated by Stratton and Warbur- ton (2003) in Figure 2, which shows that an agile entity faces volatile demand compared to the more stable demand for standard products that a lean entity faces. No change to the definition of agility is required, how- ever, as the ability to deal with fluctuations and vari- ations was introduced via the definition of flexibility. Another distinction between a lean and an agile

organization is that a lean one is cost efficient and pro- ductive, while an agile one learns fast, if not initially cost efficient and productive (Booth and Hammer 1995). The fact that leanness does not encourage learning would be hotly contested by some, but in

Figure 2 Demand/Product Matrix for Agile and Lean Supply

Demand

Product Volatile Stable

Special Agile X Standard X Lean

Note. Adapted from Stratton and Warburton.

any case, no adjustment is made to the existing def- inition of agility because learning was already incor- porated at an earlier stage in the research process. The previous section discussed the elimination of

waste and reduction of cost to be key elements of the lean enterprise. Within the realm of agility, many have done likewise, and there are many examples of cost- and time-reduction initiatives throughout the agility literature. These include the reduction of production costs (He and Kusiak 1994), handling costs (Lee 1998), relocating costs (Lee 1998), machine reconfiguration costs (Lee 1997), lost revenue due to machine relo- cation (Lee 1998), part differentiation costs (He and Kusiak 1994), and order completion time (Kusiak and He 1997). However, while ultimate leanness eliminates all waste, agility requires waste to be eliminated but only to the extent where its ability to respond to change is not hindered (Young et al. 2001). This does not remove the need to be economical but only lowers its priority. This is supported by numerous pieces of research. For example, Agarwal et al. (2006) conducted a review of the literature that studies the distinction between lean and agile supply chains (e.g., Naylor et al. 1999, Bruce et al. 2004, Mason-Jones et al. 2000, Olhager 2003), and concluded that while a lean supply chain aims to be a market winner in terms of cost, its agile counterpart aims only to be a market qualifier. Instead, an agile supply chain aims to be the market winner in terms of level of service provided, level of robustness, variety of product offered, and other fac- tors that add customer value. This emphasis on value over simple cost reduction was already discussed ear- lier, adding further merit to a reference to value in the definition of agility. Therefore, “the maximization of value through economy, quality, and simplicity” is taken from the earlier definition of leanness and inserted into a modified definition of agility below.

Final Definition of Agility the continual readiness of an ISD method to rapidly or inherently create change, proactively or reac- tively embrace change, and learn from change while contributing to perceived customer value (economy, quality, and simplicity), through its collective compo- nents and relationships with its environment.

Taxonomy of ISD Agility The definition proposed above provides an over- arching interpretation of ISD agility. This is now trans- lated into a formative taxonomy of ISD agility below, comprising the key components of the definition, out- lining the goals an ISD method or part of an ISD method must achieve if it purports to be agile. In this form, it is easier to apply in practice and to make a distinction between something that contributes to ISD agility and something that does not. The first part of the taxonomy shows there are four

primary ways in which a practice or other method component might contribute to agility. To elaborate, a method might facilitate the creation of change. If the change is created elsewhere, however, then the method might contain proactive practices that pre- empt that change, or some reactive measures can take place after the change event has occurred. Finally, the method might facilitate learning from change in order to improve creation, proaction, and reaction in the future. In other words, if early cycles of a project show that creativity on the part of the developers is low, risks have not been addressed, or reaction to changes have been poor or slow, then the method should con- tain practices or techniques to learn from these prob- lems to facilitate improvement over time. Note that change is central to all four aspects of

agility, which is not surprising given that change was a core aspect of the definition and every iteration of its development. There is nothing revelatory in this as the motivation underpinning the very emergence

Figure 3 Taxonomy of ISD Agility

1. To be agile, an ISD method component∗ must contribute to one or more of the following:

(i) creation of change (ii) proaction in advance of change (iii) reaction to change (iv) learning from change

2. To be agile, an ISD method component must contribute to one or more of the following, and must not detract from any:

(i) perceived economy (ii) perceived quality (iii) perceived simplicity

3. To be agile, an ISD method component must be continually ready i.e. minimal time and cost to prepare the component for use.

∗An ISD method component refers to any distinct part of an ISD method.

of agile methods has always been their ability to han- dle change (Fowler and Highsmith 2001, Cockburn 2001a, Fowler 2000, Abrahamsson et al. 2002, Beck 1999). The significance of this taxonomy, however, is that change can be interpreted in a wider sense; agile methods such as XP and Scrum are often cited for their ability to handle requirement changes, and not necessarily all the other changes that an ISD team might have to face. By adopting this broader interpre- tation, a method part can be considered agile even if it contributes to change other than requirements, e.g., changes in personnel, budget, contract, or third-party hardware or software. The second part of the taxonomy requires that

a method component also contributes to perceived economy, quality, or simplicity but should not per- form poorly in any of the three. For example, a 400-page requirements document detailing all speci- fications in minute detail might increase quality and simplicity but might not be agile if perceived econ- omy is poor due to rapid obsolescence of the docu- ment or through its lack of use. Finally, continual readiness of the method compo-

nent is also a prerequisite. For example, acceptance tests certainly contribute to agility in some circum- stances; but if it takes hours to prepare tests every time they run, then their contribution to agility is unclear; the fact that automated acceptance run instantaneously at any given moment explains why they make a valid contribution to agility.

Application of the Taxonomy While the definition and taxonomy of ISD agility was developed in a structured, rigorous, and transparent manner, this was a purely conceptual exercise, and their application in practice is necessary if they are to make a contribution to the field. Researchers often cre- ate “trivial,” impractical concepts, taxonomies, frame- works, and theories because the construction process is restricted by methodological structures that favour validation over usefulness (Lindbolm 1987), a prob- lem that seems to be particularly prevalent in the IS field (Benbasat and Zmud 1999, Keen 1991, Lee 1999). The use of what Weick (1989) refers to as “disciplined imagination” can be used to develop a research arti- fact or test the application of an artifact and the extent to which it achieves its purported goals. Specifically,

“thought trials” (Weick 1989, Van de Ven 2007) can achieve this by applying “mental tests” in an applied situation to solve a particular problem. To test the applicability of the taxonomy devel-

oped in this study, and whether it achieves its pur- ported goals, thought trials were conducted using two completed ISD projects (TaxSys and AccountSys) in ABC Consulting,5 a large multinational consulting firm. These projects were chosen as “rich” and “rev- elatory” cases (Yin 2003). According to both teams, each had very contrasting experiences of their agile implementation, so this was seen as an ideal opportu- nity to apply the taxonomy and test its applicability. Eight members from each project took part and were divided into two distinct focus groups,6 with the objective of analysing each practice adopted on their respective project under each heading of the agility taxonomy. The groups were required to rate each practice’s contribution as excellent, good, mediocre, no perceived effect, poor, or very poor and to provide ratio- nale or illustrative examples to support each rating. The summarized results of the exercise are con- tained in Appendix A, and a detailed rationale for two of the most notable practices, the on-site cus- tomer and stand-up meeting practices, are contained in Appendixes B and C, respectively. The researcher adhered to various “best practices” and principles for ensuring effective focus groups (Morgan 1997, Greenbaum 1998). As can be seen from Table 3, both projects were similar in terms of important character- istics such as team size, duration, and project type. Prior to the commencement of the projects, both teams had received the same in-house agile method training covering the basics of XP and Scrum.

Applying the Taxonomy to Test Agility of Commercially Labeled “Agile” Practices As outlined in the problem statement, a key prob- lem of current agile method thinking is that some practices are now commonly referred to as agile even though the connection to the concept of agility might be tenuous, and even if this link is clear it might still be too simplistic to regard the practice as agile in

5 “ABC Consulting,” TaxSys, and AccountSys are organization and project pseudonyms used to protect organization anonymity. 6 Data collected between October 2008 and December 2008.

Table 3 Profile of Project TaxSys and AccountSys

Project TaxSys Project AccountSys

Project duration 19 months 21 months Team size 20 21 Team composition 1 managing partner 1 managing partner

2 senior managers 2 senior managers 2 team leads 4 team leads 11 developers 19 developers/testers 4 testers

Location Distributed Distributed Development method Scrum/XP Scrum/XP Type of system Web-based commercial Web-based commercial

developed product product Customer type External External Focus group 2 senior managers 2 senior managers participants 1 team lead 2 team leads

5 developers 4 developers

every circumstance. The taxonomy proposed in this study is generic, and its application should be able to differentiate between a practice that truly contributes to agility and one that does not. The summarized results of the trial (Appendix A)

show that, from the focus group members’ experiences on their respective projects, many practices tradition- ally accepted as “agile” did not contribute to agility as defined by the taxonomy in this study (i.e., contribut- ing to at least one of creation, proaction, reaction, and learning, and at least one of perceived economy, quality, and simplicity). In the TaxSys case, the on-site customer, stand-up meetings, pair programming, and collective code ownership were not perecived by the that focus group as having contributed to agility, while the contribution of automated acceptance testing and collective code own- ership were absent in the case of AccountSys. Among the most surprising are the on-site customer and stand- up meetings on the TaxSys project, both of which are among the most commonly known practices of XP and are usually treated as being synonymous with agility. From the detailed rationale for both practices, it is clear why the TaxSys team assigned such poor ratings. In the case of the on-site customer (Appendix B), the individual performing this role was highly expensive, and the return on investment was poor; he did not partake in many of the key activities (e.g., 27 of 113 stand-up meetings and 6 of 14 retrospectives), was too “passive,” worked different shifts to the team so was often present for as little as two hours of the teams’

working day, and was also very slow to give impor- tant feedback. The stand-up meetings (Appendix C) were similarly problematic, with delayed starts, over- runs, dominance of certain team members, and a lack of access to supporting documents. By surfacing these issues, the thought trial showed that the taxonomy can be used to highlight a lack of agility in practices com- monly labeled as agile.

Applying the Taxonomy to Show a Practice Is Not Agile in Every Instance Another goal of a generic taxonomy is to evaluate the agility of a practice on a case-by-case basis, sensitive to project context; what might be agile in one instance might not be agile in another. The summarized results of the thought trial show quite a few cases where the taxonomy achieved this, showing a practice to be agile on one of the projects but not the other. These were the on-site customer, stand-up meetings, pair pro- gramming, and automated acceptance testing. Again, the on-site customer and stand-up meetings provide illus- trative examples. While they provided no agility on the TaxSys project, as discussed above, the AccountSys team’s experience of these practices was very differ- ent. On the AccountSys project, different people sub- stituted in and out of this role to ensure a broad level of knowledge and expertise, and they were highly involved in all activities, attending 43 of 45 stand- up meetings for example. Among many other initia- tives, they organized highly creative brainstrorming sessions with multiple stakeholders. Finally, in com- parison to the customer representative on the TaxSys project, the role was highly economical. Likewise, stand-up meetings were shown to be very beneficial for AccountSys, positively contributing to agility in many ways but very substantially in terms of learn- ing, economy, and quality. Such clear and convincing rationale under these ratings shows that neither prac- tice should automatically be deemed agile in every instance.

Applying the Taxonomy to Identify New Agile Practices The taxonomy proposed in this study should be able to identify practices that contribute to agility even though they might not be commercially labeled or publicly accepted as agile. As can be seen from the

summarized results (Appendix A), the exercise identi- fied three practices not traditionally regarded as agile did in fact contribute to agility as defined by the tax- onomy. In the TaxSys case, a staged commitment strategy was found to significantly contribute to agility, while in the case of AccountSys, the use of a multisite progress dashboard and cultural ambassadors contributed. As can be seen from the rationale underpinning these ratings (presented in detail in Appendix D) each received very high ratings in almost all components of ISD agility.

Conclusions and Future Research Identification of Conceptual Issues Regarding Agility in ISD At the outset, this paper identified a number of significant conceptual shortcomings regarding agile methods and the associated ISD literature in its cur- rent state, namely a lack of clarity, theoretical glue, parsimony, limited applicability, and naivety regard- ing the evolution of the concept of agility in fields outside ISD. As well as having an obvious impact on researchers, this paper highlighted significant implica- tions these shortcomings might have for practitioners; at present it is difficult to improve agile methods, com- pare agile methods, assess agility of traditional, or in-house methods, assess agility of methods unsuit- able for commercial agile approaches, or to facilitate a staged conversion to agility. As far as this researcher is aware, this is the first comprehensive study to explicitly identify these problems and to discuss in detail the significant implications these have for agile method practice.

Definition and Taxonomy of ISD Agility A key contribution of this research is the develop- ment of a definition of ISD agility and a forma- tive taxonomy. The taxonomy was also applied to two ISD projects in a large multinational consult- ing organization. Using thought trials, the taxonomy showed that agility of a method and its underlying components can be tested in a generic sense; prac- tices publicly accepted to be agile might not actually contribute to agility in every instance, and conversely, practices not traditionally recognized for their agility might make a significant contribution. This notion of agile capabilities, also referred to by Vidgen and Wang

(2009) in this journal issue, represents a distinct depar- ture from current thinking on ISD agility, where it is often crudely measured by the number of XP or Scrum practices used, regardless of their suitability to the project context or how they are implemented. In addi- tion, the definition and taxonomy both have impli- cations on the conceptual shortcomings of the agile method literature discussed earlier. Lack of Clarity. The term “agile” in ISD was de-

scribed earlier as so ambiguous, so multi-faceted, and defined in so many different ways that the term has lost much of its meaning. Examples of inconsistent and even contradictory interpretations were cited. To say that this study’s definition and taxonomy of agility is superior or will become universally accepted by researchers is a lofty goal. What is significant is that they were constructed in a stepped and transparent manner, whereby 16 iterations of the definition were presented, and each modification was justified by a thorough analysis of the literature pertaining to that change. Rather than presenting yet another defini- tion of agility to be added to those in existence, this stepped and transparent approach allows researchers to analyse the justification for each part of the def- inition and subsequent taxonomy. This is a feature absent from the existing range of agile method studies to date, and as far as this researcher is aware, is also absent from studies of agility across any of the other disciplines reviewed as part of this study. Lack of Theoretical Glue. As stated earlier, it is

somewhat difficult to establish the theoretical ratio- nale for labeling some practices as “agile” despite the fact that they are commonly referred to as this in prac- tice, e.g., metaphor or pair programming. In contrast, each subcomponent of the definition and taxonomy proposed in this study is justified by an extensive dis- cussion of why it can be classified as “agile,” draw- ing on arguments from previous conceptual studies of agility outside the ISD domain (e.g., Goldman et al. 1991, 1995; Preiss 1997; Dove 1994, 1995; Kidd 1994; Zhang and Sharafi 2000). Lack of Cumulative Tradition. It was argued ear-

lier that the current body of knowledge on ISD agility fails to link sufficiently to existing research both within the ISD literature and further afield. While the roots of agility are sporadically referenced in agile method lit- erature, this study documents the evolution of agility,

and also the evolution of the concepts underpinning agility, namely leanness and flexibility, which have much earlier origins. While some research has looked at each of these subconcepts in isolation, they have not been examined collectively within the broader sys- temic context. The conceptualization developed in this study provides a macro-level view of ISD agility and its subconcepts, and while frameworks of agility exist in other disciplines (e.g., Goldman et al. 1991, 1995; Preiss 1997; Dove 1994, 1995; Kidd 1994; Zhang and Sharafi 2000), this is the first of its kind in ISD. Lack of Parsimony. The earlier discussion revealed

a lot of redundancy and duplication acoss agile ISD methods and related theory. During the construction of the definition in this study, specific attention was given to removing and revising its parts, and not just continually extending, thus contributing to its parsimony. Furthermore, as discussed in the research approach, the researcher strove to maintain a con- sistent level of abstraction, something that is lacking across and even within some existing agile methods. Limited Applicability. It was argued earlier that

commercial agile methods are largely restricted to small, co-located development teams, working on noncritical systems with on demand access to devel- opers, among many other such constraints. In contrast, the definition and taxonomy of ISD agility proposed in this study can be applied in any ISD context. Every method can encourage creativity, proactiveness, learn- ing, and all other parts of the definition, although these could be achieved through very different means, depending on the project environment within which the method in question will operate.

Directions for Future Research Given the complex nature of agility and the weak theo- retical and conceptual grounding in much of the exist- ing agile method literature, this research has made a necessary first step, providing an overarching defini- tion and formative taxonomy of agility upon which much more can be done. It is hoped that this study will inspire others to investigate further this impor- tant area, and there are many fruitful directions future researchers could take. Firstly, researchers could develop behavioral sub-

constructs under the various components of the taxon- omy. If we take “learning from change” as an example,

A p p en

d ix

A . S u m m ar iz ed

F o cu

s G ro u p R es u lt s

P er ce iv ed

cu st o m er

v al u e

P ro je ct

C o n ti n u al

C o n tr ib u to r

n am

e P ra ct ic e

C re at io n

P ro ac ti o n

R ea ct io n

L ea rn

in g

E co n o m y

Q u al it y

Si m p li ci ty

re ad

in es s

to ag

il it y

T ax

Sy s

Sp ri n ts

E xc el le n t

G o o d

E xc el le n t

G o o d

Y es

P ri o ri ti za ti o n

N o p er ce iv ed

ef fe ct

N o p er ce iv ed

ef fe ct

E xc el le n t

N o p er ce iv ed

ef fe ct

G o o d

N o p er ce iv ed

ef fe ct

G o o d

Y es

O n -s it e cu

st o m er

P o o r

V er y p o o r

P o o r

M ed

io cr e

Y es

N o

St an

d -u p

m ee ti n g

P o o r

N o p er ce iv ed

ef fe ct

P o o r

V er y p o o r

N o p er ce iv ed

ef fe ct

N o p er ce iv ed

ef fe ct

Y es

N o

P ai r p ro g ra m m in g

M ed

io cr e

N o p er ce iv ed

ef fe ct

N o p er ce iv ed

ef fe ct

G o o d

P o o r

N o p er ce iv ed

ef fe ct

N o p er ce iv ed

ef fe ct

Y es

N o

A u to m at ed

ac ce p -

ta n ce

te st in g

N o p er ce iv ed

ef fe ct

E xc el le n t

N o p er ce iv ed

ef fe ct

E xc el le n t

Y es

C o ll ec ti v e co d e o w n -

er sh

ip M ed

io cr e

N o p er ce iv ed

ef fe ct

N o p er ce iv ed

ef fe ct

N o p er ce iv ed

ef fe ct

N o p er ce iv ed

ef fe ct

N o p er ce iv ed

ef fe ct

P o o r

Y es

N o

R et ro sp

ec ti v es

N o p er ce iv ed

ef fe ct

N o p er ce iv ed

ef fe ct

N o p er ce iv ed

ef fe ct

E xc el le n t

G o o d

E xc el le n t

Y es

St ag

ed co m m it m en

t st ra te g y

N o p er ce iv ed

ef fe ct

E xc el le n t

N o p er ce iv ed

ef fe ct

E xc el le n t

N o p er ce iv ed

ef fe ct

E xc el le n t

Y es

A cc o u n tS y s

Sp ri n ts

N o p er ce iv ed

ef fe ct

E xc el le n t

G o o d

E xc el le n t

G o o d

Y es

P ri o ri ti za ti o n

N o p er ce iv ed

ef fe ct

N o p er ce iv ed

ef fe ct

E xc el le n t

N o p er ce iv ed

ef fe ct

E xc el le n t

N o p er ce iv ed

ef fe ct

E xc el le n t

Y es

O n -s it e cu

st o m er

E xc el le n t

G o o d

E xc el le n t

N o p er ce iv ed

ef fe ct

Y es

St an

d -u p

m ee ti n g

G o o d

E xc el le n t

G o o d

Y es

P ai r p ro g ra m m in g

E xc el le n t

N o p er ce iv ed

ef fe ct

N o p er ce iv ed

ef fe ct

E xc el le n t

G o o d

E xc el le n t

N o p er ce iv ed

ef fe ct

Y es

A u to m at ed

ac ce p -

ta n ce

te st in g

N o p er ce iv ed

ef fe ct

E xc el le n t

N o p er ce iv ed

ef fe ct

P o o r

E xc el le n t

P o o r

N o

C o ll ec ti v e co d e o w n -

er sh

ip N o p er ce iv ed

ef fe ct

N o p er ce iv ed

ef fe ct

N o p er ce iv ed

ef fe ct

N o p er ce iv ed

ef fe ct

N o p er ce iv ed

ef fe ct

N o p er ce iv ed

ef fe ct

P o o r

Y es

N o

R et ro sp

ec ti v es

G o o d

E xc el le n t

G o o d

E xc el le n t

G o o d

Y es

M u lt is it e

p ro g re ss

d as h b o ar d

N o p er ce iv ed

ef fe ct

E xc el le n t

Y es

C u lt u ra l

am b as -

sa d o rs

an d

cr o ss -

p o ll in at io n

E xc el le n t

G o o d

Y es

A p p en

d ix

B . T h e O n -S it e C u st o m er

(O S C ) P ra ct ic e

P er ce iv ed

cu st o m er

v al u e

P ro je ct

C o n ti n u al

C o n tr ib u to r

n am

e C re at io n

P ro ac ti o n

R ea ct io n

L ea rn

in g

E co n o m y

Q u al it y

Si m p li ci ty

re ad

in es s

to ag

il it y

T ax

Sy s

Po or

Ve ry

po or

Ve ry

po or

Ve ry

po or

Po or

M ed

io cr e

Ye s

N o

L im

it ed

O SC

k n o w le d g e

o f

w id er

b u si n es s

n ee d s st ifl ed

cr ea ti o n o f n ew

id ea s.

O SC

w as

“h ig h ly

p as si v e. ”

N o

O SC

in v o lv e-

m en

t in

“s p ik es ”

(d ev

el o p er

ex -

p er im

en ta ti o n

se ss io n s) .

O SC

w as

“h ig h ly

p as si v e, ”

an d

“s h o w ed

li tt le

si g n

o f p ro ac -

ti o n ”

(P ro je ct

M an

ag er ).

M an

y ch

an g es

in cl ie n t n ee d s,

p ro ce ss es

an d

st ru ct u re

n o t

re la y ed

b y O SC

to IS D

te am

u n ti l d is co v er ed

b y th em

A v er ag

e o f 4. 3

b u si n es s d ay

s fo r

O SC

to g iv e fe ed

b ac k

o n

n ew

u se r

st o ri es .

L im

it ed

O SC

k n o w l-

ed g e

o f w id er

b u si n es s

n ee d s

an d is su

es st ifl ed

te am

le ar n in g .

O SC

o n ly

at te n d ed

27 o f 11 3 st an

d -u p

m ee ti n g s, an

d 6 o f

14 re tr o sp

ec ti v es .

O SC

d id

n o t le ar n

fr o m

m an

y m is -

ta k es ,

re p ea t-

in g

ce rt ai n

m is -

ta k es

co n ti n u -

o u sl y th ro u g h o u t

th e

p ro je ct

e. g .

u si n g w ro n g v er -

si o n o f o rg an

iz a-

ti o n p ro ce ss

d o c-

u m en

ts .

O SC

p ai d

m o re

th an

tw ic e

h ig h es t p ai d

d ev

el o p er .

T h e

O SC

, 1

p er so n

o f

18 ac co u n te d

fo r a la rg e p ar t

(1 4%

) o f p ro je ct

b u d g et .

In co m p at ib il it y

b et w ee n

O SC

an d

d ev

el o p -

er s’

w o rk in g

h o u rs

in cr ea se d

re al

co st

p er

h o u r ra te . O n ly

2 h o u rs

o v er -

la p

o n

m an

y o cc as io n s.

O n ly

O SC

ro le

w as

u se r st o ry

d ev

el -

o p m en

t an

d v al i-

d at io n .

O SC

d id

h el p

to si m p li fy

u se r

st o ri es

an d

re m o v e d u p li ca -

ti o n s an

d in co n -

si st en

ci es .

H o w ev

er , O SC

d em

an d ed

u se r

st o ri es

to b e

d o cu

m en

te d

in fo rm

at s d if fe r-

en t fr o m

d ev

el -

o p er

o r cl ie n t

o rg an

iz at io n ,

ad d in g

si g n ifi -

ca n t co m p le xi ty

an d

ef fo rt .

A cc o u n tS y s

Ex ce lle

nt G oo d

Ex ce lle

nt Ex ce lle

nt N o pe

rc ei ve d ef fe ct

Ye s

M o n th ly

cr ea ti v e

b ra in st o rm

in g

se ss io n s

o rg a-

n iz ed

an d le d b y

O SC

. Se

ss io n s in v o lv ed

m u lt ip le

cu s-

to m er

em p lo y -

ee s,

ac ro ss

sp ec -

tr u m

o f st ak

e- h o ld er

g ro u p s

(R & D , m ar k et -

in g , ac co u n ts ,

m an

u fa ct u ri n g ).

O SC

p re se n te d

to d ev

el o p er

te am

o n m u lt i-

p le

o cc as io n s.

P re se n ta ti o n s w er e

ta il o re d to

ca te r

fo r th e d ev

el -

o p er s’

la ck

o f

b u si n es s

an d

d o m ai n k n o w l-

ed g e.

R ea l- ti m e

in v o lv em

en t in

u se r st o ry

v al -

id at io n , w it h

in st an

t fe ed

- b ac k .

O SC

co n -

ti n u o u sl y

is su

ed “l iv e”

re p ri o ri ti ze d

li st s

w it h in

24 h o u rs ,

al lo w in g h ig h

p ri o ri ty

it em

s to

b e

in tr o -

d u ce d

m id -

it er at io n .

B ro ad

cu st o m er

k n o w le d g e

o f

w id er

b u si en

ss n ee d s an

d is su

es .

A tt en

d ed

43 o f 45

st an

d -u p

m ee t-

in g s.

R el at iv el y

lo w

co st

O SC

. O SC

av ai la b le

fu ll -t im

e. F u ll

o v er la p

b et w ee n

O SC

an d

d ev

el o p -

er s’

w o rk in g

h o u rs .

D if fe re n t cu

st o m er

em p lo y ee s w er e

“r ev

o lv ed

” in

an d

o u t o f th e O SC

ro le

d u ri n g th e p ro je ct ,

m at ch

in g re le v an

t ex p er ie n ce

to th e

p ar t o f th e sy st em

b ei n g d ev

el o p ed

. Q u al it y

o f p ro d u ct

in cr ea se d b y in te r-

m it te n t ac ce ss

to o th er

st ak

eh o ld er s

v ia

O SC

A p p en

d ix

C . S ta n d -U

p M ee ti n g (S U M ) P ra ct ic e

P er ce iv ed

cu st o m er

v al u e

P ro je ct

C o n ti n u al

C o n tr ib u to r

n am

e C re at io n

P ro ac ti o n

R ea ct io n

L ea rn

in g

E co n o m y

Q u al it y

Si m p li ci ty

re ad

in es s

to ag

il it y

T ax

Sy s

Po or

N o pe

rc ei ve d

Po or

Ve ry

po or

N o pe

rc ei ve d

N o pe

rc ei ve d

Ye s

N o

SU M s h ig h ly

fo rm

al , o ft en

cr it ic al , an

d so

m an

y te am

m em

- b er s w er e to o

sh y o r fe ar -

fu l to

su g g es t

n ew

id ea s.

ef fe ct

D es p it e ra is in g is su

es ,

p o te n ti al

p ro b -

le m s,

an d

re q u ir ed

ch an

g es , m an

y te am

m em

b er s fe lt

th at

o n

m an

y o cc as io n s

n o re sp

o n si v e ac ti o n

w as

ta k en

C er ta in

te am

m em

b er

u p d at es

w er e o ft en

v ag

u e an

d o f li m -

it ed

v al u e in

o p er a-

ti o n al ly

re ac ti n g

to ch

an g e.

St an

d ar d fo rm

at ,w

h er e

ea ch

te am

m em

b er

st at ed

w h at

w en

t w el l, w h at

d id

n o t

g o

w el l,

an d

w h at

th ey

w er e g o in g to

d o

n ex t fa ci li ta te d

b as ic

le ar n in g ac ro ss

th e te am

H o w ev

er , m o st

te am

m em

b er s sa id

th ey

p ai d

li tt le

o r

n o

at te n ti o n

an d

ju st

“r an

th ro u g h

th e

m o ti o n s. ”

SU M s o ft en

d o m in at ed

b y

p ro je ct

m an

- ag

er (o ft en

as m u ch

as 75

% o f m ee t-

in g ti m e) , an

d b o re

cl o se r re se m b la n ce

to a “s er m o n ” th an

ex ch

an g e o f k n o w l-

ed g e ac ro ss

te am

SU M s h el d aw

ay fr o m

d ev

el o p m en

t ar ea ,

w it h

n o

ac ce ss

to st o ry b o ar d s o r o th er

ar ti fa ct s to

re fe r to

d u ri n g m ee ti n g .

T im

e w as te d

d u e to

la te

ar ri v al s at

SU M s.

SU M s

to o k

av er ag

e o f ap

p ro x 50

m in s,

ra n g in g fr o m

0. 5 to

1. 5 h o u rs .

M an

y SU

M s d o m in at ed

b y

o v er -e la b o ra te

d is cu

ss io n o f a si n -

g le

is su

SU M s at te n d ed

b y al l

re g ar d le ss

o f p er -

ce iv ed

o f a la ck

o f

re le v an

ce o r u se fu l-

n es s.

ef fe ct

A p p en

d ix

C . (C

o n t’ d .)

P er ce iv ed

cu st o m er

v al u e

P ro je ct

C o n ti n u al

C o n tr ib u to r

n am

e C re at io n

P ro ac ti o n

R ea ct io n

L ea rn

in g

E co n o m y

Q u al it y

Si m p li ci ty

re ad

in es s

to ag

il it y

A cc o u n tS y s

G oo

d G oo

d Ex ce lle

nt Ex ce lle

nt G oo

d Ye

s Ye

s SU

M s

w er e

h ig h ly

in fo r-

m al

w it h

al l

en co u rg ae d to

sp ea k

th ei r

m in d an

d su

g -

g es t

id ea s

re g ar d le ss

o f

h o w

at y p ic al

th ey

m ay

b e.

A t ev

er y

SU M ,

o n e d ev

el o p er

h ad

to in tr o -

d u ce

a “b

re ak

- th ro u g h id ea ,”

an d

re ce iv ed

fe ed

b ac k o n it s

u se fu ln es s an

d v ia b il it y.

SU M s to o k p la ce

at 2 p .m . ea ch

d ay , al lo w in g

a p ar tn er

te am

b as ed

in th e

U .S . to

o b se rv e

p ro ce ed

in g s

an d co n tr ib u te

n ew

id ea s.

A “B

ar ri er s”

se c-

ti o n w as

in tr o -

d u ce d

in th e

SU M

w h er e

ea ch

te am

m em

b er

p ro ac -

ti v el y

li st ed

(i ) p o te n ti al

is su

es th at

m ay

in h ib it

th ei r

w o rk

an d

(i i)

te am

m em

b er s

w h o

h ad

th e

ex p er ti se

to h el p .

A ll

o b st ac le s

o r p ro b le m s

d o cu

m en

te d

b y th e p ro je ct

m an

ag er ,

an d

a li st

o f re sp

o n se

ac ti o n s ci rc u -

la te d

to th e

te am

U p d at es

h ad

to b e sp

ec i-

fi ed

cl ea rl y at

th e ta sk

le v el ,

an d

ac co m -

p an

ie d

b y

th e

p o te n -

ti al

im p ac t

o n

u se r st o -

ri es

an d

in i-

ti al

es ti m at es .

H ig h ly

in te ra ct iv e

se ss io n s fa ci li -

ta te d le ar n in g .

T ea m

m em

- b er s

o ft en

d is tr ib u te d

ex ce rp ts

o f

co d e,

d ia g ra m s,

o r o th er

ar ti -

fa ct s to

su p p o rt

th ei r p re se n ta -

ti o n .

SU M s

h el d

in d ev

el o p m en

t ar ea

to al lo w

sp ea k er s

to re fe r to

ar ti -

fa ct s ar o u n d th e

ro o m .

SU M s

al w ay

s st ar te d

o n

ti m e

re g ar d le ss

o f

w h o w as

o n ti m e

o r la te .

SU M s li m it ed

to 15

m in u te

m ax

- im

u m .

“F ac il it at o r”

ap p o in te d

to k ee p ev

er y te am

m em

b er

to ti m e.

U n re so lv ed

is su

es ta k en

“o ffl in e”

af te rw

ar d s an

d o n ly

d is cu

ss ed

b e re le v an

t te am

m em

b er s.

P er si st en

t p ro b -

le m s

an d

d ef ec ts

p re -

se n te d

at SU

M fo r

g ro u p to

ta k e

aw ay

an d

so lv e.

SU M

en su

re d

an y

ex p er t

w as

aw ar e

o f a p ro b le m

th ey

co u ld

as si st

w it h .

E ac h te am

m em

- b er

le ft

th e

SU M

w it h

a cl ea r id ea

o f

(i ) th ei r ta sk s

fo r th e

d ay ,

(i i)

w h o th ey

w o u ld

w o rk

w it h th at

d ay ,

an d (i ii ) w h ic h

te am

m em

b er s

th ey

h ad

to h el p o r ad

v is e.

A p p en

d ix

D . Id en

ti fi ca ti o n o f N ew

A g il e P ra ct ic es

P er ce iv ed

cu st o m er

v al u e

P ra ct ic e

D es cr ip ti o n

C o n ti n u al

C o n tr ib u to r

(p ro je ct )

o f p ra ct ic e

C re at io n

P ro ac ti o n

R ea ct io n

L ea rn

in g

E co n o m y

Q u al it y

Si m p li ci ty

re ad

in es s

to ag

il it y

St ag

ed co m m it m en

t st ra te g y

(T ax

Sy s)

B ec au

se o f a d el ay

in co n fi rm

at io n

o f p ro je ct

b u d -

g et , th e

te am

cr ea te d

th re e

se ts

o f re q u ir e-

m en

ts , b as ed

o n

th re e

p o ss ib le

b u d g et

to ta ls

(£ 1. 3 m , £3

m ,

£4 .7

m ).

T h es e

w er e

fa ct o re d

in to

ev er y p ri -

o ri ti za ti o n , re t-

ro sp

ec ti v e,

an d

it er at io n .

N o pe rc ei ve d

ef fe ct

Ex ce lle

nt St ag

in g

p ro ac -

ti v el y p ro te ct ed

ag ai n st

an y

d ec is io n

to re d u ce

p ro je ct

fu n d in g .

Ex ce lle

nt N o

ef fo rt

re -

q u ir ed

to d o w n si ze

sy st em

sc o p e

o r

q u al it y

in th e ev

en t

o f fu n d in g

re d u ct io n .

N o pe rc ei ve d

ef fe ct

Ex ce lle

nt E n su

re d

n o

w as te d

ef fo rt

o n p ar ts

o f th e

sy st em

th at

m ay

b e

d is -

co n ti n u ed

N o pe rc ei ve d

ef fe ct

Ex ce lle

nt Ye

s Ye

M u lt is it e

p ro g re ss

d as h b o ar d

(A cc o u n tS y s)

O n li n e d as h b o ar d

w h ic h

re p re -

se n ts

p ro g re ss

o f

m u lt ip le

te am

s d is -

tr ib u te d

ac ro ss

m an

y lo ca -

ti o n s.

R ea l- ti m e

re co rd in g

o f

ea ch

u se r st o ry ,

th e d ev

el o p er (s )

as si g n ed

to th at

st o ry , th ei r lo ca -

ti o n (s ), th e st a-

tu s o f th e st o ry ,

th e n o . o f u n it

te st s u n d er

ea ch

st o ry , an

d th e

n o . o f th o se

su c-

ce ss fu ll y te st ed

o r n o t.

N o pe rc ei ve d

ef fe ct

Ex ce lle

nt T ra n sp

ar en

cy re m o v ed

m u ch

p o te n -

ti al

u n ce rt ai n ty

o ft en

ca u se d

b y d is tr ib u te d

d ev

el o p m en

t. A ll o w ed

ea rl y

id en

ti fi ca ti o n o f

p o te n ti al

is su

es an

d co n fl ic ts

b et w ee n te am

Ex ce lle

nt A ll

d ev

el o p -

er s

in al l

lo ca ti o n s

in st an

tl y

in fo rm

ed o f

ch an

g es

in u se r

st o ry

st at u s, u n d er -

ly in g

u n it

te st s,

et c.

E n su

re d

co m p le te

re sp

o n se

to ch

an g e,

as n o ta sk

“f el l

b et w ee n

th e

te am

s” an

d re m ai n ed

ac ci d en

tl y

u n as si g n ed

Ex ce lle

nt H ig h tr an

sp ar en

cy an

d v is ib il it y

ta u g h t

ea ch

te am

ab o u t th e

o th er ’s

ab il i-

ti es , p ro g re ss ,

p re fe re n ce s, an

d cu

lt u ra l tr ai ts .

A te am

m em

- b er

co u ld

d ri ll -

d o w n

to v ie w

an d le ar n ab

o u t

te ch

n ic al

sp ec i-

fi ca ti o n s,

ac ti o n

it em

s, co d e,

an d

te st s.

Ex ce lle

nt N o fi n an

ci al

co st

o r ad

d it io n al

d o cu

m en

ta ti o n

o r p ro ce ss es

re q u ir ed

. Se

am le ss , re al -

ti m e u p d at es

fr o m

lo ca l

m o n it o ri n g

sy st em

Ex ce lle

nt L ar g e re d u ct io n

in d ef ec ts .

N o

p o ss ib il it y

fo r

re q u ir e-

m en

ts to

“f al l

b et w ee n ” th e

te am

s. T ra n sp

ar en

cy o f

p ro g re ss

in tr o -

d u ce d a “c o m -

p et it iv e

b u t

h el at h y

el e-

m en

t” b et w ee n

te am

Ex ce lle

nt D as h b o ar d

m ad

e d iv is io n

o f

w o rk

b et w ee n

d is tr ib u te d

te am

s v er y si m -

p le .

D as h b o ar d

p ro -

v id ed

a si n g le

p o in t o f re fe r-

en ce

si m p li fy -

in g

co m m u n i-

ca ti o n b et w ee n

te am

Ye s

A p p en

d ix

D . (C

o n t’ d .)

P er ce iv ed

cu st o m er

v al u e

P ra ct ic e

D es cr ip ti o n

C o n ti n u al

C o n tr ib u to r

(p ro je ct )

o f p ra ct ic e

C re at io n

P ro ac ti o n

R ea ct io n

L ea rn

in g

E co n o m y

Q u al it y

Si m p li ci ty

re ad

in es s

to ag

il it y

C u lt u ra l am

b as -

sa d o r an

d cr o ss -

p o ll in at io n

(A cc o u n tS y s)

M em

b er

o f o ff -

sh o re

te am

is re lo ca te d

to o n sh

o re

te am

fo r si x

w ee k s d u ra -

ti o n . T h e p er -

so n p er fo rm

- in g th is

ro le

w as

ch an

g ed

ev er y

si x

w ee k s

to al lo w

al l o ff -

sh o re

m em

- b er s to

sp en

d ti m e

w it h

th e o n sh

o re

te am

Ex ce lle

nt A m b as sa d o r

re fe rr ed

to p o te n -

ti al

id ea s

an d

co ll ab

o ra ti v e

ex p er im

en ta ti o n

w it h

o ff sh

o re

te am

Ex ce lle

nt A m b as sa d o r

re m o v ed

m u ch

p o te n -

ti al

u n ce r-

ta in ty

o ft en

ca u se d b y d is -

tr ib u te d

d ev

el o p m en

t. A ll o w ed

ea rl y

id en

ti fi ca ti o n

o f

p o te n -

ti al

is su

es an

d co n -

fl ic ts

b et w ee n

te am

Ex ce lle

nt A m b as sa d o r

as si st ed

w it h

co n -

fl ic t re so lu -

ti o n ca u se d

b y su

d d en

, u n fo re se en

ch an

g es

e. g . p ro b -

le m s

ca u se d

b y

d ec is io n

to sh

if t

an ex tr a

20 %

o f

w o rk

fr o m

o n sh

o re

to o ff sh

o re

te am

Ex ce lle

nt A m b as sa d o r

co n st an

tl y

d is se m in at ed

in fo rm

a- ti o n

to th e

o n sh

o re

te am

m em

b er s

re g ar d in g

th e o ff sh

o re

te am

’s ab

il i-

ti es , p ro g re ss ,

p re fe re n ce s,

an d

cu lt u ra l

tr ai ts .

A m b as sa d o r

h el d

a fo rt -

n ig h tl y se m -

in ar

o n

th e

ab o v e to p ic s.

G oo d

F in an

ci al

co st

o f re lo ca ti n g

am b as sa d o r

“i n si g n ifi ca n t”

re la ti v e

to to ta l p ro je ct

co st .

G oo d

C u st o m er s w er e

v er y p le as ed

to se e an

d b e ab

le to

co m m u n i-

ca te

fa ce -t o -f ac e

w it h

m em

b er s

o f th e o ff sh

o re

te am

th ey

w er e

p ay

in g fo r.

G oo d

A m b as sa d o r si m -

p li fi ed

co m -

m u n ic at io n

b et w ee n te am

s. A m b as sa d o r

ex p la in ed

an y

an o m al ie s

o r

is su

es w it h an

y o f th e ar ti fa ct s

d ev

el o p ed

b y

th e

o ff sh

o re

te am

Ye s

one could take subconstructs from team and organi- zational learning literature (e.g., Huber 1996, Argyris 1999) and adapt them if needed to suit an ISD con- text. A similar exercise for creation, proaction, reac- tion, economy, quality, and simplicity would result in a more detailed, operational classification of ISD agility. The thought trials in this paper applied the taxonomy to test whether it was viable in practice and whether it achieved its purported goals. A more detailed and operational classification with subcon- structs would facilitate a more prescriptive, intricate, and exhaustive exercise. The current taxonomy examines behaviors and per-

ceived outcomes that contribute to ISD agility. Future research could extend this by developing a set of met- rics to evaluate actual performance outcomes under each component of the taxonomy. It is well known that behaviors are not always rewarded by positive out- comes, so applying outcome measures of ISD agility across methods, method variants, organizations, and projects could reveal some very interesting insights and add credence to those who claim their methods or practices are agile. Researchers or practitioners could also provide fur-

ther practicial application of the taxonomy. For exam- ple, the thought trials in this study provided a snaphot of ISD agility on the TaxSys and AccountSys projects at a single point in time. Longitudinal cases could yield further insights, identifying how quickly teams can transition to agility and how effective such a transi- tion actually is. Aside from conceptual weaknesses of agility ISD, the paper also set out practical implica- tions arising from these issues, and these could form the basis for further research. Researchers could use the taxonomy as a starting reference point to analyse method improvement, to compare agile methods, and to assess agility in traditional or in-house methods, or in environments unsuitable for commercial agile methods. These efforts would be even easier once a set of behavioral and/or outcome subcontructs have been developed under each component of the taxonomy.

References Abrahamsson, P., O. Salo, J. Ronkainen, J. Warsta. 2002. Agile soft-

ware development methods: Review and analysis. VTT Pub- lications 478, Technical Research Centre of Finland, Espoo, Finland.

Agarwal, A., R. Shankar, M. Tiwari. 2006. Modeling the metrics of lean, agile and leagile supply chains: An ANP-based approach. Eur. J. Oper. Res. 173 211–225.

Ågerfalk, P., B. Fitzgerald. 2005. Methods as action knowledge: Exploring the concept of method rationale in method construc- tion, tailoring and use. T. Halpin, J. Krogstie, K. Siau, eds. Proc. EMMSAD’05: Tenth IFIP WG8.1 Internat. Workshop on Exploring Modeling Methods in Systems Anal. Design, Porto, Portugal.

Ågerfalk, P., K. Wistrand. 2003. Systems development method rationale: A conceptual framework for analysis. The 5th Inter- nat. Conf. Enterprise Inform. Systems �ICEIS 2003�, Angers, France.

Ambler, S. W. 2002. Agile Modeling: Best Practices for the Unified Pro- cess and Extreme Programming. John Wiley & Sons, New York.

Argyris, C. 1999. On Organizational Learning. Blackwell, London. Auer, K., R. Miller. 2002. Extreme Programming Applied—Playing to

Win. Addison-Wesley, Reading, MA. Aveling, B. 2004. XP lite considered harmful? J. Eckstein,

H. Baumeister, eds. Extreme Programming and Agile Processes in Software Engineering. Springer, Berlin/Heidelberg.

Avison, D., G. Fitzgerald. 2003. Information Systems Development: Methodologies, Techniques and Tools. McGraw-Hill, London.

Aydin, M., F. Harmsen, K. Van Slooten, R. Stegwee. 2004. An agile informations systems development method in use. Turkish J. Electronic Engrg. 12 127–138.

Backhouse, J., J. Liebenau, F. Land. 1991. On the discipline of infor- mation systems. J. Inform. Systems 1 19–27.

Banville, C., M. Landry. 1989. Can the field of MIS be disciplined? Comm. ACM 32 48–60.

Beck, K. 1999. Extreme Programming Explained. Addison-Wesley, Reading, MA.

Benbasat, I., R. Zmud. 1999. Empirical research in information sys- tems: The practice of relevance. MIS Quart. 23 3–16.

Boehm, B., R. Turner. 2003. Using risk to balance agile and plan- driven methods. IEEE Software 36 57–66.

Booth, C., M. Hammer. 1995. Agile manufacturing concepts and opportunities in ceramics. Ceramic Trans. 50 67–76.

Bowers, J., J. May, E. Melander, M. Baarman, A. Ayoob. 2002. Tailoring XP for large mission critical software develop- ment. D. Wells, L. Williams, eds. XP/Agile Universe. Springer, Chicago.

Breu, K., C. Hemingway, M. Strathern. 2001. Workforce agility: The new employee strategy for the knowledge economy. J. Inform. Tech. 17 21–31.

Brinkkemper, J. N. 1990. Formalisation of information systems modelling. Ph.D. thesis, Catholic University of Nijmegen, Nijmegen, The Netherlands.

Brinkkemper, S. 1996. Method engineering: Engineering of infor- mation systems development methods and tools. Inform. Soft- ware Tech. 38 275–280.

Brinkkemper, S., M. Saeki, F. Harmesen. 1999. Meta-modelling based assembly techniques for situational method engineering. Inform. Systems 24 209–228.

Bruce, M., L. Daly, N. Towers. 2004. Lean or agile: A solution for supply chain management in the textiles and clothing industry. Internat. J. Oper. Production Management 24 151–170.

Bryson, B. 2004. A Short History of Nearly Everything. Black Swan Publishing, London.

Burgess, T. 2002. Enhancing value stream agility. Eur. Management J. 20 199–212.

Cao, L., K. Mohan, P. Xu, B. Ramesh. 2004. How extreme does extreme programming have to be? Adapting XP practices to large-scale projects. R. Sprague, ed. Proc. 37th Hawaii Internat. Conf. System Sci., IEEE Computer Society Press, Washington, DC.

Checkland, P. 1981. Systems Thinking, Systems Practice. John Wiley & Sons, Chichester, UK.

Checkland, P., S. Holwell. 1998. Information Systems and Information Systems. John Wiley and Sons, Chichester, UK.

Childerhouse, P., S. Disney, D. Towill. 2000. Speeding up the progress curve towards effective supply chain management. Internat. J. Supply Chain Management 5 176–186.

Christopher, M. 2000. The agile supply chain: Competing in volatile markets. Indust. Marketing Management 29 37–44.

Christopher, M., D. Towill. 2000. Supply chain migration from lean and functional to agile and customised. Internat. J. Supply Chain Management 5 206–213.

Coad, P., J. De Luca, E. Lefebre. 1999. Java Modelling in Color. Prentice Hall, Englewood Cliffs, NJ.

Cockburn, A. 2001a. Agile Software Development. Addison-Wesley, Reading, MA.

Cockburn, A. 2001b. Crystal Clear: A Human-Powered Software Devel- opment Methodology for Small Teams. Addison-Wesley, Reading, MA.

Cockburn, A., L. Williams. 2001. The costs and benefits of pair pro- gramming. G. Succi, M. Marchesi, eds. Extreme Programming Examined. Addison-Wesley, Reading, MA.

Connors, D. T. 1992. Software development methodologies and traditional and modern information systems. Software Engrg. Notes 17 43–49.

Crispin, L., T. House. 2003. Testing Extreme Programming. Pearson, Boston.

Cummings, J., J. Espinosa, C. Pickering. 2009. Crossing spatial and temporal boundaries in globally distributed projects: A rela- tional model of coordination delay. Inform. Systems Res. 20(3) 420–439.

De Groote, X. 1994. The flexibility of production processes: A gen- eral framework. Management Sci. 40 933–945.

De Vor, R., J. Mills. 1995. Agile manufacturing. Amer. Soc. Mech. Engineers 2 977–992.

DiMaggio, P. 1995. Comments on “What Theory is Not.” Admin. Sci. Quart. 40 391–397.

Dove, R. 1994. The meaning of life & the meaning of agile. Pro- duction 106 14–15.

Dove, R. 1995. Measuring agility: The toll of turmoil. Production 107 12–14.

Drucker, P. 1995. The information that executives truly need. Harvard Bus. Rev. 73 55–63.

Dubin, R. 1976. Theory building in applied areas. M. Dunnette, ed. Handbook of Industrial and Organisational Psychology. Rand McNally, Chicago.

Dubin, R. 1978. Theory Development. Free Press, New York. Eppink, D. 1978. Managing the unforeseen: A study of flexibil-

ity. Ph.D. thesis, Free University of Amsterdam, Amsterdam, The Netherlands.

Fenwick, J. 2003. Adapting XP to an academic environment by phasing in practices. F. Maurer, D. Wells, eds. XP/Agile Universe. Springer, Berlin/Heidelberg.

Fitzgerald, B. 1997. The use of systems development methodologies in practice: A field study. Inform. Systems J. 7 201–212.

Fitzgerald, B. 1998. An empirical investigation into the adoption of systems development methodologies. Inform. Management 34 317–328.

Fitzgerald, B., G. Hartnett, K. Conboy. 2006. Customising agile methods to software practices. Eur. J. Inform. Systems 15 197–210.

Fitzgerald, B., N. Russo, E. Stolterman. 2002. Information Systems Development: Methods in Action. McGraw-Hill, London.

Fowler, M. 2000. Put your process on a diet. Software Development 8 32–36.

Fowler, M. 2001. Extreme Programming Explained. Addison-Wesley, Reading, MA.

Fowler, M., J. Highsmith. 2001. The agile manifesto. Software Devel- opment 9 28–32.

Galliers, B. 1994. Relevance and rigour in information sys- tems research: Some personal reflections on issues facing the information systems research community. B. Glasson, I. Hawryszkiewycz, B. Underwood, R. Weber, eds. Proc. IFIP TC8 Open Conf. Bus. Process Re-Engineering, Elsevier, New York.

Gerwin, D. 1993. Manufacturing flexibility: A strategic perspective. Management Sci. 39 395–410.

Golden, W., P. Powell. 2000. Towards a definition of flexibility: In search of the holy grail? Omega 28 373–384.

Goldman, S., R. Nagel. 1993. Management, technology and agility: The emergence of a new era in manufacturing. Internat. J. Tech. Management 8 18–38.

Goldman, S., R. Nagel, K. Preiss. 1995. Agile Competitors and Virtual Organisations: Strategies for Enriching the Customer. Von Nostrand Reinhold, NY.

Goldman, S., R. Nagel, K. Preiss, R. Dove. 1991. Iacocca Institute: 21st Century Manufacturing Enterprise Strategy: An Industry Led View. Iacocca Institute, Bethlehem, PA.

Goudswaard, A., M. de Nanteuil. 2000. Flexibility and working conditions: A qualitative and comparative study in seven EU member states. Technical Report (EF0007), European Founda- tion for Living and Working Conditions, Dublin, Ireland.

Gould, P. 1997. What is agility? Manufacturing Engineer 76 28–31. Greenbaum, T. 1998. The Handbook for Focus Group Research. SAGE,

London. Grunwald, H., L. Fortuin. 1992. Many steps towards zero inventory.

Eur. J. Oper. Res. 59 359–369. Gunasekaran, A., Y. Yusuf. 2002. Agile manufacturing: A taxonomy

of strategic and technological imperatives. Internat. J. Produc- tion Res. 40 1357–1385.

Gunasekaran, A., E. Tirtiroglou, V. Wolstencroft. 2002. An inves- tigation into the application of agile manufacturing in an aerospace company. Elsevier Technovation 22 405–415.

Gustavsson, S. 1984. Flexibility and productivity in complex pro- duction processes. Internat. J. Production Res. 22 801–808.

Hammer, M. 1990. Reengineering work: Don’t automate, obliterate. Harvard Bus. Rev. 68 104–111.

Hardy, C., J. Thompson, H. Edwards. 1995. The use, limitations and customization of structured systems development methods in the United Kingdom. Inform. Software Tech. 37 467–477.

Harrison, A. 1992. Just-in-Time in Perspective. Prentice Hall, London. Hashimoto, T. 1980. Robustness, reliability, resilience and vulnera-

bility criteria for planning. Water Resources Res. 8 11–47.

Hashimoto, T., D. Loucks, J. Stedinger. 1982. Robustness of water resources systems. Water Resources Res. 18 21–26.

He, D., A. Kusiak. 1994. Design of products and assembly sys- tems for agility. ASME International Mechanical Engineering Congress and Exposition, New York.

Hines, P., D. Taylor. 2000. Going lean: A guide for implementa- tion. Lean Enterprise Research Centre, Cardiff Business School, Cardiff.

Hines, P., M. Holweg, N. Rich. 2004. Learning to evolve: A review of contemporary lean thinking. Internat. J. Oper. Production Management 24 994–1011.

Hines, P., R. Silvi, M. Bartolini. 2002. Demand chain manage- ment: An integrative approach in automotive retailing. J. Oper. Management 20 707–728.

Hirschheim, R., H. K. Klein, K. Lyytinen. 1995. Information Sys- tems Development and Data Modeling: Conceptual and Philosophi- cal Foundations. Cambridge University Press, Cambridge, UK.

Holweg, M., F. Pil. 2001. Successful build-to-order strategies start with the customer. Sloan Management Rev. 43 74–83.

Hong, M., S. Payander, W. Gruver. 1996. Modelling and analysis of flexible fixturing systems for agile manufacturing. Proc. IEEE Internat. Conf. Systems, Management and Cybernetics, Beijing.

Huang, C. 1999. An agile approach to logical network analysis in decision support systems. Decision Support Systems 25 53–70.

Huber, G. 1996. Organisational learning: The contributing pro- cesses and the literatures. M. Cohen, L. Sproull, eds. Organi- sational Learning. Sage Publishers, Newbury Park, CA.

Hunter, R., R. Thayer. 2001. Software Process Improvement. IEEE Computer Society Press, Piscataway, NJ.

Iivari, J., J. Maansaari. 1998. The usage of systems development methods: Are we stuck to old practices? Inform. Software Tech. 40 501–510.

Iivari, J., R. Hirschheim, H. K. Klein. 2000. A dynamic framework for classifying information systems development methodolo- gies and approaches. J. Management Inform. Systems 17 179–218.

James-Moore, S. 1996. Agility is easy but effective agile manufac- turing is not. IEE Colloquium 179 4–10.

Jin-Hai, L., A. Anderson, R. Harrison. 2003. The evolution of agile manufacturing. Bus. Process Management J. 9 170–189.

Johnson, D., J. Caristi. 2003. eXtreme programming and the software design course. M. Marchesi, G. Succi, D. Wells, L. Williams, eds. Extreme Programming Perspectives. Addison Wesley, Reading, MA.

Kahkonen, T. 2004. Agile methods for large organisations— Building communities of practice. T. Little, ed. Proc. Agile Development Conf. 2004, IEEE Computer Society, Salt Lake City.

Katayama, H., D. Bennet. 1999. Agility, adaptability and leanness: A comparison of concepts and a study of practice. Internat. J. Production Econom. 62 43–51.

Keen, P. 1980. MIS research: Reference disciplines and a cumula- tive tradition. E. Mclean, ed. Proc. First Internat. Conf. Inform. Systems, Philadelphia.

Keen, P. 1991. Relevance and rigor in information systems research: Improving quality, confidence cohesion and impact. H. Nissen, H. Klein, R. Hirschheim, eds. Information Systems Research: Contemporary Approaches & Emergent Traditions. North-Holland, Amsterdam.

Khaled, R., P. Barr, J. Noble, R. Biddle. 2004. System metaphor in extreme programming: A semiotic approach. Presented at The 7th Internat. Workshop Organ. Semiotics, Setúbal, Portugal.

Kidd, P. 1994. Agile Manufacturing: Forging New Frontiers. Addison- Wesley, Reading, MA.

Kircher, M., D. Levine. 2001. The XP of tao: eXtreme program- ming of large open-source frameworks. Extreme Programming Examined. Addison-Wesley, Reading, MA.

Kircher, M., P. Jain, A. Corsaro, D. Levine. 2001. Distributed extreme programming. Proc. XP2001—eXtreme Programming and Flexible Processes in Software Engrg. Villasimius, Sardinia, Italy.

Kusiak, A., D. He. 1997. Design for agile assembly: An operational perspective. Internat. J. Production Res. 35 157–178.

Kussmaul, C., R. Jack, B. Sponsler. 2004. Outsourcing and off- shoring with agility: A case study. C. Zannier, H. Erdogmus, L. Lindstrom, eds. XP/Agile Universe. Springer, Calgary, Alberta, Canada.

Lamming, R. 1993. Beyond Partnership: Strategies for Innovation and Lean Supply. Prentice Hall, London.

Lee, A. 1999. Rigor and relevance in MIS research: Beyond the approach of positivism alone. MIS Quart. 23 29–34.

Lee, G. 1997. Reconfigurability consideration design of components and manufacturing systems. Internat. J. Advanced Manufactur- ing Tech. 13 376–386.

Lee, G. 1998. Designs of components and manufacturing systems. Internat. J. Advanced Manufacturing Tech. 13 157–178.

Levy, Y., T. J. Ellis. 2006. A systems approach to conduct an effective literature review in support of information systems research. Informing Sci. J. 9 181–212.

Lindbergh, P. 1990. Strategic manufacturing management: A proac- tive approach. Internat. J. Oper. Production Management 10 94–106.

Lindbolm, C. 1987. Alternatives to validity: Some thoughts sug- gested by campbell’s guidelines. Knowledge: Creation, Diffusion, Utilization 8 509–520.

Lindvall, M., D. Muthig, A. Dagnino, C. Wallin, M. Stupperich, D. Kiefer, J. May, T. Kahkonen. 2004. Agile software develop- ment in large organisations. IEEE Comput. 37 27–34.

Lings, B., B. Lundell. 2004. On transferring a method into a usage situation. B. Kaplan, ed. Information Systems Research: Work- ing Group 8.2—“Relevant Theory and Informed Practice: Looking Forward from a 20 Year Perspective.” Kluwer, Boston.

Lundell, B., B. Lings, A. Mattson, U. Arlig. 2004. Taking steps to improve working practice: A company experience of method transfer. B. Fitzgerald, E. Wynn, eds. IT Innovation for Adaptability and Competitiveness: IFIP Working Group 8.6. Kluwer, Boston.

Maskell, B. 1996. Software and the Agile Manufacturer-Computer Systems and World Class Manufacturing. Productivity Press, Portland, Oregon.

Mason-Jones, R., J. Naylor, D. Towill. 2000. Engineering of the leagile supply chain. Internat. J. Agile Management Systems 2 54–61.

Mather, H. 1988. Competitive Manufacturing. Prentice Hall, Englewood Cliffs, NJ.

McDowell, C., L. Werener, H. Bullock. 2003. The impact of pair programming on student performance, perception and persis- tence. Proc. 25th Internat. Conf. Software Engrg., IEEE Computer Society, Washington, DC.

Melnik, G., F. Mauer. 2003. Agile methods in learning environ- ments: Lessons learned. F. Maurer, D. Wells, eds. XP/Agile Universe 2003. Springer, Berlin/Heidelberg.

Metcalfe, M. 2004. Theory: Seeking a plain english explanation. J. Inform. Tech. Theory Appl. 6 13–21.

Miles, M., A. Huberman. 1999. Qualitative Data Analysis. Sage, London.

Miller, D. 1993. MIS as a discipline. ACM SIGCSE Bull. 2 27–32. Miller, R. 2002. When pairs disagree 1-2-3. D. Wells, L. Williams,

eds. XP/Agile Universe. Springer, Berlin/Heidelberg. Monden, Y. 1983. The Toyota Production System. Productivity Press,

Portland, OR. Moody, D. 2002. Validation of a method for representing large

entity relationship models: An action research study. Eur. Conf. Inform. Systems, Gdansk, Poland.

Morgan, D. 1997. Focus Groups as Qualitative Research. Sage, London.

Nagel, R., R. Dove. 1991. 21st Century Manufacturing. Enterprise Strategy. Iacocca Institute, Lehigh University, Bethlehem, PA.

Naylor, J., M. Naim, D. Berry. 1999. Leagility: Integrating the lean and agile manufacturing paradigm in the total supply chain. Engrg. Costs Production Econom. 62 107–118.

Nielsen, J., D. McMunn. 2005. The agile journey: Adopting XP in a large financial services organisation. H. Baumeister, M. Marchesi, M. Holocombe, eds. Proc. 6th Internat. Conf. eXtreme Programming and Agile Processes �XP 2005�, Springer, Berlin/Heidelberg.

Ohno, T. 1988. The Toyota Production System: Beyond Large Scale Production. Productivity Press, Portland, OR.

Olhager, J. 2003. Strategic positioning of the order penetration point. Internat. J. Production Econom. 85 319–329.

Oliga, J. 1988. Methodological foundations of systems methodolo- gies. Systems Practice 1 87–112.

Paulk, M., C. Weber, B. Curtis, M. Chrissis. 1995. The Capability Maturity Model: Guidelines for Improving the Software Process. Addison-Wesley, Reading, MA.

Piore, M. 1989. Corporate reform in American manufacturing and the challenge to economic reform. Mimeo, Deparment of Political Science, Massachusetts Institute of Technology, Cam- bridge.

Poole, C., J. Huisman. 2001. Using extreme programming in a maintenance environment. IEEE Software 18 42–50.

Poppendieck, M. 2001. Lean programming. Software Development Magazine 9 71–75.

Prahalad, C., G. Hamel. 1990. The core competence of the corpo- ration. Harvard Bus. Rev. 63 79–91.

Preiss, K. 1997. A systems perspective of lean and agile manufac- turing. Agility Global Competition 1 57–72.

Preiss, K., S. Goldman, R. Nagel. 1996. Cooperate to Compete: Building Agile Business Relationships. Von Nostrand Reinhold, New York.

Rasmusson, J. 2003. Introducing XP into greenfield projects: Lessons learned. IEEE Comput. 20 21–28.

Salipante, P., W. Notz, J. Bigelow. 1982. A matrix approach to liter- ature reviews. B. Staw, L. Cummings, eds. Research in Organi- sational Behaviour. JAI Press, Greenwich, CT.

Sarker, S. 2009. Exploring agility in distributed information systems development teams. Inform. Systems Res. 20(3) 440–461.

Scholtes, P., H. Hacquebord. 1988. Beginning the quality transfor- mation. Quality Progress 21 28–33.

Schonberger, R. 1982. Japanese Manufacturing Techniques: Nine Hid- den Lessons in Simplicity. The Free Press, NY.

Schwaber, K., M. Beedle. 2002. Agile Software Development with Scrum. Prentice Hall, Upper Saddle River, NJ.

Senge, P. 1998. The Fifth Discipline. Random House, New York. Sharafi, H., Z. Zhang. 1999. A method for achieving agility in

manufacturing organisations: An introduction. Internat. J. Pro- duction Econom. 62 7–22.

Sharafi, H., G. Colquohoun, I. Barclay, Z. Dann. 2001. Agile manu- facturing: A management and operational framework. J. Mech. Engrg. 21 857–869.

Sharp, J., Z. Irani, S. Desai. 1999. Working towards agile manu- facturing in the UK industry. Internat. J. Production Econom. 62 155–169.

Shingo, S. 1981. Study of the Toyota production systems. Japan Management Association, Tokyo.

Shingo, S. 1986. Zero Quality Control: Source Inspection and the Poka- yoke System. Productivity Press, Portland, OR.

Stapleton, J. 1997. DSDM: Dynamic Systems Development Method. Addison-Wesley, Harlow, England.

Stephens, M., D. Rosenberg. 2003. Extreme Programming Refactored. Addison-Wesley Longman Publishing Co., Boston.

Stotts, D., L. Williams, N. Nagappan, P. Baheti, D. Jen, A. Jackson. 2003. Virtual teaming: Experiments and experiences with dis- tributed pair programming. F. Maurer, D. Wells, eds. Extreme Programming/Agile Universe 2003. Springer, New Orleans.

Stratton, R., R. Warburton. 2003. The strategic integration of agile and lean supply. Internat. J. Production Econom. 85 183–188.

Succi, G., M. Marchesi. 2001. Extreme Programming Examined. Addison-Wesley Longman, Reading, MA.

Sutton, R., B. Staw. 1995. What theory is not. Admin. Sci. Quart. 40 371–384.

Swiss, J. 1992. Adapting total quality management (TQM) to gov- ernment. Public Administration Rev. 52 356–362.

Tan, B. 1998. Agile manufacturing and management of variability. Internat. Trans. Oper. Res. 5 375–388.

Towill, D., M. Christopher. 2002. The supply chain strategy conun- drum: To be lean or agile or to be lean and agile. Internat. J. Logist: Res. Appl. 5 31–44.

Upton, D. 1994. The management of manufacturing flexibility. California Management Rev. 1 72–89.

Van de Ven, A. 2007. Engaged Scholarship: A Guide for Organizational and Social Research. Oxford University Press, Oxford, UK.

Van Oyen, M., E. Gel, W. Hopp. 2001. Performance opportunity for workforce agility in collaborative and non-collaborative work systems. IEEE Trans. 33 761–777.

Vidgen, R., X. Wang. 2009. A co-evolving systems approach to the organisation of agile software development. Inform. Systems Res. 20(3) 355–376.

Vokurka, R., G. Fliedner. 1998. The journey toward agility. J. Indust. Management Data Systems 98 165–171.

Volberda, H. 1998. Building the Flexible Firm: How to Remain Com- petitive. Oxford University Press, New York.

Vonderembse, M., M. Uppal, S. Huang, J. Dismukes. 2006. Design- ing supply chains: Toward theory development. Internat. J. Production Econom. 100 223–238.

Vonk, R. 1990. Prototyping: The Effective Use of CASE Technology. Prentice Hall, London.

Wainer, M. 2003. Adaptations for teaching software development with eXtreme programming: An experience report. F. Maurer, D. Wells, eds. XP/Agile Universe. Springer, New Orleans.

Walsham, G. 2006. Doing interpretive research. Eur. J. Inform. Sys- tems 15 320–330.

Webster, J., J. T. Watson. 2002. Analyzing the past to prepare for the future: Writing a literature review. MIS Quart. 26 xiii–xxiii.

Weick, K. 1989. Theory construction as disciplined imagination. Acad. Management Rev. 14 516–531.

Welke, R. 1983. I.S./DSS: DBMS support for information system development. P. Holsapple, A. Whinston, eds. Database Man- agement Systems. Reidel, Dordrecht, The Netherlands.

Whetten, D. 1989. What constitutes a theoretical contribution? Acad. Management Rev. 14 490–495.

Williams, L., R. Kessler. 2000. The effects of “pair-pressure” and “pair-learning” on software engineering education. Proc. Thir- teenth Conf. Software Engrg. Ed. Training, IEEE Computer Soci- ety, Austin, TX, 59–65.

Womack, J., D. Jones. 1994. From lean production to the lean enter- prise Harvard Bus. Rev. 72 93–104.

Womack, J., D. Jones. 1996. Lean Thinking: Banish Waste and Create Wealth for Your Corporation. Simon and Schuster, New York.

Womack, J., D. Jones, D. Roos. 1990. The Machine That Changed the World. Rawson Associates, New York.

Wynekoop, J. L., N. L. Russo. 1995. Systems development method- ologies: Unanswered questions. J. Inform. Tech. 10 65–73.

Yin, R. 2003. Case Study Research: Design and Methods. SAGE Pub- lications, Thousand Oaks, CA.

Young, K., R. Muehlhaeusser, R. Piggin, P. Rachitrangsan. 2001. Agile control systems. J. Automobile Engrg. 215 189–195.

Yusuf, Y., E. Adeleye. 2002. A comparative study of lean and agile manufacturing with a related survey of current practices in the UK. Internat. J. Production Res. 40 4545–4562.

Yusuf, Y., M. Sarhadi, A. Gunasekaran. 1999. Agile manufactur- ing: The drivers, concepts and attributes. Internat. J. Production Econom. 62 23–32.

Zain, M., N. Kassim, E. Mokhtar. 2002. Use of I.T. and I.S. for organisational agility in Malaysian firms. Singapore Manage- ment Rev. 25 25–34.

Zhang, Z., H. Sharafi. 2000. A methodology for achieving agility in manufacturing organisations. Internat. J. Oper. Production Man- agement 20 496–512.