small business growth and internal transparency: the role

Street & Meister/Small Business Growth & Internal Transparency

MIS Quarterly Vol. 28 No. 3, pp. 473-506/September 2004 473

SPECIAL ISSUE

SMALL BUSINESS GROWTH AND INTERNALTRANSPARENCY: THE ROLE OFINFORMATION SYSTEMS1

By: Christopher T. StreetQueen’s School of BusinessQueen’s UniversityKingston, Ontario K7L [email protected]

Darren B. MeisterRichard Ivey School of BusinessUniversity of Western OntarioLondon, Ontario N6A [email protected]

Abstract

While many large businesses start out as a smallenterprise, remarkably little is known about how anorganization actually changes internally during theperiods of growth. Small business growth isknown to strain internal communication processes,for example, which likely limits growth oppor-tunities. Information systems are often called uponto remedy such deficiencies. Through a parti-cipatory action research project, we investigatedthe ways in which a small business management

1Michael Myers was the accepting senior editor for thispaper.

team developed an IS-enabled solution to addresstheir growth needs. During the progression of theproject, a new outcome of organizational effec-tiveness, internal transparency, was identified anddeveloped. Adopting a punctuated equilibriumperspective, a theoretical process model is pro-posed that sheds light on a relationship betweeninternal transparency, small business growth, andIS. The paper concludes with observations thatinternal transparency may well be a concept thatoffers significant potential for MIS research as wellas a discussion about the applicability and credi-bility of participatory action research for thisproject.

Keywords: Business growth, IT investment,small business, internal transparency, processmodels

Introduction

Small businesses are an important and integralpart of every nation’s economy and have beenlong recognized as different from large businesses(Hambrick and Crozier 1985). In order to grow,small businesses must evolve their organization,incorporating changes to management structure,operational planning, control, and communicationprocesses (Hanks 1990; Steinmetz 1969), without


474 MIS Quarterly Vol. 28 No. 3/September 2004

impairing the firm’s competitive advantage. Failureto make these changes may result in harm to thebusiness through stagnation, negative growth, lossof customers, and failure to introduce newproducts, potentially closing the business(Churchill and Lewis 1983; Hambrick and Crozier1985). Information systems are often relied on toassist growth, although small businesses often findtechnology difficult to implement due to resourceconstraints (Raymond 1985).

There is significant research about the role of ISin small businesses (e.g., Cragg and King 1993;Harrison et al. 1997; Igbaria et al. 1997). Whilethis research informs ours, this paper enters a newarea: what happens to the IS of a small businessas a growth period begins. Rather than seeking tounderstand the effect of the traits of anorganization on important phenomena such asselection, adoption, and implementation, we focuson how those traits affect an organization’s ISchanges during the business’ transition period.These transitions may be due to external factors(e.g., a general economic downturn requiring adecrease in size). Alternatively, a company’smanagement may decide that they wish to try togrow proactively, through targeting a larger ormore lucrative market. Our interest was therecursive relationship between this decision andthe IS in the business: specifically, would there bea need to change the existing systems, whattriggers this need to change, and what new orrevised demands would these decisions have onthe existing IS infrastructure?

A useful analytical framework is provided bypunctuated equilibrium theory (PET) (e.g., Gersick1991; Romanelli and Tushman 1994; Tushmanand Romanelli 1985), a theory that has beenadapted from evolutionary biology into manage-ment theory. Researchers who use PET typicallyfocus on three concepts: evolutionary periods,revolutionary periods, and deep structures. Froma PET perspective, it can be argued that organi-zational development is characterized by stableperiods of evolutionary change that are occa-sionally punctuated by periods of rapid change(Gersick 1991; Van de Ven and Poole 1995). Forexample, an organization’s IS can be seen as a

form of deep structure. Managerial behaviors canalternately perpetuate or weaken existing struc-tures (Orlikowski 2000; Orlikowski and Robey1991). PET suggests that behavioral changes inresponse to punctuations cause structuralchanges. For example, the decision to change ISstrategy and implementation in response tocorporate strategy can be seen as a change in thedeep structure as a response to short periods ofsharp, extensive change (punctuation) (Sabherwalet al. 2001). From this perspective, the decision togrow can also be thought of as a punctuation andthe information system as a deep structure thatmay change as the management team responds tothe punctuation.

To address our research question, we joined withthe five-member management team of a smallCanadian manufacturing company to develop aresearch partnership using participatory actionresearch (PAR). Before the project, the companyowner had been searching for assistance toinvestigate how IS could be used to assist his firmthrough a planned growth period. This can beseen as his initial response to the punctuation. Asexpected, there were other behavioral changes inresponse to the punctuation from other manage-ment team members. The organization providedan excellent opportunity to investigate our area oftheoretical interest: the effect on the deep struc-ture of the information system of a small businessin response to this type of punctuation. Throughthe project, a request for quotation (RFQ, adocument upon which potential vendors can bid fora contract) for a new system was developed,satisfying the company’s practical goals. In theend, a richer understanding of how growth pres-sures a company’s information resources wasreached, satisfying our research goals.

The paper proceeds as follows. Literature fromthe small business growth and information com-munication areas is presented. The description ofa research project designed to address the ques-tions posed above is followed by a narrativeaccount of the project itself. The critical eventsthat occurred during the project are then dis-cussed, and an explanation is presented thatresolves the issues raised during the project into a


MIS Quarterly Vol. 28 No. 3/September 2004 475

cohesive picture of what was happening in thecompany during the growth process. The paperconcludes with the proposal of a process modeland a discussion of the research and managerialimplications of this project.

Theoretical Perspectives

The small business and information technologyliterature has seen considerable progressaddressing specific adoption decisions and imple-mentation practices. Our contribution lies inunderstanding the process by which a smallbusiness information system is changed inresponse to the organization entering a growthphase. To do this, we will integrate three researchperspectives. First, an overview of the smallbusiness literature is provided. Second, wedevelop and define a construct, internal trans-parency, which reflects the extent to which themanagement team understands the activities andoutcomes of the organization, which is partiallydetermined by the team’s communicationbehavior. Finally, small business growth models,and their framing of management team behaviors,specifically information sharing and communicationbehavior, are reviewed.

Small Businesses

Defining a small business is a controversial topic.Like many authors, we have adopted a commonlyused definition from the U.S. Small BusinessAdministration: a small business is independentlyowned and operated and not dominant in its fieldof operation (Small Business Administration 2003).For our discussion of small business, we areexcluding microbusinesses of less than fiveemployees (Hodgetts and Kuratko 2001) andfocusing our attention on small manufacturingenterprises interested in growth. For these busi-nesses, an important source of competitiveadvantage is the ability to remain flexible andresponsive to the business environment (Barringeret al. 1998). While this adaptability is beneficialduring transition periods, small businesses are

more challenged than large companies byresource constraints such as access to financialcapital, and technical or managerial skills, whichoften significantly reduce the number and type ofoptions available to management (Hodgetts andKuratko 2001; Iacovou et al. 1995).

Small businesses are typically characterized by aflat organizational hierarchy and close proximity tocoworkers, which is believed to contribute toeffective communication practices, often com-prised of informal channels (Vinten 1999), andtypically carried out face-to-face as the need arisesrather than through regularly scheduled meetings,formalized status reports, or structured briefings.These channels are considered a significantbenefit of the small business environment,providing superior operational flexibility (Wickertand Herschel 2001) and responsiveness (Monta-zemi 1988). These communication practices allowthe small business manager to understand verywell what is going on within the firm.

This awareness is threatened, however, duringsmall business growth as changes in organi-zational structures, such as levels of organizationalcomplexity, formalization, and centralization(Churchill and Lewis 1983; Scott and Bruce 1987),break down existing ways of working. Further, asgrowth occurs, managerial capacity constraints(Jensen and Meckling 1976; Oi 1983) imply thatexisting behaviors are further reduced in frequencyas new behaviors are adopted to manage thegrowing firm. As small businesses undergo thesechanges, a differentiating factor between success-ful and unsuccessful firms is that successful firmsact in “anticipation of bigness” (Hambrick andCrozier 1985). These firms proactively lay thefoundation for the bigger enterprise before growthoccurs, thereby preempting avoidable barriers togrowth such as reactive management.

Growth stage theories provide a measure ofpredictability regarding what to expect in anti-cipation of getting bigger (Churchill and Lewis1983; Scott and Bruce 1987). However, whilethere is significant literature on the stages ofgrowth theory, it is focused at the firm level ofanalysis and describes characteristics (particularly



in relation to complexity, formalization, andcentralization) that are likely to be present at acertain growth stage. There is little literature thatdescribes the evolution of these characteristics ormanagerial actions at the start of a growth phase,and particularly so at the level of the individual ormanagement team. This is surprising as it is well-accepted that as a result of growth, managersbecome removed from business operations, whichcreates a broad set of difficulties in areas of com-munication, coordination, and control (Churchilland Lewis 1983; Greiner 1972; Olson and Terpstra1992). The dynamic process by which theseproblems arise seems to be under-studied. Asgrowth occurs, the management team becomeless involved with daily matters and begins toreceive information indirectly from many sources.As organizational complexity and formalizationincreases, informal communication practices mayno longer provide the necessary level of mana-gerial information (Bavelas 1951; Leavitt 1962).Therefore, we would anticipate that an outcome ofsmall business growth would be that manage-ment’s communication behaviors will change andthat business outcomes will be affected by thedegree to which these changes constrain orfacilitate the availability of requisite information fordecision-making purposes.

While stage models have been criticized asatheoretical and simplistic (e.g., Stubbart andSmalley 1999), they are useful in understandingthe organizational aspects of what should changein a business. However, they are ineffective inanswering our research questions, which concen-trate on why and how things change. As this studyis focused on the start of a growth period, afterinitiation but before success can be determined,growth stage models would identify the position atwhich we sit, the expected characteristics, perhapsthe characteristics of the next plateau, butunfortunately little else.

So, while our investigation is informed by growthstage theories, punctuated equilibrium theory(e.g., Gersick 1991) is a more appropriate framefor the type of small business growth underinvestigation. Newman and Robey (1992) pro-posed that PET may be a useful perspective for

modeling the development of IS within organi-zations. As well, Sabherwal et al. (2001) used PETto investigate the relationship between strategicbusiness and IS alignment. Drawing upon casestudies of three large organizations, they foundthat the evolution of an organization’s IS, includingsustained periods of misalignment, could beanalyzed using a PET lens. As mentioned earlier,an organization’s IS can be seen as a deepstructure, decisions to change corporate strategyas punctuations, and changes over time asresponses to regain equilibrium within theorganization. This study follows Sabherwal et al.in viewing a change in business strategy as apunctuation.

Small Business and IS

While small businesses have been traditionallyseen as reluctant to invest in IS (Lees and Lees1987), evidence over the past decade shows anincrease in the awareness and management of ISin small businesses by owners and managers(e.g., Ballantine et al. 1998; Bergeron et al. 2001;Hussin et al. 2002). IS research has considered avariety of these situations. First, technologyadoption problems have been investigated whereinfactors such as the role of the president/CEO(Cragg and King 1993; Thong 1999; Winston andDologite 2002), perceived usefulness or relativeadvantage (Cragg and King 1993; Thong 1999),and ease of use (Iacovou et al. 1995; Thong 1999)were identified. Similarly, Harrison et al. (1997)used the theory of planned behavior to explain theactions of small business executives with respectto IS decisions. Implementation has also beeninvestigated. The small business owner’s attitudetoward IS is understood to be an important factorin determining implementation success (Winstonand Dologite 2002). Training and on-going usersupport are other post-implementation issues ofimportance (Igbaria et al. 1998; Zinatelli et al.1996).

An important common theme in this research oftenconflates the president’s role as businessexecutive and decision maker with his role as anindividual user, which is reasonable given the



president’s central role in making organizationaladoption decisions. In general, understanding theindividual roles of managers is considered pivotalto understanding the way in which IS is adoptedand implemented in small businesses.

Internal Transparency

In general, IS researchers have found resultsconsistent with those from other managementfields that show the owner and managers of asmall business as being intimately involved in thedaily operations of the small business, gatheringand acting on information first-hand. A measur-able result of the management team’s communi-cation process should then be the degree to whichmanagement personally understands what is goingon throughout the enterprise. While many authorsdiscuss such outcomes without specificallydefining an outcome measure (arguing more isbetter), there is a consistent theme in severalliteratures that supports a construct called internaltransparency.

Two different types of transparency can bedefined: internal and external. External trans-parency corresponds to the outcome of communi-cation behaviors directed outside the organization.For example, within the accounting and financeliterature, transparency is considered as theobservability of transactions, for both investment(e.g., Phillips et al. 2002) and regulatory (e.g.,Vishwanath and Kaufmann 2002) purposes.Similarly, from the supply chain managementperspective, information exchange between supplychain partners (e.g., Lamming et al. 2001) isdescribed as a type of transparency. Within themarketing literature, information flow from thecustomer is seen to be valuable (e.g., Narver andSlater 1990). Further, the positive role of IS inincreasing transparency has also been highlighted(Day and Wensley 1988; Min et al. 2002).

Internal transparency corresponds to thesebehaviors but is applied within the organization(e.g., Alavi and Leidner 1999). In the organiza-tional behavior literature, it is also considered anoutcome of communication behaviors. One

example is when supervisors hold frequentmeetings to share information with subordinates todisseminate requisite information to meetindividual, team, and organizational goals (Beechand Crane 1999). Communication behaviors canalso decrease transparency. Failures to shareinformation through practices such as screeningout often lower the ability of decision makers tomake decisions (Pfeffer and Salancik 1978).Operations management researchers have foundsimilar results with respect to work teams (e.g.,Ang et al. 2000; Forza 1995).

The IS literature expands on the role of IS increating transparency, while frequently not con-ceptualizing the outcome. IS are used to enableinformation sharing between individuals (Alavi andLeidner 1999) and organizations (Braunstein1999). Technology is also credited with helping tokeep information up-to-date, fresh, and dynamic(Schwartz and Te’eni 2000). In this perspective,IS are seen as an enabler of transparency.

A conceptual definition of transparency can beconstructed by comparing the similarities anddifferences within this literature. First, trans-parency is regarded as an outcome measure ofcommunication behaviors (Ang et al. 2000; Beechand Crane 1999). Second, transparency is alsoseen as an outcome of an exchange processbetween two or more entities (Alavi and Leidner2001; Lamming et al. 2001). From this, we defineinternal transparency to be an outcome of com-munication behaviors within an organization thatreflects the degree to which employees haveaccess to the information requisite for theirresponsibilities. An important enabler of the com-munication behaviors is an organization’s tech-nology-based IS, as several literatures recognizethe important enabling role of IS in increasingtransparency (Alavi and Leidner 2001; Min et al.2002).

Commonalities throughout the literature allow us toconsider how to operationalize internal trans-parency. First, transparency is normally con-sidered as an ordinal measure exhibiting signs ofless and more (Forza 1995; Narver and Slater1990). Second, it is frequently discussed as a



mediator: communication behaviors create a levelof transparency which itself affects outcomes suchas performance (Braunstein 1999) or coordination(Lamming et al. 2001).

Thus we see internal transparency as an ordinalmeasure expected to mediate relationshipsbetween communication behaviors and outcomessuch as planning ability or decision making. It isnot a measure of the information processingcapabilities of an individual or an organization,although an increased ability to process infor-mation can certainly be expected to impacttransparency. Internal transparency represents atleast one way to examine how changes to com-munication behaviors in response to organizationalgrowth pressures affect, and are affected by,changes to the existing IS structures of a smallbusiness.

Summary

A positive characteristic of small businesses is thetight communication and coordination within theorganization (Montazemi 1988; Wickert andHerschel 2001). These businesses operate undersignificant constraints with respect to capital,managerial time, and expertise. At times ofgrowth, these conditions stress the organizationand may lead to changes in information sharingand communication behaviors, which will likelyhave an impact on internal transparency. In-creased use of IS might be expected to enable anincrease in the internal transparency available tothe firm as it grows. However, we are interested inthe process by which the existing IS becomesseen as no longer supporting internal transpar-ency, and how this further affects ongoing com-munication practices. Understanding this effect iscrucial because without requisite communication,the organization may find itself unable to sustainits activities, much less grow. We have developeda definition of internal transparency to help usunderstand and discuss how changes in infor-mation sharing and communication behaviorsimpact managerial decision making, which allowsus to further examine the relationships betweensmall business growth and organizational IS. In

the next section, we introduce the research siteand project that we used to understand theseprocesses.

Research Method

Our research goal is to address why and how theIS of a small business had to change in responseto a specific punctuation, particularly throughchanges in communication behavior. As wewanted to work with a real organization on a realproblem, we had to find an interested organization.A government-funding agency was the catalyst inthis search process. The agency’s primary goalwas to sponsor university-industry collaboration inthe manufacturing sector. Specifically, theyencouraged placement of researchers directly inthe organization. Earlier in 2001, the fundingofficer had spoken to one of the researchers abouttopics of interest. She had also met with thepresident and owner (Nick) of a small Canadianelectronics manufacturer, ELCO (all names arepseudonyms). Nick, aware of the program’s goals,was seeking assistance in evaluating andspecifying a new information system for ELCO.The second author’s name was mentioned duringthe conversation as a potential resource andintroductions were made. The reader should beaware that the funding officer would speak tohundreds of companies and dozens of researchersin the course of a year in order to find mutuallycompatible pairs and generate projects.

The role of the agency officer in setting expec-tations on both sides cannot be understated. Shemade it clear to each participant that both party’sinterests would jointly need to be satisfied in orderto meet the agency’s goals. Through twomeetings (July and August 2001) and a series ofe-mails, Nick and the authors worked their waythrough Rapoport’s (1970) initiative dilemma: howto specify a research project that satisfies both aresearch question and a practical business need.Nick’s requirements were an impartial assessmentof ELCO’s IS needs, a plan to fulfill those needs,and an RFQ specifying system requirements forvendors.



As researchers, we were interested in ELCO’sinformation system needs, as well as under-standing why these needs existed. For example,one issue that Nick initially brought up was that hismanagement team was having difficulty gettingstarted performing a systems analysis indepen-dently. We responded that, as it addressed ourresearch questions, we were also interested in whythey were experiencing difficulties, and where theroot cause of any problems might lie.

In retrospect, the research proposal was devel-oped quickly for two primary reasons. First, bothparties had clear objectives before meeting.Mutual goals that normally have to be balancedthrough a negotiation process (Kock 1997) existedbut were flexible to the extent that both sides wereable to satisfy their needs. For example, theauthors were not as concerned with the specificneeds of the company as they were with thegeneric process being experienced. Similarly,while Nick accepted that efforts would be directedtoward understanding the “hows and whys” of theprocess, he also insisted from the start that thecompany’s goals be met. Second, the fundingofficer made clear to each side that the partici-pants’ respective needs had to be met in order tosatisfy the agency’s goals. Third, Nick did notwant to engage a consultant as he thought thatthey would be biased toward a “big IS” solution,whereas he thought that academic researcherswould be less biased to such an outcome. It couldbe said that Nick and the authors formed the coreteam in the research project; the full researchteam would include the remainder of ELCO’ssenior management.

Research Design

To address this research problem, the authorsadopted an interpretivist perspective of scientificrealism (Psillos 1999). A scientific realist viewtypically holds that knowledge is socially con-structed. Researchers attempt to understand aprocess in terms of how the individuals involvedcomprehend and attach significance to how the

process came about, how it currently exists, andhow it is likely to exist in the future (Orlikowski andBaroudi 1991). Consistent with the intent of ourresearch questions, and with the nature of theinterpretivist view, the research process focusedon making sense of the changing organizationalsituation as it emerged (Singer 1999).

The congruence of our goals, the funding agency’sgoals, and Nick’s goals led the core team tochoose action research (e.g., Elden and Chisholm1993; Rapoport 1970; Susman and Evered 1978)as the research method for the following reasons.First, ELCO believed that it was necessary that theauthors apply their technical and process expertiseto the practice problem. Second, the authorsbelieved that action must be taken in the companyin order to answer the research questions.Particularly, it was thought that only through actualanalysis of IS needs would light be shed on thereasons why ELCO’s IS needs seemed to change.Third, it would be possible to satisfy both interestswithout compromising the integrity of either thepractical or theoretical results.

To refine our plan, we drew on the general actionresearch process proposed by Susman andEvered (1978). This research process has beenused successfully for IS research in the past (e.g.,Baskerville 1999; Davison and Vogel 2000; Olesenand Myers 1999) and was a good fit with ourneeds. Specifically, we adopted participatoryaction research (Baskerville 1999; Eden andHuxham 1996) as it allows researchers to becomedeeply involved with the organization (Olesen andMyers 1999) and engages the practitioners directlywith research questions (Eden and Huxham 1996).Greater involvement is beneficial in circumstanceswhere researchers gain investigative value from aninsider’s view of a problem context, and where thechange process itself is the subject of inquiry(Davison 1999).

ELCO did not have a standard development plantemplate that we would follow. As part of theresearch process, the authors suggested a three-stage process to structure the project. The plan



was to perform an intensive requirements analysisin the first stage, followed by a planning process inthe second stage for the purpose of developing anoutline to address the issues raised in the firstphase, and ending with a development stagewhere a request for quotation (RFQ) and imple-mentation plan would be generated. Readers willrecognize this process as a segment of a standardsystem development life cycle model (e.g.,Pressman 2001), commonly taught in manycomputer science, engineering, and IS programsas well as employed by many organizations. Bothauthors had academic and practical experiencewith these methods and felt comfortable in leadingthis process, which met with Nick’s expectations.

The second and third stages built upon datagenerated from each previous stage, reflecting theiterative nature of action research (Lau 1999).Overall, we planned for three iterations of thegeneral action research model of diagnosing,planning, taking action, evaluating the actions, anddrawing key lessons and research findings fromeach stage. These three iterations coincided withthe deliverables in the project. As we would befinalizing the next intervention after the completionof each deliverable, it seemed sensible to have theaction research cycles coincide with these as well.

Narrative-based theory development, as iscommon to the action research literature (e.g.,Chiasson and Dexter 2001; Olesen and Myers1999), typically involves moving from empiricalobservation to generalizable relationships (Van deVen and Poole 1995). This is a challengingundertaking that first requires explicating ofcharacters, events, and the causal relationships ina process as the basis of theory development(Langley 1999). Using the framework provided inFigure 1 provided us with the structure andguidance to manage Rapoport’s (1970) roledilemma, in which the researchers need toconcern themselves with the practice and theoryproblems. In the action planning and action takingstages, we were comfortable, as was expected,applying our expertise and knowledge to ELCO’soperations and working as part of the management

team. As outlined in our reporting section, weactively collaborated with ELCO’s managers inplanning the next phase of action and participatingin its execution. Following the action planning andtaking, we took the opportunity to evaluate ourresults. We did this by directly asking the man-agers for their opinions and through groupdiscussions. After taking stock, we moved to thefinal two stages, specifying learning and diag-nosing, where we concentrated on our researchquestions. In specifying learning, we focused onwhat we learned about ELCO. In diagnosing, wedirected our perspective toward abstraction. Wethen presented our developing interpretations tothe entire team to gain their feedback. Aftersuitable refinements, the diagnosing stage guidedour actions in the next action planning stage. Thisframework enabled us to move effectively fromobservations to identifying the underlying rela-tionships as required for theory development.

The research plan was formalized in a writtenproposal as a six-month project running fromSeptember 2001 until February 2002 and wasjointly submitted to the government funding pro-gram coordinator by the core team. Initially,ELCO’s owner indicated there should be two partsto the research project: first, to determine ifchanges to the IS were truly necessary, andsecond, if changes were required, what theyshould be. The research plan proposal called forthe authors to deliver the final project report in theform of a system specification, or RFQ, suitable forpotential vendors. In return, ELCO agreed toprovide direct management time equal to theamount of time put in by the authors, a figureamounting to one day per month per manager forthe duration of the project (in the end significantlymore time was committed by all). As was normalfor these projects, a research budget of $16,000(Canadian) was established to cover primarilytravel—ELCO is approximately 100 km from theuniversity—and other research expenses. Theagency expected a copy of the final report andregular feedback regarding the wider applicabilityof the analysis results to other small manufac-turers. This paper would also be considered partof the feedback.



ActionPlanning

ActionTaking

Evaluating

SpecifyingLearning

Diagnosing

Description

Commentary

Theory Building

ActionPlanning

ActionTaking

Evaluating

SpecifyingLearning

Diagnosing

ActionPlanning

ActionTaking

Evaluating

SpecifyingLearning

Diagnosing

Practice Output (RFQ)

Research Model and

Implications

ActionPlanning

ActionTaking

Evaluating

SpecifyingLearning

Diagnosing

ActionPlanning

ActionTaking

Evaluating

SpecifyingLearning

Diagnosing

Description

Commentary

Theory Building

ActionPlanning

ActionTaking

Evaluating

SpecifyingLearning

Diagnosing

ActionPlanning

ActionTaking

Evaluating

SpecifyingLearning

Diagnosing

ActionPlanning

ActionTaking

Evaluating

SpecifyingLearning

Diagnosing

ActionPlanning

ActionTaking

Evaluating

SpecifyingLearning

Diagnosing

Practice Output (RFQ)

Research Model and

Implications

Figure 1. Research Framework

Data Collection and Reflection

Project data were continually collected, analyzed,reported, and revised throughout the project. Datatook many forms, ranging from the e-mailsbetween project members over the six-monthperiod, operational, financial, and strategicplanning company records, website data, workingnotes from project meetings and presentations, aswell as transcripts from three rounds of personalinterviews between the authors and managementteam members, periodically spaced throughout theproject. The dataset represented both the formalthoughts and intentions of the management team,captured by documents such as strategic plansand formal presentations of findings, as well as theinformal responses and opinions, captured throughinterview and e-mail transcripts. More than 500pages of documents were collected over thecourse of the project.

In order to include managers in the research teamand to collect data from multiple rich sources (assuggested by Baskerville and Pries-Heje 1999)while dealing with an operating manufacturingplant, almost all project team meetings occurred atELCO. Further, most interviewing as well asobservations of the work environment took place inthe personal offices and meeting rooms at ELCO.Specific sessions that were thought to demand fewinterruptions were scheduled off-site as necessary.

For example, a strategy planning session for allparticipants was held at off-site facilities.

The Research Site

Founded in the 1950s, ELCO manufactures high-accuracy electronic equipment with an inter-national reputation for product quality. The com-pany built their product line through the intro-duction of innovative technologies, incorporatingleading edge research directly from governmentlaboratories (not related to the funding agency).The company gained maturity and marketpresence but never grew very large. In the mid1990s, the company went through a severe cashcrisis. ELCO survived through a loyal customerbase and careful management and, in 1999, Nickpurchased ELCO and became president and CEO.By 2001, 32 employees staffed engineering,manufacturing, and marketing functions andannual revenues were approximately $5 million(Canadian).

For the duration of the research project, themanagement team consisted of five individuals:Nick; Jerry, the general manager; Diane, thecontroller; Bruce, the manufacturing manager; andTim, the engineering manager. Jerry, Diane, andBruce had each been with ELCO for over 15years, while Tim came to ELCO shortly before



Nick’s purchase. While this team may seem largefor a small business (5 out of 32 employees), itrepresented the entire management team: allother employees reported directly to one of thesemanagers. It was also the same managementstructure that had managed 70 employees beforethe mid-1990s cash crunch. This firm is differentthan many studied in small business and entre-preneurial research in that it was nearly 50 yearsold. The management team understood that Nickbought the company as his last entrepreneurialendeavor and that his primary goal was to growthe company over a five-year period, then sell hisinvestment and retire.

The full research team consisted of the twoauthors and the five managers. Having a researchteam comprised of both ourselves and the existingmanagement team was consistent with both ourresearch ethics and with the goals of participatoryaction eesearch (Baskerville 1999; Singer 1999).In accordance with Eden and Huxham (1996),PAR participants should be regarded as equal andvoluntary members of the research team involvedin the constructive process of probing, solving areal organizational problem and considering theresearch question.

While distinct roles existed within the team, thesechanged over the course of the project. Forexample, the authors’ role during the openingweeks of the project involved gathering datarelated to ELCO’s operations, barriers to effectiveaction, and synthesizing a shared view of thesituation. ELCO’s managers were initially to pro-vide their interpretations and viewpoints of whatthe important problems and significant barrierswere, as well as evaluating the results and pro-viding feedback to the investigators during theanalytical portion of the project. In many ways, theauthors’ role was initially one of analyst, whilemanagement’s role was that of informant. Theseroles reversed over the course of the project asthe task moved from identifying the problems tofinding and implementing possible solutions. Theauthors gradually became advisers to themanagers as they worked toward solving ELCO’sproblems.

As mentioned earlier, Nick wanted ELCO to growso as to maximize his investment. In our initialmeetings, Nick identified two key managerialproblems: how to reduce the amount of time andeffort spent “wading through paper,” and how themanagement team could be assisted in planningand implementing their individual strategies tosupport ELCO’s growth. Nick thought the twoquestions were interrelated: reducing time spentassembling information would enable more time tobe spent on growth planning. He also thought thatimproved IS could address these problems. Forexample, he mentioned that perhaps an enterpriseresource planning (ERP) product would help, buthe was uncertain ELCO could afford it.

The managers felt that they were not developing agood growth plan for the future and were losingcontact with current operations. The managementteam worried that ELCO’s IS would be unable tosupport the upcoming growth. For reasonsprimarily related to the mid-1990s cash crisis, theIS had become outdated and incompatible acrossfunctions. For example, there was no local areanetwork; a single computer was connected to theInternet via dial-up connection.

The manufacturing and inventory functions weresupported using an early 1980s minicomputer,which was, in the managers’ view, held togetherthrough a combination of “binder twine” and thegrace of Diane, who was responsible for itsprogramming and maintenance but had no formaltraining. On the other hand, employees per-forming sales and marketing functions usedrelatively recent PCs and office productivitysoftware. The engineering department used oldand unsupported software on new PCs. Certainbusiness processes, such as accounting andfinancial management, used more than onesystem with data shared through reentry. Manyroutine management tasks required significantmanual assembly of information across multipleplatforms and numerous informal requests forinformation. During the authors’ early visits toELCO, deep paper stacks were on each mana-ger’s desk. When asked what information wascontained in a 5 to 10 page printout, the managerwould mention either one or two important



numbers or shrug uncomplainingly. The similaritywas striking between this firm in 2001 and thosedescribed six years previously by Cragg andZinatelli (1995). It was clear that ELCO had notadopted many possible technology innovations inthe past decade, in spite of the fact that theirproduct required leading edge electronicstechnology.

In preliminary discussions, the management teamexpressed goals very much in line with thosefound by Pollard and Hayne (1998) including usingIS for competitive advantage, developing IS projectmanagement skills, building responsive IS,aligning IS and business planning, coping with thetechnological change, planning and managingcommunications networks, facilitating and man-aging business process redesign, educating users,and recruiting and developing IS humanresources. One exception was that ELCO’smanagement expressed opposition to creating anysoftware development capability. Cragg and King(1993) found that enthusiasm for the technologywas a precursor to increased IS use in smallbusinesses and that resource constraints andlimited education were among the factors thatinhibited applications growth. Both were demon-strated at ELCO. Therefore, the authors drew theobservation that this was a representative smallmanufacturing firm with typical goals, limitations,and current environment.

What emerged for the authors was a perspectivethat Nick’s decision to grow the company was apunctuation for ELCO. The management team’sbelief that the IS needed to change could be seenas a weakening of one of ELCO’s deep structures.What was not clear was what behavioral changesled to this weakening, as would be suggested by aPET perspective (Gersick 1991).

Reporting the ActionResearch Cycles

For each of the three action research stages, wepresent the events and the analysis and discuss

the results. While stages occurred sequentially,the boundaries as described in Figure 1 betweeneach stage were not as unidirectional as thediagram would make them appear. While ingeneral the stages occurred in that order, theproblem resolution and theory development ac-tions often occurred concurrently, as is expected inPAR (Baskerville 1999). As the authors completedthe documentation and analysis of each stage, itwas turned back to the research team fordiscussion and revision before moving on.

Stage 1: Baseline Analysis

As the practical goal of the project was torecommend potential changes to the organi-zation’s IS, the first actions were to follow generalsystem development practices and begin withdetermining current organizational practices anddirections. We did this recognizing that themanagers felt that the IS had to change; weneeded to understand why this belief haddeveloped.

Action Planning

Project scope and data collection methods werereviewed at the first research team meeting. Wedecided to capture current practice and futurerequirements by involving ELCO’s employeesdirectly in identifying their job tasks and infor-mation requirements. The authors also inter-viewed each manager individually for their opinionson future requirements. Two generalquestions—what are your short- and long-termfunctional goals and what barriers or challenges doyou see that might make it difficult to achievethose goals—formed the basis for these open-ended interviews. By examining the congruencybetween the managers’ future plans and intentionsand comparing this to the managers’ experiences,a clear picture of where difficulties were occurringwas expected to emerge.



Figure 2. Organization-Level Process Diagram of ELCO Manufacturing

Action Taking

Meetings with employees were held two weeksafter the initial meeting. Over the next threeweeks, the authors used the data to create aprocess diagram of ELCO’s operations thatindicated boundaries between each manager’soperating departments and exchange points wherecross-functional communication was required.Cross-functional communication is two-waycommunication where the impact of activitiesbetween two functions occurs. It is a specific formof interpersonal communication. Interpersonalcommunication can occur between two peoplewhere each reports what is going on in their ownfunction but the dialogue that links the two reportsdoes not occur.

A process map was created after two rounds ofrevisions based on feedback during project teammeetings (Figure 2). Interviews with the manage-

ment team were transcribed and examined by theauthors. A multi-view perspective of the manage-ment challenges and pressures facing the com-pany was created. The authors’ interpretation ofthe data was presented to the full research team atthe end of November 2001. The revised pre-sentation, taking into account the debate andclarifications given during the meeting, is illustratedin Figure 3 and details the individual perspectives.

Evaluating

At this stage, the management team reallyappreciated the organizational snapshot. Themanagement team was pleased to have theopportunity for open group discussion aboutELCO’s operations and agreed with the results ofthe analysis. In fact, it seemed to them that itshould have been obvious.



Observations – Individual Priorities

• Nick – Short and long-term objectives, time constraints, productivity

• Jerry – Bridging between objectives (planning) and tasks (operations)

• Diane – Supporting legacy IS, implementing new IS

• Bruce – Running a smooth production process, wading through too much data to do it

• Tim – Supporting information infrastructure and engineering data

Observations – Stated or Implied Goals

• Nick – STRATEGY – increase company value, technical expertise, and lateral growth

• Jerry – OPERATIONS – Ensure goals are met and nobody is left behind

• Diane – OPERATIONS – Provide more valuable support

• Bruce – OPERATIONS – Boost productivity and streamline planning

• Tim – OPERATIONS – Manage and control engineering data to create smoother flow

Observations – Information Use Within ELCO

Based on the process diagram and notes:• Most information and reports must be ‘pruned’ to

be useful, doesn’t always go to right people (steals time)

• Information from the ‘outside world’ doesn’t flow well inside ELCO (blind to the market)

• Information sharing and feedback is primarily ad hoc

• Management team has different planning horizons (communication problems?)

Analysis – Two Views

Information ViewWe need to capture and use

information for making better decisions regarding marketing, engineering, production, etc.

Implied by Nick

Infrastructure ViewWe need to specify and

implement an IS so that we can make decisions regarding marketing, engineering, production, etc.

Implied by Dianne and Tim

Bridging View - We need bothImplied by Jerry and Bruce

Analysis – What May Be Happening

Long-term ViewCollect and leverage proper

information to make better decisions leading to higher profitability

Implied by Information View

Short-term ViewGet the infrastructure built and

start using it (risk is that haste may lead to poor planning)

Implied by Infrastructure View

Bridging View – What resources and planning do we need to join views?

Summary – Go Forward Questions

1. What is the most effective way to link short and long-term views?

2. Given change, time, and financial constraints, what does it make sense to focus on first? (i.e. entire infrastructure, marketing, engineering, production?)

3. How much change are we willing to attempt? What are we not willing to change?

4. How do we merge our planning perspectives?

Observations – Individual Priorities

• Nick – Short and long-term objectives, time constraints, productivity

• Jerry – Bridging between objectives (planning) and tasks (operations)

• Diane – Supporting legacy IS, implementing new IS

• Bruce – Running a smooth production process, wading through too much data to do it

• Tim – Supporting information infrastructure and engineering data

Observations – Stated or Implied Goals

• Nick – STRATEGY – increase company value, technical expertise, and lateral growth

• Jerry – OPERATIONS – Ensure goals are met and nobody is left behind

• Diane – OPERATIONS – Provide more valuable support

• Bruce – OPERATIONS – Boost productivity and streamline planning

• Tim – OPERATIONS – Manage and control engineering data to create smoother flow

Observations – Information Use Within ELCO

Based on the process diagram and notes:• Most information and reports must be ‘pruned’ to

be useful, doesn’t always go to right people (steals time)

• Information from the ‘outside world’ doesn’t flow well inside ELCO (blind to the market)

• Information sharing and feedback is primarily ad hoc

• Management team has different planning horizons (communication problems?)

Analysis – Two Views

Information ViewWe need to capture and use

information for making better decisions regarding marketing, engineering, production, etc.

Implied by Nick

Infrastructure ViewWe need to specify and

implement an IS so that we can make decisions regarding marketing, engineering, production, etc.

Implied by Dianne and Tim

Bridging View - We need bothImplied by Jerry and Bruce

Analysis – What May Be Happening

Long-term ViewCollect and leverage proper

information to make better decisions leading to higher profitability

Implied by Information View

Short-term ViewGet the infrastructure built and

start using it (risk is that haste may lead to poor planning)

Implied by Infrastructure View

Bridging View – What resources and planning do we need to join views?

Summary – Go Forward Questions

1. What is the most effective way to link short and long-term views?

2. Given change, time, and financial constraints, what does it make sense to focus on first? (i.e. entire infrastructure, marketing, engineering, production?)

3. How much change are we willing to attempt? What are we not willing to change?

4. How do we merge our planning perspectives?

Figure 3. Baseline Analysis Presentation, November 21, 2001



Well, I’ll tell you, it really hits home atwhat I think has been a difficult point forus. We need to find some way tocommunicate more if we are going to getanywhere. When the different perspec-tives are laid out like that, yeah, I can seewhy problems can happen. (Nick, duringthe analysis presentation)

From the interviews, it was clear that the manage-ment team felt burdened by the need to balanceday-to-day operations and planning activities.

Well, I guess you prioritize [your plans forthe future], and it’s unfortunate becauseso many times the urgent takes over andthe best gets left behind. It’s the “tyrannyof the urgent.” It’s like that. We fight thatall the time. What is the best [action totake first]? Sometimes you just don’thave the time [to plan ahead] becauseyou feel this urgent need to geteverything done at the end, the day-to-day stuff, too. (Diane, remarking on thedifficult reality of medium to long-termplanning at ELCO)

Relevant issues became apparent while the pro-cess diagram was being revised. The mostnotable was a difference in each manager’splanning horizon. A second issue was the appar-ent lack of effective communication betweenmanagers.

[As far as what is happening outside ofengineering] I would be interested in…production scheduling [information], asan example. If we’ve got a componentproblem, which is affecting the continua-tion of production…how soon do theyneed a technical problem resolved beforeit impacts production? I don’t need toknow how well production is doing interms of getting product out the door.Where it would impact me is if they havea problem which I have got to solve, andknowing what their schedule is [before-

hand], so I know how much time I haveto resolve a problem before it impactsproduction [is vital for planning in theEngineering department]. (Tim, ex-plaining why cross-functional informationis crucial in medium- to long-termplanning)

Not surprisingly, differences between individualmanagement perspectives corresponded withtrouble areas identified in the process map, suchas the barrier between the engineering andmanufacturing planning functions.

It appeared everyone was in agreement on theobservations, although some team membersquestioned a few of the conclusions.

I…was a little surprised about the com-munication problem to a certain extent.When [it was suggested] the manage-ment team had different priorities… withour different personality types and jobswe do, that is to be expected.…I guess Iwas a bit surprised when I saw…differentpeople were looking at long-term ratherthan short-term or short-term rather thanlong-term.…I never really felt there wasa big communication problem in themanagement team. Maybe we don’talways listen really well and hear whatthe other one is saying, but most thingsare discussed. (Diane, commenting afterthe baseline analysis was presented)

Specifying Learning

One thing was clear. Each manager was spendingtoo much time on collecting and working withreports and figures from multiple sources. Thiswas getting in the way of their operating duties,something with which their colleagues used tohelp. The managers concurred that decreasedcommunication caused this problem, which in turnwas caused by increased time and planningpressures. This was the first direct indication that



a significant source of difficulties might be due toa loss of internal transparency; the idea that themanagers were losing an awareness of what washappening in the functional departments aroundthem. It was at this time that the authors began todevelop and apply the concept of internaltransparency.

Diagnosing

At the end of the first stage, the authors believedthat the decision to grow resulted in an initialincrease in the amount of planning activities. Inturn, this reduced the amount of time spent incommunication, which reduced the organization’sinternal transparency. It seemed that the loss ofcommunication represented a problem for planningactivities, and placed an additional burden on thecompany’s ability to move beyond day-to-dayreactive management. We suspected that therewas a relationship between the different planninghorizons and the reduction in internal transparencybut were uncertain what that was.

Stage 2: Strategy Planning

When the research project was initially planned, itwas expected that this stage would focusexclusively on IS planning to support growth.However, it was clear that ELCO’s managers werestruggling to maintain their own functions, muchless grow the organization. Based on theobserved planning horizon differences, the authorssuggested, and the team adopted, a plan toarticulate a common business strategy that, inturn, might be used to identify relevant IS needs.

For ELCO’s managers, the implementation of anIS was viewed as much less important. They hadcome to see that defining their business prioritieswould need to be done before defining a new IS.Stage 1 had suggested that their problems werenot related directly to the current IS but ratherchanging managerial pressures. During Stage 2,the team focused on building a coherent strategyfor implementing a new IS that took into account

what was learned in the baseline analysis. Withrespect to the research question, we focused onidentifying potential causes for the reduction ininternal transparency and its relationship tobusiness planning capability.

Action Planning

To ensure that the business strategy sessionwould be completed and not fall victim tointerruptions, the authors suggested an off-siteplanning session to minimize interruptions. Theteam agreed to a one-day planning session at theQueen’s Executive Decision Centre (in Kingston,Canada, approximately a one hour drive fromELCO’s plant), using facilities specifically designedfor group decision-making and planning practicesunder the guidance of an experienced facilitator.A date in early December was chosen, and eachperson was provided with material that outlined asimple planning process.

Action Taking

The day-long session was split into two parts. Themorning section covered a SWOT (strengths,weaknesses, opportunities, threats) analysis. Theafternoon session built on the morning’s resultsand consisted of the team members (including theauthors) working alternately in pairs and as agroup to decide on the three top strategic initia-tives to be addressed in the next 18 months. Bythe end of the day, the team had reachedconsensus that the three top issues were marketdevelopment (developing new products andstronger market presence), operational manage-ment (filling in the middle ground by joining theshort-term and long-term views), and technologyinfrastructure (using IS to assist in regaininginternal transparency). Results from the sessionwere organized and distributed to the project teamin the month following the strategy session.Interviews were conducted at the end of Decemberto examine how well the management team felt theprocess went, and whether they thought they hadcreated realistic plans to address their concerns.



Evaluating

The research team members embraced the resultsof the planning session. Some had beenpreviously involved with strategy-planning exer-cises that had mixed results. In this case, how-ever, all felt a strong sense of accomplishment.

I was pleasantly surprised, because Ihave been to a lot of [planning andstrategy meetings] over the years, and Ifound in quite a number of them,unfortunately, you walk out with a lot ofgood ideas but not any plan on how youare going to do what the suggestionswere. (Jerry, shortly after the planningsession)

Even with the positive outlook, there was stillcaution about whether ELCO would be able tocarry through with the planned strategy.

This was a good start at strategicplanning for the purpose of setting theground floor requirements for a forwardmove. But, ELCO’s internal support sys-tems are not ready for…growth, meaningthe MIS system and even personnel. Alot of planning still [has] to be done.(Bruce, after the planning session)

There was relief that attention was being directedtoward problem areas, but it was tempered byconcern about whether ELCO could effectivelydeal with the problems.

There is a lot of work to move forwardwith a plan. If we don’t have the peopleor the support of an MIS system this isgoing to be a problem, just getting thework done. [If we develop our market]we will get more calls from [sales repre-sentatives] and customers concerningproduct performance issues, you know,they want to know more, so you need tosupport the customer more.…the listgoes on and on. (Bruce, commenting onthe challenges of implementing thestrategy planning results)

The forces creating a “tyranny of the urgent” wereevident in the management team’s skepticism thatchange was possible. However, no one ques-tioned that the correct problems had been targetedfor resolution.

Specifying Learning

This stage demonstrated a significant shift in theorganization. Previously, ELCO had not plannedits IS activities or attempted any form of alignmentto its business strategy. Here, the managementteam had articulated a business strategy (Priority#1) and then worked out the implications for ISplanning of those decisions. This clearly demon-strated an evolution in IS-business alignmentpractice from administrative integration to sequen-tial integration (Teo and King 1997). This wasimportant because it led to several insights aboutthe process by which the existing IS had becameunacceptable. The existing IS did not support thetype of information required to support eithermarket development (Priority #1) or operationalmanagement (Priority #2). This marked a signi-ficant change in IS planning. Until this time thereason for a new system was not identifiable, butnow it was apparent that it related to both long-term and medium-term needs.

Diagnosing

We become convinced that the different planninghorizons were not a cause of the reduction ininternal transparency but rather the opposite. Thereduction in internal transparency had led tooperational difficulties. Lower internal transpar-ency made the managers’ operational tasks harderto do, which meant that they had less time forplanning. When they had time, they did not havethe right information. Hence, two behaviorchanges were identified that weakened the existingIS structure. First, the introduction of new planningbehaviors necessitated new informationrequirements (e.g., easier access to customer andmarket information). Second, the reduction in timefor cross-functional communication caused,through a reduction of internal transparency, a



greater need for accessible and structured infor-mation than had previously been communicatedand contextualized by other managers.

Stage 3: Requirements Specification

The final stage of the project translated thestrategic plan into a request for quotation docu-ment. In this stage, the work distribution of theteam became very different. Two people (oneauthor and one manager) focused on developingthe RFQ while the remaining team membersreacted to document drafts with comments andsuggestions. We believed that the researchquestion would be addressed through identificationand validation of the types of issues that wouldassist in defining the new IS.

Action Planning

At this stage, the project transitioned fromidentifying a course of action during the strategicplanning stage to acting on a particular initiative, inthis case preparing the RFQ. From the authors’experience, two months seemed to be a reason-able time frame for this activity and coincided withthe formal end of the six-month project. However,the six-month time period could have beenextended with the agreement of the fundingagency (the agency would have readily agreed toa one or two month extension) and as such thedeadline was not the motivation behind the two-month time frame.

Action Taking

Diane and one of the authors developed the RFQitself and worked together to define the IS speci-fications required for the RFQ document.Requirements analysis consisted of a variety ofdata-gathering activities such as cataloguing andaugmenting the IS specifications from existingsystems in use, requirements-oriented analysis ofthe strategy-planning exercise, and personalinterviews with key IS users. Of particular interest

to the RFQ authors was the communicationbetween functional departments in the company,referred to as the information exchanges betweenfunctions. The requirements-gathering activitieswere conducted at the ELCO site and lasted twodays. All project team members contributed theirsuggestions and changes during three revisioncycles over the following four weeks. By the endof this process a final version of the RFQdocument was agreed upon and completed.ELCO received the RFQ and a copy was providedfor the government support agency, formallyending the practice element of the project.

Evaluating

This stage resulted in an RFQ calling for threedistinct systems (materials requirements planningor MRP, marketing management, and projectmanagement) with the capability to communicateand share information between them. The newMRP system would replace the outdated mini-computer while the marketing management andproject management systems represented entirelynew IS for ELCO. While the decision to followthrough with three separate systems rather thanone single system had been the topic of con-siderable debate, the final decision rested on therealities of operating a small business. Thepurchase cost of an integrated, multifunctionalsystem would be quite high, and the projectanalysis to this point indicated there was not asgreat a need to integrate data between systems asthere was to have efficient, open but controlled,access to company data. In the end, a decision toinvest in three subsystems was seen as aneffective compromise between functionality andcost, especially given the uncertainty aboutELCO’s future state. In Jerry’s words,

Well, we have to be practical, right? Wehave to ask ourselves, what is ourbiggest need? But at the same time Ihave to keep an eye on cash flow. Nowsure, there are times when Bruce couldreally [benefit from] having his [manufac-turing system] automatically know what ishappening in Sales, or Engineering, or



whatever, but right now my biggest worryis the time it takes to do scheduling, andright now our [computer] system just isn’tup to the task.…Now, we’ve been talkingabout sharing our departmental informa-tion more, and I am all for that so long aswe need it and act on it…[but] in the endwe have to look at what we can afford.(Jerry, during requirements analysis)

Specifying Learning

Initially, the results of this final stage of theresearch project did not appear to have mucheffect on the research team members’ view of howtechnology could assist in ELCO’s operations.The management team maintained an under-standing following the baseline analysis thattechnology was not going to make their challengesdisappear automatically. Success would hinge onhow individuals integrated IS into their operatingactivities.

If we don’t understand what it is we aretrying to do, putting a system in placeisn’t going to fix it. I think putting a sys-tem in could help a lot, but only if you doexcellent training. Otherwise, you aregoing to get garbage-in, garbage-out.You have to understand what it is youare trying to do in terms of moving workthrough a process. (Nick, during the lastround of RFQ revision)

However, one subtle change was apparent in theway the management team discussed the use ofIS. Formalizing the project’s results into a speci-fication document for new technology systemsrequired everyone to be very explicit in what theyexpected IS to accomplish for ELCO, and thedegree to which they had confidence in specifyingELCO’s future needs. The RFQ revision processrequired them to share and discuss their cross-functional plans much more than had been done inthe past year. The managers rarely madedistinctions at the beginning of the project betweentheir interpersonal and cross-functionalcommunication habits or patterns. For example,

when asked about how well and how often themanagement team discussed plans and issues,the typical response was

Well, we have a management meetingusually once a week. These meetingsare really just the formal who’s doingwhat, what’s coming up discussions, butwe also see each other in the building allthe time.…I mean, there are only 32people here; you can’t help but talk toeach other. I would say we communicatequite well. (Diane, during the baselineanalysis stage)

However, at the same time, it was also clear therewere particular times when effective communi-cation was not occurring:.

I know there have been times whenthings have been missed, deadlinesextended, and sometimes it turns out thatthere was an [engineering report] orsales order misplaced that really affects[operations for] Bruce or Tim.…wecertainly talk to each other on a dailybasis, but when it comes to knowingwhat each person’s department is doing,we don’t always have time to talk aboutthe details like that. (Diane, during thebaseline analysis stage)

It seemed that the strains of reacting to Nick’soriginal decision to jump-start ELCO’s growth—the punctuation—caused significant behaviorchanges that lead to weakening of the existing IS.

Diagnosing

As the RFQ developed it became clear to the coreresearch team that there were two types ofimportant communication patterns going on insideELCO. There was both an interpersonal as well asa cross-functional communication process, andwhile the interpersonal communication was stilleffective, the cross-functional communication wasnot. Through deliberations, the new systems werenot only seen as being able to provide the



operational capabilities required to organize andplan the company’s operations, they were alsoseen as a way in which to regain the level ofinternal transparency that was lacking.

Nick: “When we talk about the projectmanagement system [for example]…Ithink we agree that [expecting thesystem to give us successful projects] isnot valid. I guess what we need to do isvalidate the need for the project manage-ment system…”

Author: “I’ll tell you what I think theproject management system…has tooffer [ELCO]. The best capability it willoffer is [when the other managers use it]to see how the product developmentprocess is progressing, increasing thetransparency of the engineering depart-ment to everybody else.”

Nick: “Totally agree. Particularly [for]me.” (Nick and an author, during theRFQ revision process)

They also realized that an ERP was not a feasiblesolution, as it was not possible to develop a visionof an integrated system when there were manyalternative views of ELCO’s future—it couldcontract, grow a little, grow significantly, or stay thesame. In the end, the RFQ specified a solutionthat no one believed would be the perfect long-term solution for ELCO but rather defined an ISthat would allow ELCO to regain its internaltransparency and provide some basic planninginformation while respecting cash flow and otherresource requirements.

Going through the RFQ generation process forcedELCO’s management team to apply this newunderstanding in a way that introduced them to anew way of thinking about how they sharedinformation. They believed that IS should becapable of assisting their operation by enabling ahigher degree of cross-functional communicationbut still in the informal manner in which they werecomfortable and successful.

Summary

This project delivered its goals on two dimensions.First, the project team was able to develop an RFQthat satisfied ELCO’s requirements. Second, theauthors found results that highlighted howplanning, communication, operating behaviors,and an IS in a small business are interdependentand vulnerable to exogenous punctuations to theorganization. We also found it interesting how ournascent theoretical perspectives influenced thenext action research stage and how each stagerefined our developing perspectives. Indeed, wewere surprised in the last stage (we thought mostdiscovery would occur in the earlier PAR cycles),when the importance of the distinction betweencross-functional and interpersonal communicationemerged.

A Punctuated EquilibriumModel

The decision to grow can be seen as a punctuationto the small business. In the following sections,the sequence of behavioral changes that results ina weakening of the deep structure of the existingIS and causes poor operating and planningoutcomes is demonstrated. This model isdeveloped from the diagnosing sections of theaction research cycles. We want to be cautious inproclaiming this as a general model but believeproposing it serves two purposes. First, otherresearchers can look for concurrence betweentheir findings and ours. Second, there are manysmall businesses that could be seen to be similarto ELCO. Not every small business wouldexperience this dynamic process the same way,but it would seem reasonable that there would beothers that would.

An interesting result of this research project wasthat the IS became viewed as obsolete almostimmediately upon the launch of the growth stra-tegy before any organizational or structuralcharacteristic usually associated with growthchanged. The model in Figure 4 describes our=



Beha

vior

Cha

nges Planning

Operating

Communicating

Increase in time spent (desired activity)

Decrease due to increase in time spent planning

Cross-functional information is not communicated; Internal Transparency reduces

Time required increases due to lower Internal Transparency

Time decreases due to increased operating time

Change system to accommodate immediate shortcomings

Weakens due to need for information that used to come from informal communication

Weakens due to new need for planning related information


Deep Structure: Information Systems

Punctuation:Decision to Grow

Cross-

Beha

vior

Cha

nges Planning

Operating

Communicating

Increase in time spent (desired activity)

Decrease due to increase in time spent planning

Cross-functional information is not communicated; Internal Transparency reduces

Time required increases due to lower Internal Transparency

Time decreases due to increased operating time

Change system to accommodate immediate shortcomings

Weakens due to need for information that used to come from informal communication

Weakens due to new need for planning related information


Deep Structure: Information Systems

Punctuation:Decision to Grow

Cross-

Figure 4. Punctuated Equilibrium Model

interpretation of why this happened. Four man-agerial responses occur sequentially to thepunctuation: an increase in time planning (asdesired), leading to a decrease in time com-municating cross-functional information, followedby an increase in time spent managing operationsand finally, and contrary to the initial response, adecrease in the amount of time spent planning.

Response #1: Increased TimeSpent Planning

Planning behavior increases in direct response tothe punctuation, the stated goal of growing. Inorder to prepare ELCO for growth, the managersneeded to increase the amount of time spentplanning. As part of these new tasks, the existingIS structure weakens as different information thannormally provided through the IS is required by themanagers. Another implication of the increasedtime spent in planning (and in some casestraveling to see potential clients) is that the seniormanagement team had less time to communicatecross-functional information.

Response #2: Decreased Time SpentCommunicating Cross-FunctionalInformation

As more time was spent planning, ELCO’smanagement decreased the amount of time thatthey spent communicating cross-functional infor-mation. This was seen as a subtle change but itturned out to be one with significant effects. It oc-curred as an instance of the managerial capacityproblem (Jensen and Meckling 1976; Oi 1983)wherein the addition of new planning tasks re-quires other tasks to be removed from activities—in this case, cross-functional communication. Theinformal networks that communicated cross-functional information and maintained internaltransparency disappeared. This spoke clearly tothe fragility and necessity of the social context ofan information system in a small business, muchin line with the work of Brown and Duguid (2000)amongst others. ELCO’s management team washighly interdependent; each individual’s actionsinfluenced the planning actions of the others. Thesocial context had provided the necessaryinformation in the past, but changes to that context



had corresponding negative effects oncommunication practices. This is in concordancewith the sociotechnical nature of IS developmentand implementation as supported by Mumford andher colleagues (reported in Mumford 2001). Notonly is development a sociotechnical phenom-enon, maintenance of existing systems wouldappear to be so too. Extending on the existingliterature, we also found that the effects areasymmetric in that changes in the social contextaffected cross-functional information before inter-personal—managers continued to talk but notabout what they were doing that affected othermanagers. This might be partly explained if cross-functional communication (how my activities affectyours) were reported through informal channelsand other information (reporting on my activities)were reported at meetings that were maintained.

In retrospect, the symptoms of internal trans-parency loss during the transition period werequite evident. What is interesting is why no onesaw it at the time or at least realized it was aproblem. Even after the potential problem areawas identified, however (Nick, for example, didthink that an improved information system mighthelp), doubts still existed whether the root causereally did relate to management communicationpractices, as evidenced by Diane’s early doubtsabout whether communication problems actuallyexisted. While the baseline analysis suggestedcommunication problems, the communicationhabits within the management team were littledifferent than in the past. The difference was thatthe informal practices changed whereas the formalstructures did not change. Eventually, the socialsystem in which the IS worked changed to such adegree that the legacy system becameinadequate.

Response #3: Increased Time SpentManaging Operations

The legacy system became inadequate as theresults of reduced internal transparency becameapparent. It meant that managers had to spendtime gathering and analyzing information thatpreviously was provided by their colleagues. This

further weakened the existing IS by demandingnew ways of accessing existing information.Hansen (1995) foresaw this effect when he statedthat “any constraint on a small group’s internalcommunication structure seriously (interferes) withthe group’s ability to perform complex tasks.”Communication constraints were also considereda cause of lower levels of task accomplishment, aneffect later confirmed by others (e.g., Snyder andMorris 1984). Diane had perceptively picked up onthe very dynamic that was at the root of theproblem: while the interpersonal communication,which provided information about what eachperson was doing, was essentially unchanged, thecross-functional communication, the provision ofinformation regarding the actions and results thatwere being delegated, was sharply decreased.

I mean, there are only 32 people here;you can’t help but talk to each other. Iwould say we communicate quite well…

we certainly talk to each other on a dailybasis, but when it comes to knowingwhat each person’s department is doing,we don’t always have time to talk aboutthe details like that. (Diane, during thebaseline analysis stage)

Most published accounts of the operating charac-teristics of a small business depict an environmentwhere individuals are naturally aware of what isgoing on around them, what we called internaltransparency. We found that cross-functionalcommunication breakdowns resulted in the loss ofinternal transparency, which created operatingdifficulties. The managers were still communi-cating personal plans and developments but notwhat was going on around them. The existing IScould not make up the gap. The result was thatthe managers reduced the time dedicated toplanning.

Response #4: Decrease in Time forPlanning for Growth

As the managers reduced the time dedicated toplanning, the necessity to replace or enhance



ELCO’s IS became obvious. The link betweendecreased communication and planning difficultieshas been established by authors such as Bavelas(1951) and Leavitt (1962) but a mechanismexplaining the relationship to IS has not. Usingthis model, we suggest that planning difficultiesare, at least partially, directly attributable toreduced internal transparency. These effects weremanifested within ELCO as difficulty in planning forgrowth.

This behavior response was in direct opposition toNick’s stated goals, and contrary to Response #1,the original reaction to the punctuation. In spite ofthe short term increase in planning time, ELCO’sIS could not support the organization’s informationneeds and the time planning was returned tooperational management. Change of some sorthad become essential. ELCO’s IS changed inresponse to the punctuation. Before Nick pur-chased ELCO, its information needs were servicedby an IS that was neither modern nor up-to-date.It held together and was made functional throughthe communicative efforts of the managers. Alltechnology-based IS are a mix of social andtechnical components and this system was noexception. As Nick moved the company intogrowth mode, the social system changed; com-munication practices altered and the technicalflaws of the system became apparent. Formalmeetings continued to convey information but notthe cross-functional information requisite tomaintain internal transparency. At the start of thisproject, the system was viewed as inadequate andin need of replacement, even though ELCO’soperations had not changed in any obviousmanner. The managers blamed the IS for itsfailings when in reality it was they who hadchanged.

Deep Structures: Impact onExisting IS

In response, ELCO has defined a satisficingsolution. We call this a satisficing solution as itmeets ELCO’s current direct needs within budgetconstraints, but it is not necessarily the ideal long-

term solution. It was difficult to design the perfectsystem for the future due to uncertainties about thefuture state of ELCO’s business. Many firmswould be unable to define the ideal system at thisstage. Rather, they may need to be content tomake changes that compensate for the decreasein cross-functional communication and internaltransparency while minimizing expense andchange. Given that successful growth firms seemto put into place the foundations for the largerenterprise before they are absolutely essential butdo not overly formalize the organization too quickly(Hambrick and Crozier 1985), a satisficing solutionmay help navigate this delicate balance.

From this model, it is clear that the existing IS—adeep structure—is weakened very quickly bybehavioral changes in response to the punctua-tion. First, newly initiated planning activities mayrequire additional functionality. In turn, cross-functional communication likely needs to be sup-ported using more formal methods. Not compen-sating for these weaknesses may result innegative outcomes for operational and planningtasks. Therefore, our findings would suggest thatfirms embarking on growth need to considerlimited adaptation of their IS before starting to planfor future growth.

Discussion

IS processes that had adequately supported ELCOin the past silently failed when they began planningfor growth. This finding suggests that thetransition effects on IS may occur very early. Thisis, of course, troublesome because the organiza-tion might need to consider IS investments that itcan neither afford nor clearly specify. ELCO dealtwith these problems by adopting an approach thatadapted their infrastructure at the lowest cost andminimal degree of change.

The ELCO project illustrates three lessonsregarding transition pressures during small busi-ness growth, informal communication practices,and the internal transparency of an organization.First, increasing operational business pressures



such as managerial time constraints provokechanges in informal communication practices(Vinten 1999). Second, cross-functional communi-cation practices may be prone to difficulty duringtransitions. Third, these breakdowns lead todecreased internal transparency, and theaccompanying difficulties. Understanding andavoiding the negative effects that business growthcan create for informal communication practices insmall business is a necessary step to successfulsmall business development.

Research Implications

Internal transparency and its central role in smallbusiness growth and IS management is potentiallyan important concept that should be introducedinto new studies. While many authors movedaround similar ideas, this paper names andsituates it for the first time within the IS literature.It will be very interesting to see if, as we propose,decreasing internal transparency generally resultsin decreased operational and planning per-formances. Internal transparency is particularlysalient for small business and IS studies, as thesocial structures upon which many small busi-nesses rely to maintain their internal transparencycome under pressure through growth due tomanagerial capacity constraints (Barringer et al.1998). It should also be noted that internaltransparency is something that it is hard to realizeis missing (if you knew it was missing, you wouldprobably try to take action to restore it, as ELCOdid). In this situation, the introduction of theauthors as action researchers allowed the injectionof a perspective that demonstrated the loss ofinternal transparency.

As small business growth tests the capabilities ofinformal communication channels, formal systemsbecome more essential to fulfill the informationneeds of an increasingly complex organization.The expected role of IS in improving communi-cation within a small business is similar to theresults seen from other authors (Levy and Powell1998; Shin 1999). This paper makes a distin-guishing contribution in identifying that the poten-

tial need for IS change occurs very early, beforemany other structural or organizational changesare required. Intention to grow seems to be thefirst critical point, not actual growth. This is in linewith general findings that small business man-agers should prepare organizational structures forthe larger organization before they are absolutelynecessary (Barringer et al. 1998; Churchill andLewis 1983). Our findings show that this is alsotrue for the IS and true at a very early stage. Thissuggests a process by which IS and businessstrategy alignment (e.g., Sabherwal et al. 2001) isundermined very early on through the changes inmanagerial behavior. The point in time at whichthis change occurs indicates when the need forrealignment planning occurs. Echoing the resultsof Sabherwal et al., who described recurringpatterns of strategic realignment that took placeover a long period of time, we also came tounderstand that ELCO’s challenges actually beganto take root at a point when the company stillappeared relatively stable.

As shown in the literature review, small businesstechnology adoption research often theorizes theowner/president as an individual decision maker.However, in our case, while Nick played animportant role, the management team as a wholemade decisions regarding technology and whattechnologies specifically to adopt. Whether this isa change from previous findings that emphasizethe role of the CEO (DeLone 1988) or anotherexample of a diminishing influence (Harrison et al.1997), it raises an interesting question about theappropriate unit of analysis for these types ofdecisions. Group decision making frameworksmay be more appropriate as presidents/ownersrealize that buy-in is important for technology-based change projects (Kotter 1996).

Deciding on the types of information best suited tobe shared through IS depends on an examinationof the networks within the small business. Theinformal communication practices within a smallbusiness can be broken into two constituent parts:the interpersonal and the cross-functional. Thesetwo base types are affected differently during thetransition process. We found interpersonal com-



munication at the top management level, thesharing of information regarding each manager’spersonal responsibilities and immediate actions, tobe robust in the transition periods. Small businessmanagers can be expected to discuss theiractivities, given that they work closely andfrequently together. On the other hand, cross-functional communication at the top-managementlevel, the sharing of information regarding thestatus and future plans of each manager’s func-tional responsibilities or departments, was not asrobust. In transition phases, sharing of informationthat is not immediately at issue often does notoccur. It is this cross-functional information, theinformation contained in manufacturing and engi-neering schedules, sales forecasts, or accountingreports, that may fail to be shared during transi-tions. This cross-functional information is wellsuited to being maintained using IS. Further,interpersonal communication technologies such asinstant messaging, e-mail, or bulletin-board stylesystems are probably not as helpful to the smallbusiness during growth stage transitions relative tothe cross-functional systems. Within our study, themanagers did not bemoan the fact that it wasdifficult to send e-mail to each other (they all hadto use the same terminal to send e-mail) but didworry about cross-functional information. As seenin Stage 3, it may be that the most desired aspectof IS functionality includes the aggregation anddissemination of information between areas ofresponsibility.

Managerial Implications

A small business contemplating growth must beaware of the effects of transition phases oninformal communication and internal transparencywithin the company. This research speaks clearlyto the need to develop procedures to ensure thatInternal Transparency does not unduly suffer andbecome a significant problem. One approach is toplan a transition to include more formal practices,which might include increased IS usage, toexchange cross-functional information.

Evaluation of the capabilities of the organization’sIS is required. In addition to basic communication

support, critical evaluations of proper managerialaccess and system security are important. Twooverriding questions at this stage should bewhether the IS is capable of capturing andproviding the correct management informationneeded to provide internal transparency to topmanagement in a timely manner, and are peopleusing it?

Finally, there is another awareness issue; anunderstanding that the communication practices ofthe post-transition small company will frequentlychange again during the next transition period.Using internal transparency as a guide, modifi-cations to a business’s structural and organiza-tional characteristics should require correspondingmodifications to communication practices. Just asthe structural and organizational characteristicsthat create success in one stage must change forthe next, fundamental changes in communicationpractices must occur as well.

Action Research

Action research can be an exhilarating experiencefor the researchers and useful for the practitioners.However, the research outputs must have abroader interest and theoretical significance if thework is to be truly differentiated from, as manycritics characterize it, consulting. While reflectingon why a reader should believe that our work trulyembodies the principles of action research, werecalled Mumford’s (2001) succinctly stated threeproblems: getting in, staying in, getting out.

Getting In: Our introduction to ELCO wasfacilitated by the funding officer. She was able toensure that researcher and practitioner came tothe project with an appreciation of each other’sneeds. Rapoport’s (1970) initiative dilemma neverbecame a substantial concern. We also devel-oped a document that defined expectationssurrounding the project. Given that many acade-mics criticize practically focused funding agencies,our experiences may show that an effectivefunding officer, with a true appreciation for the wayin which research and practice can inform eachother, can be a great aid in the action research



process. While Nick invited us in, we had to buildrelationships with ELCO’s managers, which we didthrough individual meetings and a gradual ap-proach to getting to know them. In some otherforms of research, researchers come in for oneday, try and capture everything and never have theopportunity to develop an empathetic andunderstanding relationship. Action researchallowed, and demanded, that we take the time todevelop an effective working relationship.

Staying In: ELCO’s management team wasmutually supportive and had a good workingrelationship. While the organization’s IS was beingcriticized, the criticism was not directed at thecontroller, Diane. Indeed, Diane was viewed ashighly competent for being able to keep the systemrunning. Therefore, there really was no bad guy.We were fortunate in that we did not have toencounter a situation where we were in the middlebetween competing factions. The researcherswere never seen as problems, most likely becausethe project was moving forward and achievingmutually agreed goals on time. The managerswere comfortable with why activities were takingplace, a benefit of using PAR.

Getting Out: Our formal involvement with ELCOended when the RFQ was completed. One mightargue that we left too early to adequately judge theefficacy of the action research process, but Nickfelt that he was able to manage the acquisitionprocess from that point onward. As a postscript tothis project, the first author was invited back toELCO in February 2003, to take part in an ISselection meeting (the second author had moveda significant distance away). In the interveningyear between the end of this project and the finalselection meeting, Dianne and Bruce released theRFQ for vendor responses, accepted proposals,evaluated eight of them, and short-listed three forfurther consideration. By this time, the decisionwas down to two competing products. Bothpackages incorporated an integrated manufac-turing planning system with a marketing manage-ment system. A decision to not include a projectmanagement system was made with the intentionof revisiting it in the future. At the end of thismeeting, Dianne and Bruce were directed to

perform a final check on the references of previouscompanies who had purchased one particularsystem, and, if satisfactory, to engage the vendorand pursue purchase.

While the final recommendation, three separatesystems, was not followed exactly, it was ELCO’sresponsibility to make the final decision. Duringthe final selection meeting in February 2003, theproject team discussed once more how importantit was to recognize they were an interdependentfunctional team, how one person’s actions signifi-cantly affected everyone else, and why trans-parency in their actions was crucial as they grew.This discussion came much more easily to thegroup than it had a year and a half previously.Considering the progress they made on their ownin the year between the end of the project and thefinal selection meeting, understanding and inter-nalizing internal transparency made a significantdifference in their company. In retrospect, thegreatest achievement of the project may havebeen the understanding they gained of internaltransparency, and how their IS could enable it, butnot create it.

In addition to Mumford’s comments, we reviewedthe action research literature in general and thePAR literature in particular and established severalcommon themes of desirable PAR characteristics.A well-known framework by Lincoln and Guba(1985) introduces four criteria for qualitativeresearch: credibility, fittingness, auditability, andconfirmability. In Table 1, we have situated com-mon action research criteria within this frameworkto provide a broader perspective. Further, wehave elaborated on criteria commonly referred toin the IS action research literature in the followingparagraphs.

Genuine Problem: Many authors have stated theimportance of solving a genuine problem throughaction research (e.g., Baskerville and Wood-Harper 1996; Kock 1997; Lau 1999). Eden andHuxham (1996) identified a common theme ofaction research that the “research output resultsfrom an involvement with members of an organi-zation over a matter which is of genuine concern.”Nick defined our practice problem before he metus. Indeed, the funding officer knew to bring the



Table 1. Evaluating the Participatory Action Research ProcessCriteria for QualitativeResearch (Lincoln and

Guba 1985) Desirable AR Characteristics Examples From This Study

Credibility: faithfuldescriptions or interpreta-tions of a human exper-ience that the peoplehaving that experiencewould immediatelyrecognize it

The observations are recorded andanalyzed in an interpretive frame

Critical reflection on how data weresocially constructed throughresearch team interaction (Kleinand Myers 1999)

Stakeholder viewpoints are com-pared and contrasted (Stringer1999)

Researchers continuallydeveloped and presentedindividual views and prompteddiscussions to resolve views andcreate understanding

Diane’s apparently conflictingremarks regarding communi-cation behaviors was highlightedthrough interpretive frame

The research should be set in amultivariate social situation

Credibility is established throughtriangulation of information frommultiple data sources (Stringer1999)

Sufficient data must be collected toprovide rich, deep insight (Myers1999)

Research was conductedentirely on-site, involving thecomplete management team andtwo researchers

Data collection includedtranscribed interviews, officialand unofficial company docu-ments, e-mails, observationnotes, planning session results,over 500 pages of data

Changes in the social setting areanalyzed

Participants’ roles are expected tochange over time (Lau 1999)

Research participant rolesnoticeably changed duringproject

Fittingness: findings fitinto contexts outside thestudy situation

The research should illuminate atheoretical framework that explainshow the actions led to thefavorable outcome

Details regarding the changes thatoccurred should be related togeneral concepts to explain thenature of the changes and whythey might have occurred (Kleinand Myers 1999)

Purpose of research is to extendunderstanding of an issue2

Planning actions became easieras communication practicesimproved, which led to theinsights about internaltransparency

A general PE model wasproposed



Table 1. Evaluating the Participatory Action Research Process (Continued)Criteria for QualitativeResearch (Lincoln and

Guba 1985) Desirable AR Characteristics Examples From This Study

Auditability: whenanother researcher canclearly follow the “decisiontrail” used by theinvestigator

Future researchers can clearlyfollow the “decision trail” used bythe investigator (Lau 1999)

The methodology and resultssections provide a reliable audittrail regarding: how the site wasselected; how the projectparameters were decided upon;what data was collected, when,and by whom; project duration;and the interpretive data analysismethod followed

Confirmability:Engagement with thethings to be known issought in the interests oftruth

Direct research participant actionsintervene in the research setting(Baskerville and Wood-Harper1998)

Principle of interaction betweenresearchers and subjects (Kleinand Myers 1999)

Researchers seek to engageprincipal stakeholders actively(Stringer 1999)

Changes were introducedthroughout the project to testtheir effect and usefulness (e.g.,staff members defined their owninformation requirement needs,research team used results fromeach stage in later phases

Data collection includesparticipatory observation

Researchers develop researchcontext by observing the actualenvironment and the actors withinit as they live and work (Stringer1999)

Researchers kept personalproject logs and discussed groupinteractions

Researchers verified commentsand remarks, checking that whatwas said was what was actuallyoccurring

practitioners and academics together becauseboth had articulated similar interests. We believethat this project satisfies the criterion of solving agenuine problem.

Rich Data Collection: In order to understanddeeply the process under study, researchers needto take care when recording data and observations(Baskerville and Wood-Harper 1996; Singer 1999).To this end, research was conducted entirely on-

site, involving the complete management team andtwo researchers. Data collection includedtranscribed interviews, official and unofficialcompany documents, e-mails, observation notes,and planning session results—over 500 pages ofdata. We also discussed the data collection andsummary methods with the practitioners. Forexample, the presentation provided in Figure 3 isthe product of collaborative data aggregation,based on records kept by the researchers.



Genuine Collaboration: The importance of trueinteraction and dialogue between researcher andpractitioner is essential to most forms of qualitativeresearch (Klein and Myers 1999) and actionresearch is not an exception (Singer 1999). Wewere struck by the amount of time spentdiscussing what was going on in the project as weattempted to understand ELCO and its managers.Recognizing our ethical research responsibilitieswhile using an action research approach left uscareful not to ascribe motivations and beliefs tooquickly. Our role as committed and active parti-cipants—we wanted to see ELCO succeed, not forfinancial reasons but because we felt an affinitywith their goals—contributed to our focus onunderstanding and the collaborative experience.

This research project would not have beenpossible to do using any other methodology. Thefunding agency insisted that the company andresearchers work together in order to maximize theopportunities for knowledge transfer between thetwo groups. Also, Nick would not have beenwilling to commit ELCO to this project without theactive involvement of the authors in problemsolving. The action research methodology pro-vided us with guidelines and criteria that allowedus to participate in the project and address ourresearch questions. Indeed, as Baskerville andWood-Harper (1996) suggested, the ELCOmanagement were pleased to be an interestingresearch site.

Research Versus Consulting: While a commoncriticism of action research is its potential to be“consulting masquerading as research” (as notedby Baskerville and Wood-Harper [1996] amongothers), in this situation action research wasviewed by Nick as different than consulting. Recallthat Nick specifically wanted to avoid consultants.When action research was explained to him, hefelt comfortable that this method allowed him to doso. He never seemed to see us as consultants, afeeling perhaps helped by the fact that we were notsubmitting invoices and he was not writing checks.

More importantly, we hope that the concept ofinternal transparency and its relationship to smallbusiness growth, IS, and managerial behaviors will

be seen as research contributions. As discussedin the research implications, our findings shouldhave implications well beyond ELCO. We believethat our findings would not have been possiblewithout the interplay between theory and practicedemonstrated in this project.

Time Pressures: As our project was structuredwithin a specified time period, this might potentiallyharm the research integrity of the work (Baskervilleand Wood-Harper 1996). To this we can only saythat we did not find ourselves driven to meet thetimeline or that we sacrificed addressing theresearch questions to expediency, but were able tostay true to our goals (Seashore 1976) with thesupport of our industry partners who were alsointerested in our success. Further, while the RFQwas delivered at the end of the six- month project,the research perspective has continued to bedeveloped over the past year and through theprocess of developing this manuscript.

Practitioner Involvement in Research: Parti-cipatory action research differs from canonicalaction research or action learning in that there is“the notion that some members of the organizationbeing studied should actively participate in theresearch process rather than just be the subjectsof it” (Whyte 1991, quoted in Eden and Huxham1996). From Eden and Huxham, “this suggests atwo-way relationship; the research becomesinvolved in and contributes to the practitioner’sworld, and the practitioner becomes involved inand contributes directly to the form of the researchoutput.”

There were times when the authors were clearlyidentified as being in the practitioner’s world, suchas when we discussed deliverables, businessneeds, and constraints with Nick, when weproposed and took part in the strategy planningsession, and during the RFQ-creation process. Atthese times, it would have been difficult for anuninformed observer to tell that we were doingresearch.

There were also times when the entire ELCOmanagement team was clearly in the researchworld, such as when we discussed the small



business experience, growth, and internal trans-parency during interview sessions. The ELCOmanagers were active in interpreting and makingsense of what was going on at the conceptuallevel. They commented directly on our developingresearch perspectives. Indeed for the last twocycles, they had to ask questions in order tounderstand the reasoning behind our interventions.Again, at these times, an uninformed observerwould have believed that the management teamwas participating as researchers only.

As researchers and practitioners were bothinvolved in considering research and practiceproblems, the entire team was concerned withboth sides. Further, the cyclical nature of actionresearch meant that in order to take appropriateactions, the theory development was important tothe entire team. The authors certainly spent moretime on theory development, but ELCO’smanagers clearly contributed to the developmentprocess. Our perspectives and developing theorychanged in response to the reactions of ELCO’smanagement, a flexibility that we view as astrength of the method, as did Davison and Vogel(2000).

Limitations

While we believe the evidence for our claims isstrong, it is difficult to prove the absence of bias.As action research is essentially a problem-discovery or problem-solving approach where theparticipants continually gather, analyze, and refinetheir understanding of an issue, there is thepossibility that had the research environment beendifferent, the results would have significantlychanged as well. Nevertheless, as shown in theresearch site description, ELCO was a fairly typicalsmall manufacturing firm. The authors wereclassically trained in systems analysis and designtechniques and had worked in design projectspreviously and could be seen as representative.

The Hawthorne effect (Roethlisberger and Dickson1939) is a well-known limitation in action researchstudies (Baskerville and Stage 1996). The

limitations imposed by the Hawthorne effect werethat the problem itself, communication and internaltransparency issues, and their effects on planningand growth fundamentally changed as theresearch group identified and focused theirattention on it. It cannot be ruled out that theissues addressed in this research project were inpart created or modified by the researchparticipants themselves during the project. Thisvalidity threat is also fundamental to the actionresearch methodology; however, the essentialaction is to introduce changes to the environmentthroughout the exercise. In any event, the resultsrepresent the resolution strategies, the keyactions, and the outcomes of those actions asexperienced by the actual people living andworking in the research environment.

Finally, as much of our reported data and con-clusions are based on our working relationship withELCO’s managers, there is the possibility that wesimply did not get the true story from one or moreparticipants. However, while there may have beensome suspicion or concern about motives at thestart of the project, we felt that the team becameopen and honest very quickly. For example, wewere conscious of what Rapoport (1970) termedthe rejection phenomenon, a negative practitionerreaction to our intervention, but did not see it andas such believe that we obtained the managers’real perspectives. Additionally, the data wereinterpreted and acted upon by the research team.The study is, therefore, limited by our biases andactions.

Conclusion

Communication practices within a successful smallbusiness normally provide the required levels ofinternal transparency during stable periods.However, from the earliest part of a transitionperiod, communication methods change. Failureto properly adapt showed itself as a crisis ofplanning at ELCO. A decrease in internaltransparency is often itself caused by decreases inthe cross-functional communication that arecaused by increased managerial pressures. Small



business management often finds itself in acircular holding pattern as day-to-day operationsand transition planning consume more and more oftheir personal time and attention, leaving less andless time to inform their management colleaguesof what is happening in their part of the company.This tyranny of the urgent can have drasticallynegative effects on the Internal Transparency ofthe organization, ultimately leading to moredifficulty in planning for the future.

What surprised us in our research was how quicklyinternal transparency fell away and the need for anIS arose. This need translated into a requirementfor the organizational IS to be used to maintain theinformal cross-functional communication channelsthat used to be accomplished through face-to-facecommunication among top management. Using ISto provide a window where management caninquire on and observe the interdependentoperations of their management colleagues on anas-needed, ad hoc basis offers at least twoimportant benefits. First, proper access to cross-functional information offers the capability for time-constrained managers to gain access to requiredoperational information according to their ownneeds and schedule. Second, the as-needed, adhoc access also facilitates and maintains theinformal communication processes thatcharacterize successful small businesses.

We thought about what might have happened atELCO without this project. They might havepurchased more IS and internal transparencymight have increased by luck in picking andimplementing the right software. Poor IS wouldhave continued to be seen as the cause of theproblem and the next time a similar problem arose,the same solution may have been attempted.Technology would always have been seen as thesolution. Alternatively, they might have realizedthat there was a problem with the social system,gone on a corporate retreat, and resolved tocommunicate cross-functional information better.This might have worked fine until time pressuresbuilt again and caused communication to decline.There was a genuine requirement to improve theIS in a focused way. Finally, through the actionresearch process, we were able to shed light on

the root problem and to address ELCO’s difficultiesdirectly and facilitate the development of a betterunderstanding of the underlying processes. Thisbenefitted ELCO as well as similar firms and theresearch community, all in line with the basic goalsof action research.

Acknowledgements

The authors gratefully acknowledge the assistanceof ELCO (a pseudonym) and its managers fortheir time and devotion to this study and Materialsand Manufacturing Ontario (MMO) for providingfunding, as well as the anonymous reviewers andassociate editor for their insightful suggestions interms of both theory and method, and finally toDavid Goldsmith for motivating some of ourreflections.

References

Alavi, M., and Leidner, D. . “Knowledge Manage-ment Systems: Issues, Challenges, and Bene-fits,” Communications of the Association forInformation Systems (1:7), 1999.

Alavi, M., and Leidner, D. E. “Review: Knowl-edge Management and Knowledge Manage-ment Systems: Conceptual Foundations andResearch Issues,” MIS Quarterly (25:1), 2001,pp. 107-136.

Ang, C.-L., Davies, M., and Finlay, P. N. “Mea-sures to Assess the Impact of InformationTechnology on Quality Management,” TheInternational Journal of Quality & ReliabilityManagement (17:1), 2000, pp. 42-66.

Ballantine, J., Levy, M., and Powell, P. “Evalu-ating Information Systems in Small andMedium-Sized Enterprises: Issues and Evi-dence,” European Journal of InformationSystems (7:4), 1998, pp. 241-251.

Barringer, B. R., Jones, F. F., and Lewis, P. S. “AQualitative Study of the Management Practicesof Rapid-Growth Firms and How Rapid-GrowthFirms Mitigate the Managerial Capacity Prob-lem,” Journal of Developmental Entrepre-neurship (3:2), 1998, pp. 97-122.



Baskerville, R. “Investigating Information Systemswith Action Research,” Communications of theAssociation for Information Systems (2:19),1999.

Baskerville, R., and Pries-Heje, J. “KnowledgeCapability and Maturity in Software Manage-ment,” Database for Advances in InformationSystems (30:2), 1999, pp. 26-43.

Baskerville, R., and Stage, J. “Controlling Proto-type Development Through Risk Analysis,” MISQuarterly (20:4), 1996, pp. 481-504.

Baskerville, R., and Wood-Harper, A. T. “A CriticalPerspective on Action Research as a Methodfor Information Systems Research,” Journal ofInformation Technology (11), 1996, pp. 235-246.

Baskerville, R., and Wood-Harper, A. T. “Diversityin Information Systems Action ResearchMethods,” European Journal of InformationSystems (7), 1998, pp. 90-107.

Bavelas, A. “An Experimental Approach toOrganizational Communication,” Personnel(27), 1951, pp. 366-371.

Beech, N., and Crane, O. “High PerformanceTeams and a Climate of Community,” TeamPerformance Management (5:3), 1999, pp. 87-102.

Bergeron, F., Raymond, L., and Rivard, S. “Fit inStrategic Information Technology ManagementResearch: An Empirical Comparison of Per-spectives,” Omega (29:2), 2001, pp. 125-142.

Braunstein, J. “Ford Motor’s Software Revolution,”Manufacturing Engineering (123:5), 1999, pp.64-69.

Brown, J. S., and Duguid, P. The Social Life ofInformation, Harvard Business School Press,Boston, MA, 2000.

Chiasson, M., and Dexter, A. S. “Systems Devel-opment Conflict During the Use of anInformation Systems Prototyping Method ofAction Research: Implications for Practice,”Information Technology & People (14:1), 2001,pp. 91-108.

Churchill, N. C., and Lewis, V. L. “The FiveStages of Small Business Growth,” HarvardBusiness Review (61:3), 1983, pp. 30-39.

Cragg, P. B., and King, M. “Small-Firm Com-puting: Motivators and Inhibitors,” MIS Quar-terly (17:1), 1993, pp. 47-60.

Cragg, P. B., and Zinatelli, N. “The Evolution ofInformation Systems in Small Firms,” Informa-tion & Management (29:1), 1995, pp. 1-8.

Davison, R. M. “GSS and Action Research in theHong Kong Police,” Information Technology &People (14:1), 1999, pp. 60-77.

Davison, R. M, and Vogel, D. “Group SupportSystems in Hong Kong: An Action ResearchProject,” Information Systems Journal (10),2000, pp. 3-20.

Day, G. S., and Wensley, R. “Assessing Advan-tage: A Framework for Diagnosing CompetitiveSuperiority,” Journal of Marketing (52), 1988,pp. 1-20.

DeLone, W. H. “Determinants of Success forComputer Usage in Small Business,” MIS Quar-terly (12:1), 1988, pp. 51-61.

Eden, C., and Huxham, C. “Action Research forManagement Research,” British Journal ofManagement (7:1), 1996, pp. 75-86.

Elden, M., and Chisholm, R. F. “EmergingVarieties of Action Research: Introduction tothe Special Issue,” Human Relations (46), 1993,pp. 121-142.

Forza, C. “The Impact of Information Systems onQuality Performance: An Empirical Study,”International Journal of Operations & Produc-tion Management (15:6), 1995, pp. 69-83.

Gersick, C. “Revolutionary Change Theories: AMultilevel Exploration of the PunctuatedEquilibrium Paradigm,” Academy of Manage-ment Review (16:1), 1991, pp. 10-36.

Greiner, L. E. “Evolution and Revolution asOrganizations Grow,” Harvard Business Review(50:4), 1972, pp. 37-46.

Hambrick, D. C., and Crozier, L. M. “Stumblersand Stars in the Management of Rapid Growth,”Journal of Business Venturing (1), 1985, pp. 31-45.

Hanks, S. H. “The Organization Life Cycle: Inte-grating Content and Process,” Journal of SmallBusiness Strategy (1:1), 1990, pp. 1-13.

Hansen, E. L. “Entrepreneurial Networks and NewOrganization Growth,” Entrepreneurship Theoryand Practice (19:4), 1995, pp. 7-19.

Harrison, D. A., Mykytyn, P. P., Jr., and Riemen-schneider, C. K. “Executive Decisions AboutAdoption of Information Technology in SmallBusiness: Theory and Empirical Tests,” Infor-



mation Systems Research (8:2), 1997, pp. 171-195.Hodgetts, R. M., and Kuratko, D. F. Effective

Small Business Management, Harcourt College,Fort Worth, TX, 2001.

Hussin, H., King, M., and Cragg, P. “IT Alignmentin Small Firms,” European Journal of Infor-mation Systems (11:2), 2002, pp. 108-127.

Iacovou, C. L., Benbasat, I., and Dexter, A. S.“Electronic Data Interchange and Small Organi-zations: Adoption and Impact of Technology,”MIS Quarterly (19:4), 1995, pp. 465-485.

Igbaria, M., Zinatelli, N., and Cavaye, A. L. M.“Analysis of Information Technology Success inSmall Firms in New Zealand,” InternationalJournal of Information Management (18:2),1998, pp. 103-119.

Igbaria, M., Zinatelli, N., Cragg, P. B., and Cavaye,A. L. M. “Personal Computing AcceptanceFactors in Small Firms: A Structural EquationModel,” MIS Quarterly (21:3), 1997, pp. 279-305.

Jensen, M., and Meckling, W. “Theory of the Firm:Managerial Behavior, Agency Costs andOwnership Structure,” Journal of FinancialEconomics (3), 1976, pp. 305-360.

Klein, H. K., and Myers, M. D. “A Set of Principlesfor Conducting and Evaluating Interpretive FieldStudies in Information Systems,” MIS Quarterly(23:1), 1999, pp. 67-94.

Kock, N. “Negotiating Mutually Satisfying ISAction Research Topics with Organizations: AnAnalysis of Rapoport’s Initiative Dilemma,”Journal of Workplace Learning (9:7), 1997, pp.253-262.

Kotter, J. P. Leading Change, Harvard BusinessSchool Press, Cambridge, MA, 1996.

Lamming, R. C., Caldwell, N. D., Harrison, D. A.,and Phillips, W. “Transparency in Supply Rela-tionships: Concept and Practice,” The Journalof Supply Chain Management (37:4), 2001, pp.4-10.

Langley, A. “Strategies for Theorizing from Pro-cess Data,” Academy of Management Review(24:4), 1999, pp. 691-710.

Lau, F. “Towards a Framework for ActionResearch in Information Systems Studies,”Information Technology & People (12:2), 1999.

Leavitt, H. J. “Unhuman Organizations,” HarvardBusiness Review, July-August 1962, pp. 90-98.

Lees, J. D., and Lees, D. D. “Realities of SmallBusiness Information System Implementation,”Journal of Systems Management (38:1), 1987,pp. 6-13.

Levy, M., and Powell, P. “SME Flexibility and theRole of Information Systems,” Small BusinessEconomics (11), 1998, pp. 183-196.

Lincoln, Y., and Guba, E. Naturalistic Inquiry,Sage, Newbury, CA, 1985.

Min, S., Song, S., and Keebler, J. S. “An Internet-Mediated Market Orientation (IMO): Building aTheory,” Journal of Marketing Theory andPractice (10:2), 2002, pp. 1-11.

Montazemi, A. R. “Factors Affecting InformationSatisfaction in the Context of the Small Busi-ness Environment,” MIS Quarterly (12:2), 1988,pp. 239-256.

Mumford, E. “Advice for an Action Researcher,”Information Technology & People (14:1), 2001,pp. 12-27.

Myers, M. D. “Investigating Information Systemswith Ethnographic Research,” Communicationsof the Association for Information Systems(2:23), 1999, pp. 1-19.

Narver, J. C., and Slater, S. F. “The Effect of aMarket Orientation on Business Profitability,”Journal of Marketing (54:4), 1990, pp. 20-35.

Newman, M., and Robey, D. “A Social ProcessModel of User-Analyst Relationships,” MISQuarterly (16:2), 1992, pp. 249-266.

Oi, W. “Heterogeneous Firms and the Organi-zation of Production,” Economic Inquiry (21),1983, pp. 147-171.

Olesen, K., and Myers, M.D. “Trying to ImproveCommunication and Collaboration with Infor-mation Technology: An Action Research Pro-ject Which Failed,” Information Technology &People (12:4), 1999, pp. 317-332.

Olson, P. D., and Terpstra, D. E. “OrganizationalStructural Changes: Life-Cycle Stage Influ-ences and Managers’ and Interventionists’Challenges,” Journal of Organizational ChangeManagement (5:4), 1992, pp. 27-40.

Orlikowski, W. J. “Using Technology and Con-stituting Structures: A Practice Lens forStudying Technology in Organizations,” Organi-zation Science (11:4), 2000, pp. 404-428.

Orlikowski, W. J., and Baroudi, J. J. “StudyingInformation Technology in Organizations:



Research Approaches and Assumptions,”Information Systems Research (2:1), 1991, pp.1-27.

Orlikowski, W. J., and Robey, D. “InformationTechnology and the Structuring of Organi-zations,” Information Systems Research (2:2),1991, pp. 143-169.

Pfeffer, J., and Salancik, G. R. The External Con-trol of Organizations, Harper and Row, NewYork, 1978.

Phillips, T. J., Luehlfing, M. S., and Vallario, C. W.“Hazy Reporting,” Journal of Accountancy(194:2), 2002, pp. 47-52.

Pollard, C. E., and Hayne, S. C. “The ChangingFace of Information System Issues in SmallFirms,” International Small Business Journal(16:3), 1998, pp. 70-87.

Pressman, R. S. Software Engineering: A Prac-titioner’s Approach, McGraw-Hill Higher Educa-tion, New York, 2001.

Psillos, S. Scientific Realism: How ScienceTracks Truth, Routledge, New York, 1999.

Rapoport, R. N. “Three Dilemmas in ActionResearch,” Human Relations (23:6), 1970, pp.499-513.

Raymond, L. “Organizational Characteristics andMIS Success in the Context of Small Business,”MIS Quarterly (9:1), 1985, pp. 37-52.

Roethlisberger, F., and Dickson, W. Managementand the Worker, Harvard University Press,Cambridge, MA, 1939.

Romanelli, E., and Tushman, M. L. “Organiza-tional Transformation as Punctuated Equili-brium: An Empirical Test,” Academy of Man-agement Journal (37:5), 1994, pp. 1141-1166.

Sabherwal, R., Hirschheim, R., and Goles, T.“The Dynamics of Alignment: Insights from aPunctuated Equilibrium Model,” OrganizationScience (12:2), 2001, pp. 179-197.

Schwartz, D. G., and Te'eni, D. “Tying Knowledgeto Action with kMail,” IEEE Intelligent Systems(15:3), 2000, pp. 33-39.

Scott, M., and Bruce, R. “Five Stages of Growth inSmall Business,” Long Range Planning (20:3),1987, pp. 45-52.

Seashore, S. “The Design of Action Research,” inExperimenting with Organizational Life: TheAction Research Approach, A. Clarke (ed.),Plenum, New York, 1976, pp. 103-117.

Shin, N. “Does Information Technology ImproveCoordination? An Empirical Analysis,” LogisticsInformation Management (12:1/2), 1999, pp.138-144.

Singer, E. T. Action Research (2nd ed.), Sage,Thousand Oaks, CA, 1999.

Small Business Administration. “Guide to SBA’sDefinitions of Small Business,” March 20, 2003(available online at http://www.sba.gov/size/indexguide.html, modification date; accessedApril 4, 2003 ).

Snyder, R. A., and Morris, J. H. “OrganizationalCommunication and Performance,” Journal ofApplied Psychology (69:3), 1984, pp. 461-465.

Steinmetz, L. L. “Critical Stages of Small Busi-ness Growth,” Business Horizons (12:1), 1969,pp. 29-44.

Stringer, E. T. Action Research: A Handbook forPractitioners, Sage, Newbury Park, CA, 1999.

Stubbart, C. I., and Smalley, R. D. “The DeceptiveAllure of Stage Models of Strategic Processes,”Journal of Management Inquiry (8:3), 1999, pp.273-287.

Susman, G. I., and Evered, R. D. “An Assessmentof the Scientific Merits of Action Research,”Administrative Science Quarterly (23:4), 1978,pp. 582-603.

Teo, T. S. H., and King, W. R. “Integrationbetween Business Planning and InformationSystems Planning: An Evolutionary-Contin-gency Perspective,” Journal of ManagementInformation Systems (14:1), 1997, pp. 185-214.

Thong, J. Y. L. “An Integrated Model of Informa-tion Systems Adoption in Small Businesses,”Journal of Management Information Systems(15:4), 1999, pp. 187-134.

Tushman, M. L., and Romanelli, E. “Organiza-tional Evolution: A Metamorphosis Model ofConvergence and Reorientation,” in Researchin Organizational Behavior (Volume 7), B. M.Staw (ed.), JAI Press, Greenwich, CT, 1985,pp. 171-222.

Tyre, M. J., and Orlikowski, W. J. “Windows ofOpportunity: Temporal Patterns of Technolo-gical Adaption in Organizations,” OrganizationScience (5:1), 1994, pp. 98-118.

Van de Ven, A. H., and Poole, M. S. “ExplainingDevelopment and Change in Organizations,”Academy of Management Review (20:3), 1995,pp. 510-540.



Vinten, G. “Corporate Communications in Small-and Medium-Sized Enterprises,” Industrial andCommercial Training (31:3), 1999, pp. 112-119.

Vishwanath, T., and Kaufmann, D. “TowardTransparency: New Approaches and Their Ap-plication to Financial Markets,” The World BankResearch Observer (16:1), 2002, pp. 41-58.

Whyte, W. (ed.). Participatory Action Research,Sage, London, 1991.

Wickert, A., and Herschel, R. “Knowledge-Management Issues for Smaller Businesses,”Journal of Knowledge Management (5:4), 2001,pp. 329-337.

Winston, E. R., and Dologite, D. “How DoesAttitude Impact IT Implementation: A Study ofSmall Business Owners,” Journal of End UserComputing (14:2), 2002, pp. 16-29.

Zinatelli, N., Cragg, P. B., and Cavaye, A. L. M.“End User Computing Sophistication and Suc-cess in Small Firms,” European Journal ofInformation Systems (5:3), 1996, pp. 172-181.

About the Authors

Chris Street is a doctoral candidate at Queen’sUniversity and research associate with the Centrefor Knowledge-Based Enterprises, Queen’s Uni-versity. His research interests deal with the effectsof information technology on organizational growthand development, with a focus on small businessstrategies.

Darren Meister is an assistant professor of Infor-mation Systems and Robert V. Brouillard FacultyFellow at the Richard Ivey School of Business,University of Western Ontario. His interests focuson the integration of technology with inter- andintra-organizational processes, knowledge man-agement, and technology adoption by profes-sionals.

small business growth and internal transparency: the role

Documents