a multi-project model of key factors affecting organizational benefits from enterprise systems;...

Seddon et al./Key Factors Affecting Organizational Benefits

RESEARCH ARTICLE

A MULTI-PROJECT MODEL OF KEY FACTORS AFFECTINGORGANIZATIONAL BENEFITS FROM ENTERPRISE SYSTEMS1

By: Peter B. SeddonDepartment of Information SystemsUniversity of MelbourneVictoria [email protected]

Cheryl CalvertCorporate Business SystemsMonash UniversityVictoria [email protected]

Song YangDepartment of Information SystemsUniversity of MelbourneVictoria [email protected]

1Carol Saunders was the accepting senior editor for this paper.

The appendices for this paper are are located in the “Online Supplements”section of the MIS Quarterly’s website (http://www.misq.org).

This article contains references to the products of SAP AG. SAP, R/3, SAP NetWeaver,Duet, PartnerEdge, ByDesign, Clear Enterprise, SAP Business Objects Explorer, andother SAP products and services mentioned herein as well as their respective logos aretrademarks or registered trademarks of SAP AG in Germany and other countries. Business Objects and Business Objects logos, BusinessObjects, Crystal Reports, CrystalDecisions, Web Intelligence, Xcelsius, and other Business Objects products and servicesmentioned herein as well as their respective logos or trademarks are registeredtrademarks of SAP France in the United States and other countries.

SAP AG is neither the author nor the publisher of this publication and is not responsiblefor its content. SAP Group shall not be liable for errors or omissions with respect to thematerials. The only warranties for SAP Group products and services are those that areset forth in the express warranty statements accompanying such products and services,if any. Nothing herein should be construed as constituting an additional warranty.

Abstract

This paper develops a long-term, multi-project model of fac-tors affecting organizational benefits from enterprise systems(ES), then reports a preliminary test of the model. In theshorter-term half of the model, it is hypothesized that once asystem has gone live, two factors, namely functional fit andovercoming organizational inertia, drive organizational bene-fits flowing from each major ES improvement project. Theimportance of these factors may vary from project to project.In the long-term half of the model, it is hypothesized that fouradditional factors, namely integration, process optimization,improved access to information, and on-going major ESbusiness improvement projects, drive organizational benefitsfrom ES over the long term. Preliminary tests of the modelwere conducted using data from 126 customer presentationsfrom SAP’s 2003 and 2005 Sapphire U.S. conferences. All sixfactors were found to be important in explaining variance inorganizational benefits from enterprise systems from theperspective of senior management.

Keywords: Enterprise system success, packaged software,functional fit, overcoming organizational inertia, change man-agement, IS implementation, IS project management, integra-tion, process optimization, improved access to information

Introduction

Enterprise systems are large-scale, real-time, integratedapplication-software packages that use the computational,data storage, and data transmission power of modern informa-tion technology to support processes, information flows,reporting, and business analytics within and between complexorganizations. Because they impound deep knowledge of

MIS Quarterly Vol. 34 No. 2 pp. 305-328/June 2010 305


new ways of designing and executing organizational pro-cesses, these complex software packages can cause consider-able assimilation difficulties for client organizations (Robeyet al. 2002). Some people equate the terms enterprise systemand ERP, but in this paper the term enterprise system is usedto refer to all large organization-wide packaged applicationsincluding enterprise resource planning (ERP), customerrelationship management (CRM), supply chain management(SCM), data warehousing, and any application components ofthe software platforms on which these applications are built(e.g., SAP’s NetWeaver and Oracle’s Fusion).

Worldwide investment in enterprise systems (ES) has beenextensive. According to AMR Research, such investment inES was U.S. $36 billion in 2004 (Reilly 2005). Focusing juston ERP, Gartner (Hestermann et al. 2009) estimated that theworldwide ERP software market was U.S. $24 billion in2008. Individual firms have also spent millions of dollarsacquiring and implementing ES; for example, DisneyCorporation reported at a presentation at SAP’s annual userconference, Sapphire 2003,2 that it spent $400 million on itstwo-year SAP ES consolidation project. However, as withmost large IT projects, not all ES projects go smoothly.Widely reported disasters include the FoxMeyer Drug com-pany, a U.S. $5 billion per annum revenue pharmaceuticalcompany that went bankrupt and sued SAP and AndersenConsulting for U.S. $500 million after its failed SAPimplementation (Scott 1999), and Hershey’s, a U.S. $4 billionper annum revenue confectionary maker, that spent U.S. $112million implementing an SAP system, and which lostU.S. $150 million in revenue as a result of logistics problemsin the first year after go live (Carr 2002). With such largeexpenditures on ES, and significant risks of failure, it isimportant for managers to understand what makes some ESinvestments more successful than others. Hence the researchquestion posed in this paper: What key factors explainvariance in organizational benefits from enterprise systems?

Our answer to this question is the multi-project model offactors affecting organizational benefits from ES use shown

in Figure 1. We call this the organizational benefits fromenterprise systems model, or OBES for short. The variablesin OBES are defined in Table 1. In OBES, the term multi-project refers to the series of projects depicted on the right-hand side of Figure 1. Although we are aware that groups ofprojects are often coordinated through some sort of over-arching program or Program Office, the term project in thispaper refers to individual ES projects. Normally such projectsgo live with different functionality or at different times or indifferent geographic locations.

The OBES model consists of two variance models,3 one along-term model of factors affecting organizational benefitsfrom ES use (on the left), the other a shorter-term model offactors affecting organizational benefits from individual majorES business improvement projects post go live (on the right).Major business improvement projects are large projects thatlead to changes in the way that work is done in the business;this is in contrast to cost-reduction projects (e.g., the mergingof two systems) or so-called “technical upgrades” that leadto improvements in the ES infrastructure that are invisible tothe business. The OBES model is split into two parts becausemost firms that invest in ES find themselves embarking on notjust one, but a series of major business improvement projectsover the course of some years (i.e., the initial implementation,followed by various upgrade, extension, and consolidationprojects). The post-go-live consequences of each of theseprojects need to be modeled separately because some projectsare likely to have better outcomes than others.

Although longitudinal variation of benefits is not exploredempirically in the current study, in formulating the OBESmodel we also sought to explain how the dependent variable,organizational benefits from ES use, from the perspective ofsenior management, changes over time. The benefits-versus-time graph for individual projects on the right of Figure 1 isbased on Ross and Vitale (2000), Gattiker and Goodhue(2005, Figure 4, p. 576), and Cotteleer and Bendoly (2006).It shows a typical dip in performance immediately followingproject go live, with benefits per period rising in the next fewyears. Not all firms experience this dip, but many do. TheOBES model asserts that the drivers of increasing benefits in

2The Sapphire conferences are a series of annual conferences organized bySAP, the world’s largest vendor of enterprise systems, in various locationsaround the world. Sapphire conferences provide a vehicle for SAP to informtheir customers of new product developments and for their customers to tryout new software and exchange information about implementation experi-ences and what they are doing with SAP software. At a typical 3-day U.S.Sapphire conference, there are over 10,000 attendees, many paying somethousands of dollars each to attend. Consistent with Ramiller et al.’s (2008,p. 9) observation that practitioners’ interest in ES (specifically, ERP) peakedin 1999 then fell markedly by 2003, attendance at SAP’s U.S. Sapphireconferences rose to 15,000 in 1998, dropped to 7,000 in 2003, then rosesteadily to 15,000 in 2008 before dropping to 10,000 in 2009.

3Webster and Watson (2002, p. xix) say “variance theories incorporateindependent variables that cause variation in dependent variables.” Indiagrammatic representations of variance theories, the higher the score for theindependent variable at the tail of an arrow, the higher the score expected forthe dependent variable at the head of the arrow. Two alternatives to variancemodels are process models (Mohr 1982), which identify a series of steps thatif executed in the specified order lead to a predictable outcome, andconfiguration models (Ragin 1987), which assert that the presence or absenceof certain combinations of independent variables affect an outcome.

306 MIS Quarterly Vol. 34 No. 2/June 2010


Figure 1. OBES Model of Factors Affecting Organizational Benefits from ES

Table 1: Definitions of Factors in the OBES Model (Figure 1)

Factor Definition

Organizationalbenefits fromsystem use,from theperspective ofseniormanagement

The dependent variable in the OBES model, organizational benefits from system use, from the perspec-tive of senior management, is an overall measure of senior management’s perception of the benefits fromthe IT-based application. Such benefits—which may be assessed either for the ES investment overall, orfor individual ES projects—usually revolve around the software enabling (1) faster, more accurate processcoordination and execution, including links with business partners up and down the supply chain, and(2) greater accuracy of and visibility into organizational data, resulting in more tightly controlled organiza-tional processes, improved asset utilization, and improved decision making. In almost all cases, suchbenefits vary over time, as depicted in the two graphs in Figure 1.

Short-term model: Project-focused independent variables likely to have different outcomes for each project:

Functional fit(FF)

Functional fit is the extent to which the functional capabilities embedded and configured within an ESpackage match the functionality that the organization needs to operate effectively and efficiently. Sayingthat software has good functional fit is equivalent to saying that (1) the processes supported by the ESare efficient and effective for the organization, and (2) the software helps people in the organization gettheir jobs done. FF is conceptualized as being delivered and measured project by project. Note that FFdoes not consider the capacity or desire of people in the organization to use or work with the system,which is captured by OOI.

Overcomingorganizationalinertia (OOI)

Overcoming organizational inertia is the extent to which members of the organization have been moti-vated to learn, use, and accept the new system. During initial implementation and subsequent upgradeprojects, considerable change-management effort, training, and support are needed to overcome organi-zational inertia. OOI is conceptualized as being measured project by project.

Long-Term Organizational Benefits ModelShort-Term Model: Factors Driving

Benefits from Each Project

Expected long-term benefitsfor many major projects

years

On-Going ES Business-Improvement Projects

Integration

ProcessOptimization

ImprovedAccess to

Information

Organizational Benefits from System Use, from the Perspective of Senior Management

On-Going Major ES Business- Improvement

Projects

H3

H4

H5

H6

Pursuit of the three goals above guides decisions about which ES-improvementprojects are to beundertaken in future.

Major Project “n”

Continuous improvement

Benefits from theprevious system(s)

Implementation project Shakedown

Benefits from the implementation

6 - 12 months Go live Time

Go live

Functional Fit (FF)

Overcoming organizational inertia (OOI)

Organizational Benefits from System Use, from the perspective of Senior Management for this project

(post go-live)

H1

H2

MIS Quarterly Vol. 34 No. 2/June 2010 307


Table 1. Definitions of Factors in the OBES Model (Figure 1) (Continued)

Factor Definition

Long-term independent variables:

Integration Integration of information systems is the unification of processes, systems, and/or data from multiplecomputer-based systems, not necessarily in the one organization. According to Ross et al. (2006, pp. 27-28), “Integration links the efforts of organizational units through shared data. This sharing of data can bebetween processes to enable end-to-end transaction processing, or across processes to allow the com-pany to present a single face to customers.…The biggest challenge of integration is usually around data.”

Processoptimization

Process optimization is any attempt to improve the efficiency and effectiveness of an organization’s pro-cesses, ultimately in support of its strategic goals. By working with key clients to build software to helpclient firms standardize and optimize their processes, ES vendors offer their customers the promise ofaccess to “best practice” process templates from other leading organizations. However, ES packagesalso contain facilities for tailoring processes to specific local needs.

Improvedaccess toinformation

Improved access to information is any step taken to increase the provision of timely, accurate, relevantinformation (including previously hidden information) to key organizational decision makers. The term isintended to capture the same idea as Davenport et al.’s (2002) “informate.”

On-goingmajor ESbusinessimprovementprojects

On-going major ES business improvement projects is a measure of the number and extent of investmentin major business improvement projects that an organization has undertaken for improving and extendingits enterprise system. Major business projects are those that lead to changes in the way that work is donein the business (as opposed to infrastructure changes that are invisible to the business). Examplesinclude implementation of a CRM system after an ERP system, an upgrade to an existing ERP systemthat leads to changed processes, or a new data warehousing project. These projects are represented inthe OBES model by the series of major business projects depicted on the right of Figure 1.

this project-oriented view of ES benefits are increased func-tional fit (e.g., as the result of minor projects) and success inovercoming organizational inertia. Wagner and Newell(2007) identify similar mechanisms.

The long-term benefits graph on the left of Figure 1 is basedon Davenport et al.’s (2002) empirical findings (Figure 15, p.26), where benefits per period are shown rising more thanfour years after the initial implementation. We argue that thetwo primary drivers of increasing benefits in this long-termmodel are the firm’s on-going investments in ES projects andincreased benefits from each project as depicted in the graphon the right of Figure 1. With respect to new projects,choices of which projects to undertake are guided, in themain, by the pursuit of the three factors on the left-hand sideof the long-term model (i.e., greater integration, processoptimization, and improved access to information).

The goal for this paper is to present, justify, and conduct apreliminary test of the OBES model. In the remainder of thispaper, the OBES model is synthesized from the literature,then assessed using data from 126 customer presentations attwo leading industry conferences, namely SAP’s 2003 and2005 Sapphire USA conferences. The contribution of thispaper is this multi-project model of factors affecting organi-zational benefits from enterprise systems, combined with theevidence that the model seems to fit the data very well.

Is a New ES Benefits Model Needed?

This section uses a literature review, summarized in Table 2,to argue that an integrated model of factors affecting benefitsfrom ES would be a valuable contribution to the IS literature.The papers cited in Table 2 were identified in a systematicanalysis conducted in September 2007 using Google Scholarand Harzing’s “Publish or Perish” search tool. Since GoogleScholar’s goal is to index close to the full population ofacademic publications, high average citations per year providea reasonably objective indicator of the important publicationsin any topic area. Using the search terms, enterprise systemand software, enterprise resource planning and software, andERP and software, Google Scholar returned 2,370 distinctpublications,4 with a total of 28,500 citations. After sortingthis list into descending order by average citations per yearsince publication, the top 200 of those publications, ac-counting for 13,600 citations, were analyzed in depth. Ofthose 200 publications, the six primary focal areas consideredrelevant to the current study are shown in the middle column

4As an indication of the extent of data quality problems in GoogleScholar—which are very real—there was no year of publication for 290 ofthe 2,370 publications.



Table 2. Example Studies of Enterprise System Projects and Benefits

BroadTheme Primary Focus Example Publications (see Notes below)

Projects 1. Implementation projectsoverall

Markus and Tanis (2000) 33, Hong and Kim (2002) 30, Bancroft et al. (1998)19, Markus et al. (2000) 18, Ross and Vitale (2000) 16, Motwani et al. (2002)12, Scott and Vessey (2002) 12, Sumner (2000) 11, Parr and Shanks (2000)11, Rajagopal (2002) 10, Mabert et al. (2003a, b) 9, 9 (resp.), Somers andNelson (2004) 9, Strong and Volkoff (2004)

2. Project critical successfactors (CSFs)

Bingi et al. (1999) 35, Umble et al. (2003) 28, Holland and Light (1999) 25,Robey et al. (2002) 23, Nah et al. (2001) 20, Al-Mashari et al. (2003) 18,Akkermans and van Helden (2002) 16, Sarker and Lee (2003) 14, Somersand Nelson (2001) 12, Parr et al. (1999) 8, Finney and Corbett (2007)

3. Functional fit Hong and Kim (2002) 30, Soh et al. (2000) 28, Scheer and Haberman (2000)15, Keller and Teufel (1998) 14, Dalal et al. (2004) 9, Soffer et al. (2003) 9,Somers and Nelson (2003) 6, Sia and Soh (2002), Luo and Strong (2004),Rosemann et al. (2004), Wei et al. (2005), Soh and Sia (2005), Light (2005),Financial Executives International & CSC (2006), Keil and Tiwana (2006)

4. Overcoming organizationalinertia:a. Resistance/user

acceptance/ changemanagement

Markus and Tanis (2000) 33, Markus (2004) 11, Aladwani (2001) 9, McAffee(2002) 9, Abdinnour-Helm et al. (2003) 8, Boudreau and Robey (1999) 5,Lapointe and Rivard (2005), Staehr et al. (2006), Calvert (2006), Liang et al.(2007)

b. Learning/knowledgetransfer

Robey et al. (2002) 23, Ko et al. (2005) 16, Boudreau and Robey (2005) 9,Boudreau (2003), Volkoff et al. (2004), Staehr et al. (2006)

Impacts 5. Organizational benefitsfrom enterprise systems

Markus and Tanis (2000) 33, Shang and Seddon (2002) 9, Skok and Legge(2002) 8, Wu and Wang (2007) 8, Deloitte Consulting (1998), Davenport etal. (2002), Gable et al. (2003), Sedera and Gable (2004), Gefen andRagowsky (2005), Cotteleer and Bendoly (2006), Harris and Davenport(2006), Staehr (2007)

6. Factors affecting variancein organizational benefits

Hong and Kim (2002) 30, Al-Mashari et al. (2003) 18, Davenport et al. (2004)9, Bradford and Florin (2003) 9, Somers and Nelson (2003) 6, Gattiker andGoodhue (2005) 4, Shang (2001), Staehr et al. (2006)

Notes: 1. The average citation count per year since publication, from Google Scholar in September, 2007, is shown after most citations.2. Citations are listed in descending order by this average annual citation count. 3. Additional author-selected studies that influenced the formulation of the OBES model are shown in italics.

of Table 2. Forty percent of the 200 publications were classi-fied into these six primary focal areas.5

The studies cited in the right-hand column of Table 2 arearranged in descending order by average citations per year.Some publications, e.g., Markus and Tanis (2000) and Hong

and Kim (2002), appear in more than one band because theyaddress more than one issue. Toward the end of each list ofcitations, we have added in italics a number of additionalstudies (with no average citation count) that were also influ-ential in formulating the model in Figure 1. The OBES modelwas synthesized from a combination of the publications inboth Table 2 and the broader IS literature on factors affectingbenefits from organization-wide applications of IT.

The studies in row 6 of Table 2 (i.e., those on factors affectingvariance in organizational benefits from ES) define thebenchmark against which the contribution of this study should

5Focal areas considered not relevant include ES in general (12%, includingDavenport (1998) with over 1,000 citations), supply chain (11.5%), EStechnology (9%), eBusiness (7%), and technical integration (5%). Workingindependently, Moon (2007, Table 3) classified the ERP literature undersimilar headings to those in Table 2.



(a) Copyright © 2002, Elsevier; used by permission.(b) Copyright © 2004, Emerald; used by permission.(c) Copyright © 2005, Regents of the University of Minnesota; used by permission.

Figure 2. Three Very Different Models of Factors Affecting Organizational Benefits from ES

be judged. Models from three key studies of factors affectingorganizational benefits—Hong and Kim (2002), Davenport etal. (2004), and Gattiker and Goodhue (2005)—are shown inFigure 2. These studies were singled out because they presentthree very different explanations of the same dependentvariable, organizational benefits from ES use. Hong andKim’s model (see Figure 2a) focuses on project-oriented, notlong-term, drivers of ES project success. It assumes that theproject has been successful in going live, and includes projectcost and time as indicators of project success. Hong andKim’s H1, H2, and H3 all address the positive association

between functional fit and project success, and how to achievefunctional fit (through either software or organizationaladaptation). Empirically, based on their sample of 105responses from project-team members in 34 firms, Hong andKim report strong support for their H1 (p = 0.002, AdjustedR² = 0.24). In their H4, Hong and Kim posit that organiza-tional resistance moderates the relationship between fit andimplementation success. Empirically, they found that thishypothesized H4 interaction term was not significant, but thatthere was a strong negative correlation between organiza-tional resistance and success (r = -0.48, p < 0.01). Summa-

Contingency Variables• ERP Adaptation Level (H2)• Process Adaptation Level (H3)• Organizational Resistance (H4)

Organizational Fit of ERP

• Data Fit• Process Fit• User Fit

ERP Implementation Success

• cost• time• performance• benefits

H1

(a) Hong and Kim (2002), Figure 2, p. 28.

N = 105 respondents from 34 firms. H4 was not significant.

(b) Davenport et al. (2004), Figure 2, p. 18.

N = 163 firms globally.

Integrate

Optimize

Informate

BenefitsRealized

ImplementExtensively

Invest inthe ES

Spend Timewith the ES

0.18 0.16

0.23

0.2

0.22

0.19

0.18

0.15

R2 = 0.13

(c) Gattiker and Goodhue (2005), Figure 3, p. 575.

N = 111 manufacturing plants.

Interdependence

Differentiation

Customization

Time elapsed sinceimpementation

CoordinationImprovements

Data Quality

Task Efficiency

Local (Plant) Level Overall Benefits

*p < .05**p < .01***p < .001

.81***

-.09*

-.19*

.14**

.30***

.19***

.57***

.15**

.36***

.52***



rizing, Hong and Kim’s model paints a very project-orientedview of factors affecting organizational benefits from ES. Asa general model of factors affecting organizational benefitsfrom ES, its major limitation is that it does not recognizeexplicitly the multi-project nature of ES ownership, nor doesit seek to identify the long-term drivers of organizationalbenefits from ES.

Davenport et al.’s (2004) model (see Figure 2b) takes alonger-term, non-project-oriented view of factors affectingorganizational benefits from ES that identifies three quite dif-ferent benefit drivers compared to Hong and Kim. Its threevariables, integrate, optimize, and informate6 were confirmedas important in Harris and Davenport’s (2006) recent globalsurvey of 371 executives. Empirically, based on multiple-regression analysis using data from a global sample of 163organizations, Davenport et al. (2004) report significant pathcoefficients of 0.19, 0.18, and 0.15, respectively, betweentheir three variables and organizational benefits (R² = 0.13).As a general model, the limitation of Davenport et al.’s modelis that it does not recognize explicitly7 that projects are themechanism through which organizations achieve their longer-term integration, process optimization, and improved accessto information goals. As a consequence, it does not considerexplicitly any project-related factors such as functional fit andchange management that affect organizational benefits fromES.

The third model in Figure 2—Figure 2c, from Gattiker andGoodhue (2005)—presents yet another very different explana-tion of key drivers of organizational benefits from ES. Itsthree antecedents of benefits are data quality, task efficiency,and coordination improvements. Task efficiency and coordi-nation improvements are indicators of the extent to which anES helps people in the organization get their jobs done, a keyconcept in our definition of functional fit. Data quality is alsoessential if management is to rely on reports derived from theES. In addition, Davenport et al.’s long-term benefit drivers(integration, process improvement, and informating) areimplicit in Gattiker and Goodhue’s model: its interdepen-dence–coordination–improvement link attributes benefits tointegration, its task efficiency implies that there are benefits tobe had from process optimization, and its data qualityvariable implies benefits from improved access to informa-tion. Empirically, using data from 111 U.S. manufacturing

plants, Gattiker and Goodhue found that their three ante-cedents explained 71 percent of the variance in plant-levelbenefits from ES (Table 11, p. 574). The highly significantpaths (coefficients of 0.52 and 0.36, both p < 0.001) for taskefficiency and data quality, respectively, show that theseattributes of what we have termed functional fit are importantdrivers of benefits from ES. As a general model, the limita-tion of Gattiker and Goodhue’s model is that it does notconsider explicitly the multi-project nature of ES ownershipshown in Figure 1, the possibility that different ES projects atthe one plant or at different plants might have very differentoutcomes, or the difficulties of overcoming organizationalinertia in each project.

Summarizing, the purpose of this section has been to demon-strate that the current ES research literature contains a numberof partial explanations of factors affecting organizationalbenefits from ES. What is missing is an integrated view. Itwould therefore seem valuable, we suggest, to try to integrateinsights from the various explanations in the ES literature—aswell as the prior IS literature on factors affecting benefitsfrom other types of information systems—into a coherentwhole. The OBES model in this paper provides one suchintegrated view.

Hypothesis Development: Synthesizingthe OBES Model from the Literature

This section describes how we used the extensive literature onorganization-wide investments in IT, including investments inES, to derive the six hypotheses in the OBES model. We alsoexplain why the various factors in the model contribute toorganizational benefits from ES. We commence by ex-plaining our choice of dependent variable.

Organizational Benefits from ES Use

Building on the work of Cameron and Whetten (1983),Grover et al. (1996) argue that success measurement does notmake sense unless the evaluator clearly defines the stake-holder from whose perspective success is to be evaluated. Inthe OBES model, the success perspective adopted is that ofsenior management (i.e., the top-management team).

The benefits of interest in OBES are similar to those reportedby Tallon et al. (2000), Davenport et al. (2002), Shang andSeddon (2002), Staehr et al. (2002), and Harris and Davenport(2006). For example, based on their most recent and largestglobal survey (371 executive respondents), Harris and Daven-port (p. 4) report that the most frequent ES benefits sought by

6Informate—a term they borrow from Zuboff (1988)—is defined as “trans-forming enterprise solutions data into context-rich information and knowl-edge” (Davenport et al. 2002, p. 6).

7Harris and Davenport (2006) do discuss project-related factors, including theimportance of change management (p. 9) and project management (p. 14), butthese factors are not included in their model, in Figure 2b.



management were (in descending order of frequency ofidentification by respondents) better management decisionmaking; improved financial management; faster, moreaccurate transactions; cost reduction; improved inventory andasset management; ease of expansion/growth and increasedflexibility; and cycle-time reduction. Note that the benefitsconstruct in OBES is conceptualized as before subtracting thecosts of implementing and running the organization’s ES. Ofcourse, the ultimate dependent variable of interest to seniormanagement is net benefit (DeLone and McLean 2003)—thatis, benefits less costs, but because costs are usually driven byfactors different from those that drive benefits, in this paperwe focus only on benefits and the factors that affect them.

Finally, the single benefits construct on the left of Figure 1represents benefits from all ES projects that have gone live inthe organization to date, whereas the benefits constructs onthe right are for benefits from each different project, post golive. Shang and Seddon suggest that a good way to assessorganizational benefits from ES is to ask business-unitmanagers about benefits in their part of the organization. Forthe long-term model, an appropriate question might be: Howsatisfied is your business unit with benefits from the organi-zation’s overall investment in ES? For the project side ofOBES, an appropriate question might be: How satisfied isyour business unit with the benefits from this ES project?

It is hypothesized in OBES that variance in organizationalbenefits from ES use—with benefits assessed from theperspective of senior management—is driven by variance ineach of the six independent variables discussed below.

Functional Fit

As defined in Table 1, functional fit is the extent to which thefunctional capabilities embedded and configured within an ESpackage match the functionality that an organization needs tooperate effectively and efficiently. Although an ES supports(and sometimes frustrates) the work of many individuals inmany parts of the organization, the unit of analysis chosen forassessing functional fit is the organization, not the indi-vidual.8 New functionality is delivered in each new majorbusiness ES implementation project, so although it ismeaningful to talk of the overall fit of ES-based software

systems (possibly implemented over many projects) for thebusiness, in OBES it is argued that functional fit should beassessed project by project.

Functional fit was selected as the first hypothesized keybenefit driver in OBES because organizations invest in ES fortheir functionality. Authors such as Dalal et al. (2004), Hongand Kim (2002), Rosemann et al. (2004), Soh et al. (2000,2003), Soh and Sia (2005), Scheer and Haberman (2000), andSomers and Nelson (2003)—see Table 2, row 3—have arguedrepeatedly that functional fit is a key goal in ES implementa-tion and, conversely, that misfit causes problems. Greaterfunctional fit helps a multitude of people across the enterpriseto play their part in the collective organizational endeavor.For example, an accounting clerk may use the ES to record apurchase invoice, an inventory clerk may use the ES to recordreceipt of raw materials, a marketing manager may use the ESto access details of last month’s sales to customer X, and abusiness-unit head may rely on monthly reporting to identifyprofit trends. Each of these people—and for many organi-zations there are thousands of such people—relies on enteringdata into, or retrieving data from, the system to do his or herjob. The greater the functional fit, the more efficient andeffective the organizational processes supported by the systemand the more the system helps users across the organizationget their jobs done.

The last 40 years of IS research and practice have shownrepeatedly the importance of achieving functional fit inorganization-wide applications of IT such as ES. Becausefunctional fit is so important, tools such as data-flow diagrams(DeMarco 1978; Gane and Sarson 1979), event-driven pro-cess chains (Scheer 1994), UML (Object Management Group2007), and business process execution language (BPEL;OASIS 2007) have been developed as increasingly sophisti-cated ways of representing the functional capabilities of acomputer-based system.

In practice, the importance to organizations of achievinggreater functional fit with packaged software is evident in therange of techniques that ES vendors have developed for tryingto help their customers achieve greater functional fit. Theseinclude configuration (changing parameters in various pre-defined tables); changing program code in various ways(Brehm et al. 2001); development of industry-specific solu-tions that extend base systems such as ERP by addingindustry-specific functionality (e.g., for the mining, retail,apparel and footwear, and banking industries); use of portalprograms to share access to multiple systems; use of today’semerging visual process-composition tools (Bönnen et al.2008); and use of data warehouses and reporting “front ends”(e.g., dashboards) to simplify information retrieval andreporting (Volitich 2008). The fact that so many methods

8Functional fit might be assessed by asking senior business-unit managersabout the efficiency and effectiveness of the overall processes supported bythe system, and by asking a random sample of individuals from within thebusiness unit how useful they find the ES to be in helping them do their jobs. When aggregating the second assessment, greater weight might be assignedto opinions of heavy users and those whose ES use has a greater impact onthe organization.



have been developed to try to meet various organizations’needs from their ES is clear evidence that functional fit ishighly valued by ES customers.

Empirically, as shown in row 3 of Table 2, studies such asthat of the Financial Executives International and CSC(2006), Hong and Kim (2002), Keil and Tiwana (2006), Siaand Soh (2002, 2007), Soh and her colleagues (Soh et al.2000, 2003; Soh and Sia 2005), and Somers and Nelson(2003) have provided evidence that functional fit is a keydriver of organizational benefits from ES.

In short, logical argument, evidence from practice, andfindings from many researchers over the past 40 years allpoint to the same conclusion: achieving functional fit betweensoftware capabilities and organizational needs is probably theprimary determinant of benefits from organization-wideapplications of IT such as ES. Further, since new func-tionality is delivered through major business ES implemen-tation projects, we hypothesize that

H1: The greater the functional fit (FF) resultingfrom each ES implementation project, thegreater the organizational benefits from ES use.

Overcoming Organizational Inertia

Overcoming organizational inertia (OOI) is defined inTable 1 as the extent to which members of the organizationhave been motivated to learn, use, and accept the new system.As for functional fit, the unit of analysis chosen for assessingOOI is the organization, not the individual. Further, sincedifferent change-management, training, and usability issuesarrive with each new major ES implementation project, it alsomakes sense to assess OOI project by project. The positiveoutcome of OOI is assimilation, defined by Purvis et al.(2001) and Liang et al. (2007) as the extent to which the useof technology diffuses across the organizational work pro-cesses and becomes routinized in the execution of those pro-cesses. Failure to overcome organizational inertia often mani-fests as resistance to the system (Lapointe and Rivard 2005).9

Overcoming organizational inertia was selected as our secondhypothesized key benefit driver for ES projects in the first few

years after go live because organizational change is sodifficult (Armenakis and Bedeian 1999; Kotter 1996) and nomatter how good the technical system, unless people in theorganization are motivated to use the system, and have suffi-cient knowledge of how to use the system effectively (Purviset al. 2001), the organization is unlikely to gain the benefitsit might from the system. Liang et al. (2007) appear to havechosen assimilation as the dependent variable in their studyfor similar reasons.

Just as the past 40 years of IS research and practice discussedabove testify to the importance of achieving functional fitwith organization-wide applications of IT, so the same 40years of research also demonstrate the importance ofovercoming organizational inertia in ES implementationprojects. As early as 1978, DeMarco was saying

The lesson of the 60’s is that no system is going tosucceed without the active and willing participationof users. Users have to be made aware of how thesystem will work and how they will make use of it.They have to be sold on the system. Their expertisein the business area must be made a key ingredientto system development. They must be kept aware ofprogress, and channels must be kept open for themto correct and tune system goals during development(DeMarco 1978, p. 6).

Similar evidence of the difficulty of effecting organizationalchange when new organization-wide applications of IT areintroduced has been echoed in the work of Keen (1981; socialinertia), Davis and Olson (1985; human and organizationalfactors), Joshi (1991, 1992; equity theory), Markus and Ben-jamin (1996; change agentry), Purvis et al. (2001; assimi-lation, Markus (2004; technochange), Boudreau and Robey(2005; improvised learning), and Lapointe and Rivard (2005;resistance to change). In the context of enterprise systems,authors such as Aladwani (2001), Lapointe and Rivard(2005), Markus and Tanis (2000), and McAfee (2002) (seerow 4(a) of Table 2), and Ko et al. (2005), Robey et al.(2002), Volkoff et al. (2004), and Staehr et al. (2006) (see row4(b) of Table 2) have similarly argued that no matter howgood the software, and no matter how well it has been con-figured and tested, unless people in the organization aremotivated to use the system, and in addition have sufficientknowledge of how to use the system effectively, the organi-zation is unlikely to gain its desired benefits from the system.

Summarizing, if organizations are to achieve benefits fromorganization-wide applications of IT such as ES, the peoplethe ES is intended to support must have sufficient knowledgeof the system to be able to use it effectively and be motivated

9It seems likely that OOI and assimilation will be easier to achieve if there isgood functional fit (due to greater benefits to individual users, whichmotivates them to use the system more), and that greater assimilation leadsto greater benefits. However, there is not space in this study to explore theseadditional propositions.



to use it. Further, the evidence from the past 40 years ofresearch shows that overcoming organizational inertia hasbeen a major obstacle in achieving benefit from many typesof organization-wide applications of IT. All this suggests thatOOI is so important that is should be selected as the secondkey driver, after functional fit, of benefits from ES. Thisargument is formalized in the following hypothesis:

H2: The greater the success in overcoming organiza-tional inertia (OOI) in each ES implementationproject, the greater the organizational benefitsfrom ES use.

Integration

As defined in Table 1, integration of information systems isthe unification of processes, systems, and/or data frommultiple computer-based systems, not necessarily in the oneorganization. A good example of a move to a more integratedsystem is the implementation of a single instance of SAP R/34.6c by Kraft Foods Inc. across 21 countries in Europe duringthe period 2002–2004. According to Ziskasen (2008), theresulting single instance of SAP ERP supports “harmonizationof business processes across Europe” for 11,000 usersprocessing three million sales orders for U.S. $7 billion inrevenue per year.

ES provide two distinct types of integration: back-end andfront-end integration. Back-end integration shares data acrossapplications through use of four mechanisms: a commondatabase; real-time inter-system messaging, for example,through use of enterprise application integration (EAI)middleware (Linthicum 2000, Lee et al. 2003); systems thatextract, transform, and load data from various sources intoone or more data warehouses; and old-fashioned batchupdates. Front-end integration provides a common user inter-face that simplifies user learning and use of a system throughthree mechanisms: standards that define how applicationsshould look and feel (e.g., the standard menus and icons usedin the Microsoft Office suite); portal programs that enableuser organizations to build their own user interfaces; andintegrated reporting engines that use standard report-specification languages to support powerful data analytics(often called business intelligence), search, and dashboards bydrawing data seamlessly from many different systems. Thesemethods are often combined. Historically, for instance, ERPsystems have used a single database and a standardized userinterface for both transaction entry and reporting.

Integration was selected as our third hypothesized key benefitdriver in OBES because it is a frequently identified distin-

guishing feature of ES (Davenport 1998; Deloitte 1998;Markus and Tanis 2000) that often enables process optimi-zation and/or improved access to information. Integration canlead to benefits via four distinct pathways. First, organiza-tions usually find it simpler to work with one accurate versionof the truth than with data from many distinct computersystems. In an integrated system, less human effort is wastedresolving uncertainty about the accuracy and comparability ofinformation sourced from various systems10 and updatingmultiple systems. Second, as discussed in relation to H4below, the end-to-end visibility enabled by real-time integra-tion often provides the foundation needed for processoptimization (Barki and Pinnsoneault 2005; Kohli 2007).Third, as discussed in relation to H5 below, greater informa-tion visibility enabled through all forms of integration (notjust real-time integration) can help improve decision-makingby reducing the time and effort required to discover andaccess valuable information (e.g., that might previously havebeen hidden in different systems). Fourth, systems with acommon user interface are easier to learn and use thansystems that require users to learn different interfaces fordifferent systems. Such front-end integration should make iteasier to overcome organizational inertia (H2).

The following two studies provide empirical evidence ofbenefits flowing from ES-enabled integration. First, Daven-port et al. (2004)—in their previously mentioned 2001–2002survey of 163 large mainly U.S. and European organizations(see Figure 2b)—report a small but significant path coeffi-cient of 0.19 between integration and organizational benefits.Second, in their event study of stock-market reactions to 116announcements of ERP implementations in the United Statesduring the period 1997–2001, Ranganathan and Brown (2006)found that firms that announced ERP projects with two ormore value-chain modules or spanning multiple sites (i.e.,more integrated systems) had significantly positive cumula-tive abnormal returns during the announcement window.

Although it is not possible to explore empirically the fol-lowing proposition in this paper, evidence is beginning toemerge that integrated computer-based systems are mostbeneficial when they link interdependent parts of an organi-

10It is easy to underestimate the effort required to reconcile data fromdifferent systems (e.g., accounting reports from two companies that usedifferent charts of accounts) or lists of customers in two separate billingsystems (e.g., fixed line and mobile) run by one telecommunications provider.In an integrated system, this reconciliation effort has been done and processesare in place so that future reconciliation effort is not required. The entropyof the integrated system is therefore lower. This appears to be an importantreason why managers prefer integrated over non-integrated computer-basedsystems. (Thanks to Prithvi Bhattacharya for discussion of this point.)



zation (i.e., parts of an organization that need to shareinformation). Gattiker and Goodhue (2005), for instance (seeFigure 2c), argued that manufacturing plants that were moreinterdependent would benefit more from ERP-enabledintegration than plants that were less interdependent. Empi-rically, using data from 111 U.S.-based manufacturing plantsusing ERP systems, they found a very high association (pathcoefficient of 0.81) between level of interdependence betweenplants and improvements in coordination achieved throughuse of a shared ERP system. Likewise, Ross et al.’s (2006)key 2 × 2 classification of operating models (p. 39) suggeststhat only organizations with so-called coordination andunification models will benefit strongly from ES-enabledintegration. Modeling and testing the conditional propositionsdiscussed in this paragraph is an interesting topic for futureresearch, although beyond the scope of this paper.

Summarizing, the above arguments suggest that integration isa key enabler of benefits from organization-wide applicationsof IT such as ES. It is, therefore, hypothesized that

H3: The greater the level of ES-enabled integration,the greater the organizational benefits from ESuse.

The primary mechanism for increasing integration is throughmajor ES improvement projects (not just ES businessimprovement projects).

Process Optimization

Process optimization is defined in Table 1 as any attempt toimprove the efficiency and effectiveness of an organization’sprocesses. ES can be used to improve processes in four ways.First, ES are highly configurable business-process platforms(Keller and Teufel 1998; SAP 2008; Scheer 1994). Thismeans that user organizations can improve processes by con-figuring, reconfiguring, and extending ES-supported pro-cesses with relative ease.11 Second, by working with keyclients to build software to help integrate, optimize, andstandardize their processes, ES vendors offer both their keyand non-key customers the promise of access to best practiceprocess templates from other leading organizations. Forexample, SAP recently worked with Ford and CaterpillarLogistics to develop what they claim is a best practicelogistics system to manage spare parts distribution for theautomobile and earth-moving machinery industries (SAP2006). Importantly, these process options are now available

to all of SAP’s logistics-solution customers. Third, ES-enabled increased visibility into an organization’s processesallows better coordination of processes, leading, for example,to reduction of buffer stocks of inventory that were previouslynecessary due to difficulties in coordinating various parts ofa business. Inventory reduction is a frequently reported bene-fit of ERP systems (Davenport et al. 2002). Fourth, becauseof their integration and cross-checking, ES often imposegreater control on process execution than was possible withearlier systems. In most organizations, senior managementregards such control as a good thing (Harley et al. 2006).

Process optimization was selected as our fourth hypothesizedkey benefit driver in OBES because ES can be used to supportprocess improvement, and process improvement has beenshown to be a major driver of organizational benefits. Theextensive literature on total quality management (TQM)(Garvin 1988; Samson and Terziovski 1999; Walton 1986),business process reengineering (Davenport 1993; Hammer1996, 2007; Hammer and Champy 1993), and today’s Six-Sigma projects (Breyfogle 2003), all argue that processimprovement is an important way to achieve greater organiza-tional benefits. Process optimization is regarded by the ITindustry as so important that it has been the top-rankedbusiness priority in Gartner’s (2008) global surveys of over1000 CIOs for the past four years.

Empirically, the literature on TQM and BPR summarizedabove shows that process optimization can be an importantdriver of organizational benefits. With respect to ES, Daven-port et al. (2002, 2004), Gattiker and Goodhue (2005), andStaehr et al. (2006)—three of the ES papers highlighted inrow 6 of Table 2—all report that process optimization was animportant driver of benefits from ES.

In short, since process optimization has been shown to be animportant benefit driver in so many studies, and ES are knownto be particularly helpful for process optimization, it ishypothesized that

H4: The more steps taken to improve ES processoptimization, the greater the organizationalbenefits from ES use.

The primary mechanism for improving process optimizationis through major ES business improvement projects, discussedin H6 below.

Improved Access to Information

Improved access to information, a synonym for Davenport etal.’s (2002) informate, is defined in Table 1 as any step taken

11Visual process composition tools based on services-oriented architecturepromise to make this even easier in the future (Bönnen et al. 2008).



to increase the provision of timely, accurate, relevant infor-mation (including previously hidden information) to keyorganizational decision makers. Improved access to relevant,accurate information leads to increased organizational bene-fits because it enables decision makers to make betterdecisions (Kohli 2007; Kohli and Grover 2008), possibly evencontributing to competitive advantage (Davenport and Harris2007).

Improved access to information was selected as our fifthhypothesized key benefit driver in OBES because ES-enabledimproved decision making appears to be one of the mostfrequently reported and highly valued outcomes of ES use.For example, in their 2001–2002 global survey of 163 ES-using organizations, Davenport et al. (2002) reported thatimproved decision making was much valued:

Organizations in our Accenture survey desiredimproved decision making more than any otherbenefit of enterprise solutions. Driven by the desirefor accurate, consistent, complete, real-time informa-tion, executives are seeking the same type of effi-cient, transparent and “frictionless” real-time deci-sion making capability that many manufacturersachieved with just-in-time manufacturing (p. 21).

Harris and Davenport (2006) reported this same finding intheir 2005 global survey of 371 ES-using organizations.Further, the link between ES use and better decision makingappears to be causal. Harris and Davenport, for instance,report that 59 percent of organizations identified “analyticsfor decision making” as a distinctive and useful capabilitydelivered by their ES (p. 14). Moreover, and consistent withGartner’s 2008 report that the most frequently identifiedtechnical priority for CIOs in the past three years has been“business intelligence applications,” the recent spate of take-overs of vendors of “business intelligence” software12 by theworld’s major ES vendors suggests that the ES vendorsbelieve that adding greater capacity to access relevant infor-mation will further enhance their product suites.

Based on the knowledge that provision of accurate, relevantinformation has been a primary focus of much IS research

(Davis 1974; Keen and Scott-Morton 1978; Wixom andWatson 2001) and success measurement (DeLone andMcLean 1992; Ives et al. 1983) for over 30 years, plus theabove empirical evidence that ES enable much-valuedimproved decision making, it is hypothesized in OBES that

H5: The higher the level of ES-enabled improvedaccess to information, the greater the organiza-tional benefits from ES use.

As with H4, the primary mechanism for improved access toinformation is through major ES business improvementprojects, discussed in the next section.

On-Going Major ES BusinessImprovement Projects

The final construct in the OBES model, on-going major ESbusiness improvement projects, is defined in Table 1 as a mea-sure of the number and extent of investment in major businessimprovement projects that an organization has undertaken forimproving and extending its enterprise system. There are fourkey attributes of this construct. First, in formulating OBES, welimited the on-going projects construct to major businessimprovement projects only. This excludes infrastructure pro-jects and technical upgrades that may lead to reduced cost, butdon’t deliver new functionality to the business. Our interest isin on-going major business improvement projects becausethese are the projects that deliver significant new functionalityto users (and typically involve the need for additional training,change management, and support).

Second, investments in these major business projects seem tobe the key driver of increased value from ES over the longterm (as depicted in the rising long-term benefits graph inFigure 1). Assuming management invests in projects thatpromise positive net present value, the greater the investmentin these projects, the greater the expected organizationalbenefits from ES. For example, if a firm invests a total of $30million in major ES projects in one year (in one large or anumber of smaller projects), it might expect to receive, say,$10 million in benefits each year for the next four years,possibly more. As additional investment continues year afteryear, benefits per year would be expected to rise.

Third, increased organizational maturity in running ESprojects and in using ES should also lead to increased benefitsfrom later compared to earlier projects. Although thelearning-curve effect is unlikely to be as pronounced as, say,in manufacturing (Argote and Epple 1990), learning by boththe ES project team and the organization overall about how toimplement software—which is not the same as the learning

12In April 2006, Microsoft purchased ProClarity for an undisclosed sum. InFebruary 2007, Oracle purchased Hyperion for U.S. $3.3 billion. In October2007, SAP announced the purchase Business Objects for U.S. $6.7 billion,saying “SAP and Business Objects believe that customers will gainsignificant business benefits through the combination of new, innovativeofferings of enterprise-wide business intelligence solutions along withembedded analytics in transactional applications.” In November 2007, IBMannounced the purchase of Cognos for U.S. $5 billion. Finally, in July 2009,IBM announced its intention to purchase SPSS for $1.2B.



how to use the software (H2)—means that an organizationthat has completed, say, five projects is likely to do a betterjob with its sixth project than an organization that is about tocommence its first. Thus it is expected that organizationalbenefits from ES will be correlated with both the number andscope of completed projects.

Finally, because integrating, optimizing, and improved accessto information (IO&I) discussed in H3, H4, and H5 areactually implemented through projects, one would expectconsiderable correlation between measures of IO&I and on-going major ES business improvement projects. However therelationship between these variables is not simple becauseintegration may be an enabler of optimization and improvedaccess to information in some projects but not others, andsome projects may be in pursuit of goals other than IO&I(e.g., compliance with changes in government regulations).

Summarizing, on-going major ES business improvementprojects was selected as our sixth and final key benefit driverin OBES because the benefits achievable from integration,optimization, improved access to information, and other goals(e.g., compliance with government regulations) are actuallyrealized through on-going improvement projects that delivernew functionality to users. Further, as discussed earlier, ifmanagement invests wisely, the greater the investment in suchES improvement projects, the greater the benefit the organi-zation should expect from ES. Therefore, and consistent withthe views expressed in Davenport et al. (2002, 2004) andStaehr et al. (2006), it is hypothesized that

H6: The greater the investment in on-going majorES business improvement projects, the greaterthe organizational benefits from ES use.

The OBES Model in a Nutshell

The OBES model in Figure 1 is a synthesis of many ideasfrom the literature, particularly those from the ES literaturecited in Table 2. It is an attempt to identify the most impor-tant factors that drive benefits from ES over many years of ESownership, and to place them as logically as possible in anomological model. In particular, its structure reflects themulti-project nature that seems to be a key characteristic ofES ownership today.

Focusing on the right-hand (projects) side of the model, thereare strong logical grounds, and strong support in the IS litera-ture, for the view that both functional fit and overcomingorganizational inertia are key drivers of benefits from an ESin the first few years after go live. Further, since functionalfit means the system has the required functionality, and

overcoming organizational inertia means that people in theorganization are motivated and able to use the new systemonce it has gone live, good scores on these two factors appearto be all that is necessary for producing strong organizationalbenefits from an ES. We therefore expect these two factorsto be the primary determinants of benefits from all ESimplementation projects.

In addition, the left-hand side of the OBES model attempts toconvey the idea that on-going efforts to integrate, optimize,and provide improved access to information (the factors fromDavenport et al. (2002, 2004)) lead an organization to under-take various major ES improvement projects. As each projectgoes live, the high-level goals of integrating, optimizing, andproviding improved access to information are translated intoreal functionality supporting real processes that help or hinderpossibly thousands of individuals in doing their (possiblynew) jobs.

Note that in formulating OBES we chose not to include themany so-called critical success factors (CSFs) that affect anES project team’s success in delivering a full-scope, thor-oughly tested working system, on time, and within budget. Most of these CSFs (e.g., the 55 identified in Finney andCorbett 2007), appear to be antecedents of the variables onthe project side of the OBES model. For example, based onanalysis of 133 customer presentations from Sapphire USA2007, Liu and Seddon (2009) report that having a “balancedteam” of knowledgeable business users and IT staff was a keyfactor in achieving functional fit, and having a good “commu-nication plan” was a key factor in overcoming organizationalinertia.

Also, in formulating OBES we chose not to explore relation-ships between the six independent variables. For example, thediscussion above suggests that integration is often a keyenabler of both process optimization and improved access toinformation, greater functional fit reduces difficulties in over-coming organizational inertia, and it is the combination of FFand OOI (subject to successful go live) that matters forachieving benefits from ES implementation projects. Explo-ration of these and other extensions to OBES has been left forfuture research.

Obviously, since the OBES model is concerned with verycomplex organizational phenomena, it is not suggested thatthe six factors in the model explain all the variance in organi-zational benefits from ES use. Nor is it suggested that thefactors in the OBES model are new; they are not. The reasonfor citing so much of the early IS literature in arguing theimportance of the OBES factors was to make the point that ISresearchers have known about the importance of these successdrivers for organization-wide applications of IT for many



decades. It is argued, however, that OBES is a significantcontribution to the ES literature because no prior study (e.g.,those in row 6 in Table 2) has proposed a model with as muchexplanatory power as OBES. Although many of the studiesin Table 2 have explored aspects of issues related to obtainingbenefits from ES in more depth than in this paper, thecontribution of this paper is to have assembled in one place amodel that has more descriptive power than any prior paper.

Finally, note that unlike studies such as Taylor and Todd(1996) and Rai et al. (2002), which rely on the extent ofvariance explained to choose between competing models, ourclaim that OBES explains more than the studies in Table 2 isbased on logic. Our claim is that since it is possible to pointto important phenomena associated with deriving benefitsfrom ES that the prior models in row 6 of Table 2 do notaddress, yet which OBES does, the OBES model has moreexplanatory power than any of them.

Of course it remains to be seen whether empirical support forthe OBES model is as strong as the literature review andtheoretical considerations above suggest it should be. Theremainder of this paper, therefore, presents details andfindings from a preliminary test of the OBES model inFigure 1. The purpose of this test is to provide a reality checkon the meaningfulness of the hypotheses, the overall modelstructure, and to make a preliminary assessment of thestrength of support for the model.

Research Methodology

The method chosen for conducting our preliminary test of theOBES model in Figure 1 was qualitative (content) analysis bytwo pairs of coders of a total of 130 customer presentationsfrom two of SAP’s Sapphire conferences: 60 from SapphireUSA 2003, and 70 from Sapphire USA 2005. As explainedin the “Introduction,” Sapphire conferences are organizedannually by SAP, the world’s largest vendor of enterprisesystems, both to inform their customers of new productdevelopments and for their customers to exchange informa-tion about their experiences with SAP software. The detaileddiscussion that follows focuses on our analysis of data fromSapphire 2003, but an equally rigorous—though less fine-grained—analysis was also conducted of presentations fromSapphire 2005. Results from both analyses are presentedshortly.

At Sapphire 2003 there were over 100 presentations fromsenior business and IS managers from customer organizationssuch as Adidas, Audi, Barclays Bank, Bosch, Chevron

Texaco, Disney, Hershey Foods, Lockheed Martin, Shell,Sony, and Texas Instruments. Streaming video of each 45-minute presentation, together with PowerPoint slides and fulltranscripts of each presentation, were available from the SAP“community” website13 for some months after the conference.From the above-mentioned 100-plus customer presentations,we selected all 60 presentations that discussed either or bothof benefits realized from the enterprise system, and projectsuccess factors.14 The organizations are quite large. Of the30 organizations that reported revenues in their presentations,27 had 2002 revenues above U.S. $1 billion per annum.Combined, the presentations are a very rich source of infor-mation about the goals, issues, and outcomes of ES projectsin large organizations. Sapphire 2005 was similar: there werelots of presentations from many large customers, each withdetailed and interesting stories of their experiences with theirSAP software.

Sapphire presentations relate to all of SAP’s product lines,including enterprise resource planning (ERP), customerrelationship management (CRM), supply chain management(SCM), and data warehousing (BW). The presentations them-selves—the data for this study—involve 1,630 PowerPointslides and 434 single-spaced pages of transcribed presenta-tions from Sapphire 2003, and 2,056 PowerPoint slides and600 single-spaced pages of transcribed presentations fromSapphire 2005. Although the weighting of topics in eachpresentation varies enormously, the “typical” presentationdescribes the business, the reasons for selecting and imple-menting the software, the organization’s application architec-ture, time lines and staffing for the project, details of someaspect of the functionality of the software, benefits from theproject, and lessons learned.

Appendix A lists the organizations and presenter roles in all130 presentations initially selected for analysis. Businessmanagers comprised 34 percent of presenters, with CIOsrepresenting 22 percent and senior ES managers or projectmanagers representing 45 percent. As shown in Appendix A,speakers in four Sapphire 2005 presentations discussed essen-tially the same topic as in 2003, although at later stages ofdevelopment. To avoid double counting, these four presen-tations were ultimately dropped; that is, the percentages in theTable 5 for 2005 are based on a sample of 66, not 70, presen-tations.

13Http://www.sap.com/community.

14Presentations excluded did not discuss topics relevant to the model. Examples of excluded presentations were where customers discussed projectsthat had not yet gone live or technical issues such as their experiences usingvarious SAP tools such as Solution Manager.



The 60 sets of Sapphire 2003 PowerPoint presentations andtranscripts were content analyzed independently by coders 1and 3 using the definitions in Table 1 and the strength-of-evidence (SoE) criteria defined in Table 4. Examples of SoEjudgments assessing SoE as 1, 2, and 3 for H3 (Integration)are presented in Table 3. Examples of SoE judgments for theother five hypotheses are included in Appendix B. Using onerow per presentation, and one pair of columns for each of thesix OBES factors (H1-H6), slide numbers and transcript pagenumbers were recorded in a spreadsheet when evidence oflinks to benefits was identified. Strength-of-evidence judg-ments from the two coders were then compared anddifferences reconciled through discussion until no differencein SoE scores exceeded 1.15 The lower half of the analysisspreadsheet showing details for cases 25 through 60 andcolumn totals (rows 63–65) from coder 1 after discussion andreconciliation is shown in Figure 3. Scores were identical 85percent of the time. Eight percent of the time, coder 1’sscores were higher than coder 3’s; eight percent of the time,coder 3’s scores were higher than coder 1’s. In other words,the disagreements were not systematically biased one way orthe other.

A similar process to that described above was used foranalyzing the Sapphire 2005 presentations. First, coders 1and 2 independently coded all 70 customer presentations(including the four later dropped) using a binary “evidence/noevidence” classification.16 Their initial level of agreementwas about 70 percent of 70 × 7 = 490 judgments.17 Second,after about ten hours of discussion, the two coders achieved100 percent agreement in their identification of presentationswhere there was at least some support (i.e., where SoE was atleast 1) for each hypothesis. This analysis of the 2005 data

was actually done before the analysis of the 2003 data.Further, since it takes about two weeks of intense work fromtwo people to analyze a set of Sapphire presentations, and theoverall results from the two conferences both provide con-siderable support for all six hypotheses (see columns (f) and(g) in Table 5), it was decided that there was little benefit tobe gained by reanalyzing the Sapphire 2005 results using thefiner-grained SoE = 1, 2, or 3 classification used for Sapphire2003. Therefore, Table 5 presents results using two differentgranularities of analysis.

Results

Results from the analyses of customer presentations from bothSapphire 2003 and 2005 are presented in Table 5. Rows 1through 6 are for H1 through H6. Row 7 reports frequency ofdiscussion of efforts to achieve successful go live.18 Columns(b) through (f) summarize the 60 × 7 = 420 pairs of SoE judg-ments made by coders 1 and 3 during the analysis of Sapphire2003. Column (b) shows the Gamma statistic (Siegel andCastellan 1988) measuring inter-rater correlation for SoEcoding for each hypothesis after reconciling opinions;agreement is high. The percentages in columns (c) through (f)are the averages of percentages from coders 1 and 3. Column(f) shows total percentages (SoE scores of 1, 2, or 3) wherethere was any support for the six hypotheses (rows 1–6) orevidence of interest in achieving successful go live (row 7).Column (g) shows results from coder 1’s and coder 2’sanalysis of the 66 Sapphire 2005 presentations. These percen-tages are directly comparable to those from Sapphire 2003 incolumn (f). Finally, we use the fact that no firm wouldembark on an ES implementation project without expecting tosuccessfully go live, yet only 78 percent of 2003 presentationsand 53 percent of 2005 presentations discussed implemen-tation project issues (see row 7 of Table 5), to support ourclaim that the percentages in Table 5 probably also understatethe importance of each of the six OBES factors (H1–H6) inthe various organizations.

Since the percentages in columns (f) and (g) of Table 5 are sohigh, it may be concluded that the evidence from these 126Sapphire 2003 and 2005 presentations provides support for allsix hypotheses depicted in Figure 1. To be more precise,

15As explained in Appendix B, coding strength-of-evidence scores for thepresentations requires much judgment (e.g., is the evidence in this presen-tation strong or just moderate?). For this reason, there was initially only 40%agreement between coders 1 and 3. Coder 1 was lower than coder 3 in 38%of judgments, and higher in 22%. After discussion, 47% of coder 1’s SoEscores were revised up and 27% were revised down, and 8% of coder 3’sscores were revised down and 15% were revised up. In 15% of cases, afterconsidering each other’s arguments—and provided the difference in SoEscores was no more than one—the two coders agreed to disagree.

16Robey et al. (2002, Table 2, p. 26) use a similar dual-rater, binary,frequency-of-mention technique for assessing the importance of variousmotivations for implementing ERP.

17Seven, not six, judgments were made for each of the 70 cases because inaddition to looking for evidence of support for each hypothesis, coders 1 and2 also recorded evidence of interest in achieving successful project go live. These data are used in the “Results” section to support our claim that that thestatistics in Table 5 probably understate the importance of the six key factorsfor driving benefits in the study organizations.

18Examples of strength of evidence judgments of project management interestin delivering a working system at project go live (necessary for realizingorganizational benefits from ES) are shown in Appendix C.



Table 3. Example of Strength-of-Evidence Judgments for H3*

Factor SoE ExampleExplanation of Strength-of-Evidence

Classification

3. Integra-tion

1 “Our transportation analysts and customer service reps are moreeffective because they’re not involved nearly as much in trying toresolve issues around freight rating. Our process integration withR/3 allows for much better visibility into the rating process. Weno longer have this black box interface between two systemswhere we really don’t totally understand what transpired. Wenow understand the process and can see it. And we have builtcommon transportation processes and systems in North Americaand we have applied this capability to one of our businesses inEurope.” (case 6, transcript p. 5)

In this quotation, the vice president of Logistics andIT from Armstrong World, a global manufacturerand marketer of interior furnishings employing14,000 people, explains that messaging-basedintegration of SAP R/3 with Rand McNally’s Mile-maker has reduced the level of manual interventionrequired to determine freight charges. On the scalefrom 1 to 3, the strength of this evidence thatintegration leads to organizational benefits from ESuse was judged to be limited ( i.e., 1).

2 “We did also an application integration with online ordering. Thisis not an SAP-system, by the way, because it is not implementedyet. We integrated this because of two specific functions whichis team-ware and employee sales. Customer service departmentalso does employee sales. This online ordering system is also,of course, integrated or connected to AFS R/3.” (case 3, tran-script p. 2) (Note: AFS is SAP’s Apparel and Footwear IndustrySolution.)

In this quotation, the CIO from Adidas, a U.S.$6.5 billion sporting goods firm, explains whyAdidas integrated their CRM solution with an onlineordering and SAP’s R/3 AFS, using messaging. Since this integration with AFS was “of course”desirable, the strength of this evidence thatIntegration leads to organizational benefits from ESuse was judged to be moderate (i.e., 2).

3

Copyright © by Coca Cola Enterprises Inc Reprinted by permission of Coca-Cola

Enterprises. All rights reserved.

In discussing this slide, the speaker from Coca ColaEnterprises, explained that her firm used SAP’sdata warehouse (the server at the top of the dia-gram) to integrate information from multiple differentsystems (the four servers at the bottom of the dia-gram) to reduce purchasing expenditure. Since thiswas obviously a major integration effort, whichwould not have been conducted unless Coca ColaEnterprises believed it would lead to benefits, thestrength of this evidence that integration (in thiscase using a data warehouse, not real-time mes-saging between systems as in the above twoexamples) leads to organizational benefits wasjudged to be strong (i.e., 3).

(Aside: Since the data warehouse provides accessto much richer information for many people, e.g.,slide 33 says “Never had access to this sort of infor-mation before,” this slide was also one of nine in theCoca Cola Enterprises presentation that led to anSoE rating of 3 for H5, the hypothesis that improvedaccess to information is a driver of organizationalbenefits from ES use.)

*Similarly detailed examples of SoE judgments for the five other hypotheses are presented in Appendix B. A total of 60 × 7 = 420 of these SoEjudgments were made independently by the two raters, then reconciled to a difference not exceeding 1.

CCE’s Current Implementation - Continued

© C

om

pa

ny

Nam

e / S

AP

AG

20

03,

Titl

e o

f P

rese

nta

tion

, Sp

eake

r N

ame

/ 2

2

Upload from Flat Files

EBP RS/6000

INFINIUM AS/400

SAP R/3 RS/6000

Source Header ODSSource Item ODSSource Schedule Line ODSDetail Reporting ODS for Indirect Purchase

Freight Detail ODS

Catalog ID ODSShopping Cart ODSEFORMS ODS

Plant X- Ref ODSExpense Distribution ODS

Source Header ODSSource Item ODSSource Schedule Line ODSDetail Reporting ODS for Indirect PurchaseFreight Detail ODSFreight Indicator ODSCatalog ID ODSShopping Cart ODSEFORMS ODSGR Notification ODSPlant X-Ref ODSExpense Distribution ODS

Catalog ID

Shopping Cart Header

Shopping Cart Item

GR Notification

EFORM

National Vendor #Budget

Tax CubeIndirect Purchase CubeDirect Purchase CubeGR Notification Cube

Freight CubeStatistics Cube

Tax CubeIndirect Purchase CubeDirect Purchase CubeGR Notification CubeFreight CubeStatistics Cube

Master Data TextMaster Data Attribute

Master Data Hierarchy

Master Data TextMaster Data Attribute

Master Data Hierarchy

BASIS AS/400

Freight DetailFreight indicatorPO HeaderPO Item PO Schedule LinePlant X-Ref TaxExpense Distribution

Other Source Systems

BW RS6000

SAP Server



Table 4. Criteria Used for Scoring Strength of Evidence in the Qualitative Context Analysis

Criterion SoE Score

No evidence concerning the importance of the factor in the presentation 0

Factor mentioned explicitly but in passing (e.g., by one dot point on one slide) or implicitly, but notdiscussed at length (see example 1 in Table 3)

1

Factor clearly identified as a benefit driver (e.g., by a full slide) or by one or more paragraphs ofthe transcript (see example 2 in Table 3)

2

Strong evidence of the importance of the benefit driver (e.g., discussed at length) or the point wasemphasized by speaker (see example 3 in Table 3)

3

Figure 3. Lower Half of Coder 1’s Analysis Worksheet (Cases 25 through 60)*

*Three examples of strength-of-evidence decisions (see Table 4) are presented in Table 3. Table 5 shows the average percentages from tehbottom three rows of this and coder 2’s equivalent spreadsheet.



Table 5. Summary of Evidence Supporting the Six Hypotheses

Hypothesis

Sapphire 2003 DataSapphire2005 Data

Inter-ratercorrelation

postreconciliation (Gammastatistic)

Percent of the 60 presentations containingevidence supporting H1–H6 (and project go

live)

% of the 66presentationswith SoE > 0(supportingH1–H6 and

project go live)

Mentionedas a factor(SoE = 1)

Clearlyidentified(SoE = 2)

StrongEvidence(SoE = 3)

Total %(SoE = 1,2, or 3)

(a) (b) (c) (d) (e) (f) (g)

1. Functional fit 0.96 13 44 37 94 95

2. Overcoming organizational inertia 0.97 18 26 28 73 52

3. Integration 0.99 18 33 44 94 95

4. Process optimization 0.99 25 47 14 86 88

5. Improved access to information 0.98 18 14 12 43 67

6. On-going projects 1.00 28 40 7 75 68

Project go live 1.00 9 38 32 78 53

Notes:1. “SoE” stands for strength-of-evidence. Definitions of SoE = 0, 1, 2, and 3 are given in Table 4.2. Achieving successful project go live may be assumed to be a goal for all ES implementation projects, but was only discussed explicitly in 78%

and 53% of presentations in 2003 and 2005, respectively. This suggests that the percentages in this table probably also understate theimportance of the factors in H1 through H6.

1. The evidence is strongly consistent with the view thatorganizations invest in ES to help them execute somepart of their business processes more effectively. That iswhy software functionality (H1), integration (H3), andprocess optimization (H4) were discussed so often.

2. The evidence also suggests strongly that two major diffi-culties organizations face in gaining benefits from theirinvestments in ES are overcoming organizational inertia(H2) and successfully going live (row 7).

3. The evidence is consistent with the view that organiza-tions invest in ES to help them gain access to better infor-mation for decision making (H5). Further, and consistentwith the earlier quotation from Davenport et al. (2002),comments such as “outrageously beneficial” and “themost accurate, timely, actionable data that the companyhas seen in years” from some presenters (see theexamples of the SoE analysis for H5 in Appendix B)show that improved access to information is a veryimportant source of benefits for some organizations.

4. ES ownership usually involves an ongoing series ofmajor business projects (H6).

Discussion and Limitations

In assessing the results reported above, the first question onemight ask is whether it is valid to use presentations fromSapphire conferences as a source of data for testing a variancemodel such as that presented in Figure 1. On the negativeside, (1) all firms have a vested interest in showcasing thepositive aspects of their products, so it is likely that SAP wasselective in choosing the topics for presentation at Sapphire,(2) the speakers would have felt that their role as Sapphirepresenters obliged them to focus on the good outcomes oftheir implementations and to gloss over the problems theirorganizations had experienced as they implemented theirsystems, and (3) the presenters would have had a tendency topresent themselves and their firms in a good light.19 Thismight mean that conclusions based on content analysis of thepresentations reveal little of value about the factors that reallydrive benefits from ES. On the positive side, as is apparentfrom the examples in Table 3 and Appendix B, the Sapphire

19Self-reporting bias is an issue in all single-informant studies (e.g., surveys),but is likely to be stronger at a Sapphire conference than, say, in ananonymous survey, because respondents and their firms are “on display” ina highly visible setting.



presentations provide rich and detailed first-hand accountsfrom senior managers of their experiences in a wide range ofmajor corporations, the presenters knew they were talking toknowledgeable peers, which would have tended to keep themfrom “stretching the truth” too far, and some of the commentsin the presentations are quite frank about difficulties theirorganizations experienced.

After weighing the pros and cons above, we have concludedthat the evidence from the Sapphire conferences is useful forresearching in the area of benefits from ES and benefitdrivers. The strongest evidence to support this conclusion isthat, in a parallel study, Yang and Seddon (2004) analyzed thesame 60 Sapphire 2003 presentations to identify benefits fromES and project critical success factors (CSFs). The resultantlists of benefits and CSFs are very similar to those reported byprior authors—Deloitte (1998), Davenport et al. (2002), andHarris and Davenport (2006) for benefits, and factors from the45 studies summarized in Finney and Corbett (2007) forCSFs—yet those studies used a range of different researchmethods. Since results from content analysis of these sameSapphire presentations produced results in two closely relatedtopics so similar to those of prior researchers, it seems reason-able to conclude that the results from the content analysisreported in this paper may also be trusted.

A second question is whether the model in Figure 1 is a validrepresentation of cause and effect for the phenomena de-scribed in the 126 presentations. Here, we acknowledge thatthere are undoubtedly other factors not explored in thisstudy—for example, information quality (Gattiker and Good-hue 2005), organizational context, or alignment of the ES withbusiness strategy (Lee and Myers 2004; Somers and Nelson2003)—that may be important drivers of organizational bene-fits from ES, and, as discussed earlier, there may be relation-ships between the factors (e.g., greater integration probablyenables greater process optimization and improved access toinformation). Unfortunately there is not space to addressthese issues in this paper.

We can, however, make two positive claims about the validityof the OBES model. First, because it was possible to explorecausality at the individual-case level, we believe that the sixrelationships described by the OBES model are both impor-tant and causal. Second, because most firms undertake mul-tiple ES-improvement projects (see row 6, Table 5), and thefunctional fit and overcoming organizational inertia outcomesof each project could differ markedly from project to project,the two-part, multi-project structure of OBES is important.

Since most of the organizations studied were very large andfrom a wide range of industries, and causal arguments sup-

porting the model in Figure 1 do not appear to be either SAP-specific or peculiar to the organizations studied, the resultsreported here are likely to be applicable to other large organi-zations using either SAP or non-SAP enterprise-system soft-ware in Western-style organizations around the world. How-ever, because the needs and resources of small organizationsare so different, we are not confident that the results above areapplicable to small Western-based organizations. Nor are weconfident that they apply to large or small organizations incountries such as China with highly collectivistic decisionprocesses (Hofstede 2001).

With regard to the possible applicability of OBES toorganization-wide applications of IT other than ES (e.g.,custom-built software), or software such as corporate e-mailsystems, we are also very cautious. After the implementationof, say, a corporate e-mail system, logic says that functionalfit and overcoming organizational inertia will be important.However, e-mail systems do not constrain the processesthrough which people in an organization interact, as ES do.Nor is it clear that either integration with other systems isimportant, or that concepts like process optimization andaccess to better information for decision making (e.g., throughbusiness intelligence) have any meaning in an e-mail context.Thus we do not claim that OBES is applicable to allorganization-wide applications of IT.

Conclusion

The objective of this study was to synthesize from the litera-ture and conduct a preliminary test of a model of keyenterprise system (ES) success factors that reflects the multi-project reality that organizations today all seem to face as theyseek to derive increasing benefits from their enterprisesystems. The two-part organizational benefits from ES(OBES) model in Figure 1 summarizes and integrates insightsfrom the research literature on organization-wide applicationsof IT over the last 40 years, particularly the ES researchliterature summarized in Table 2.

A two-part model was required because most organizationsthat implement ES find themselves wanting or needing toconduct additional projects after the initial implementation(e.g., to upgrade the initial software and/or to improve andextend the original system). Since projects can have varyingsuccess, it was necessary to separate out the project-relatedfactors on the right from the longer-term drivers of organiza-tional benefits from ES use on the left. The right-hand side ofthe model hypothesizes that two factors identified repeatedlyin the literature—functional fit and overcoming organizationalinertia—are key determinants of organizational benefits from



each major ES implementation project in the first few yearsafter go live. The left-hand side of the model hypothesizesthat three factors from Davenport et al. (2002, 2004)—integration, process optimization, and improved access toinformation—together with a fourth factor, on-going majorES business improvement projects, explain variance inorganizational benefits from ES use over the longer term. Aseach ES business improvement project goes live, the high-level goals such as integrate, optimize, and improved accessto information that motivated the project are translated intoreal systems and real processes that help or hinder individualsas they attempt to do their jobs.

A preliminary test of the OBES model was conducted byanalyzing 126 customer presentations from SAP’s 2003 and2005 Sapphire USA conferences. The tests involved lookingfor evidence in each presentation of whether any or all of thesix independent variables in the model were described asimportant determinants of what senior management wouldview as benefits from their investment in ES. Details andexamples of the analysis were presented in Table 3,Appendices B and C, and Figure 3, with results summarizedin Table 5. Although Table 5 is silent about the amount ofvariance explained, the high frequencies reported in Table 5indicate that all six factors in the model are important driversof organizational benefits from ES.

The contribution of this paper on this important topic is itssynthesis of so much of the prior literature, its choice ofwhich factors to include and exclude, the arrangement ofthose factors in the two-part OBES model, and the evidencethat those factors are important, not the novelty of thosefactors. OBES is an important contribution because it pro-vides a richer overall explanation of the drivers of organiza-tional benefits from ES than any prior paper summarized inTable 2, there is strong theoretical justification for eachhypothesis in the model in both the ES and the broader ISliterature, and it has preliminary empirical support. Theexplanatory power of OBES comes from its two-part structureand its roots in over 40 years of IS research. The two-partstructure is necessary because, as is clear from row 6 inTable 5, many organizations undertake multiple on-going ESimprovement and upgrade projects and some projects arelikely to have better outcomes than others.

Finally, four directions for future research have been identi-fied in the course of this study. First, the test above is prelim-inary. It is important to test OBES with different types ofdata, in a range of other settings, both to continue assessingthe importance of the OBES factors and to identify other keyfactors—for example, information quality, organizationalcontext, assimilation (Liang et al. 2007), strategic alignment,project complexity, or degree of radical change or transforma-tion in a project—that might need to be added to the model.

Second, it would be valuable to identify key factors thatexplain variance in the independent OBES constructs. Forinstance, as mentioned above, Liu and Seddon (2009) suggestthat many implementation project CSFs influence ES successthrough their impact on the OBES benefit drivers. Third, assuggested in the final paragraph in the justification of H3,there are good reasons for believing that integration is notalways beneficial to an organization. It would be interestingto identify contexts where the claim made in OBES thatintegration is always beneficial breaks down. Fourth, as dis-cussed earlier, it would be interesting to explore relationshipsamong the OBES independent variables. In short, there isplenty of scope for future research in enterprise systems!

Acknowledgments

Thanks to Shari Shang, Christina Soh, Sia Siew Kien, RensScheepers, Toomas Tamm, Kalle Lyytinen, colleagues at theresearch seminars at the University of Melbourne and the Universityof Sydney, students in Enterprise Systems at the University ofMelbourne, and the anonymous reviewers for comments andfeedback on this and earlier versions of this paper. Thanks also toSAP for sharing their Sapphire presentations with their user com-munity, and to the speakers from the various firms who agreed toallow slides from their presentations to be included in this paper.

References

Abdinnour-Helm, S., Lengnick-Hall, M. L., and Lengnick-Hall,C. A. 2003. “Pre-implementation Attitudes and OrganizationalReadiness for Implementing an Enterprise Resource PlanningSystem,” European Journal of Operational Research (146:2), pp.258-273.

Akkermans, H., and van Helden, K. 2002. “Vicious and VirtuousCycles in ERP Implementation: A Case Study of Interrelationsbetween Critical Success Factors,” European Journal of Infor-mation Systems (11:1), pp. 35-46.

Aladwani, A. M. 2001. “Change Management Strategies for Suc-cessful ERP Implementation,” Business Process ManagementJournal (7:3), pp. 266-275.

Al-Mashari, M., Al-Mudimigh, A., and Zairi, M. 2003. “EnterpriseResource Planning: A Taxonomy of Critical Factors,” EuropeanJournal of Operational Research (146:2), pp. 352-364.

Argote, L., and Epple, D. 1990. “Learning Curves in Manu-facturing,” Science (247:4945), pp. 920-924.

Armenakis, A. A., and Bedeian, A. G. 1999. “OrganizationalChange: A Review of Theory and Research in the 1990s,”Journal of Management (25:3), pp. 293-315.

Bancroft, N. H., Siep, H., and Sprengel, A. 1998. ImplementingSAP R/3 (2

nd ed.), Greenwich, CT: Manning Publications.

Barki, H., and Pinsonneault, A. 2005. “A Model of OrganizationalIntegration, Implementation Effort, and Performance,” Organiza-tion Science (16:2), pp. 165-179.



Bingi, P., Sharma, M. K., and Godla, J. 1999. “Critical IssuesAffecting an ERP Implementation,” Information Systems Man-agement (16:3), pp. 7-14.

Bönnen, C., Herger, M., and Wodtke, D. 2008. SAP NetWeaverVisual Composer (2nd ed.), Boston, MA: Galileo Press.

Boudreau, M-C. 2003. “Learning to Use ERP Technology: ACausal Model,” in Proceedings of the 36th Hawaii InternationalConference on System Sciences, January 6-9, Los Alamitos, CA: IEEE Press.

Boudreau, M-C., and Robey, D. 1999. “Organizational Transitionto Enterprise Resource Planning Systems: Theoretical Choicesfor Process Research,” in Proceedings of the 20th InternationalConference on Information Systems, P. De and J. I. DeGross(eds.), Charlotte, NC, December 13-15, pp. 291-299.

Boudreau, M-C., and Robey, D. 2005. “Enacting Integrated Infor-mation Technology: A Human Agency Perspective,” Organiza-tion Science (16:1), pp. 3-18.

Bradford, M., and Florin, J. 2003. “Examining the Role ofInnovation Diffusion Factors on the Implementation Success ofEnterprise Resource Planning Systems,” International Journal ofAccounting Information Systems (4), pp. 205-225.

Brehm, L., Heinzl. A., and Markus, M. L. 2001. “Tailoring ERPSystems: A Spectrum of Choices and their Implications,” inProceedings of the 34th Hawaii International Conference onSystems Sciences, January 3-6, Los Alamitos, CA: IEEE Press.

Breyfogle, III, F. W. 2003. Implementing Six Sigma: SmarterSolutions Using Statistical Methods (2nd ed.), New York: Wiley.

Calvert, C. 2006. “A Change-Management Model for the Imple-mentation and Upgrade of ERP Systems,” in Proceedings of the17th Australasian Conference on Information Systems, Adelaide,December 6-8 (available online at http://aisel.aisnet.org/acis2006/18).

Cameron, K. S., and Whetten, D. A. 1983. “Some ConclusionsAbout Organizational Effectiveness,” in Organizational Effec-tiveness: A Comparison of Multiple Models, K. S. Cameron, andD. A. Whetten (eds.), New York: Academic Press, pp. 261-277.

Carr, D. F. 2002. “Hershey’s Sweet Victory,” Baseline, December,(available online at http://www.baselinemag.com/article2/0,3959,760615,00.asp, accessed August 2009).

Cotteleer, M. J., and Bendoly, E. 2006. “Order Lead-Time Improve-ment Following Enterprise-IT Implementation: An EmpiricalStudy,” MIS Quarterly (30:3), pp. 643-660.

Dalal, N. P., Kamath, M., Kolarik, W. J., and Sivaraman, E. 2004.“Toward an Integrated Framework for Modeling EnterpriseProcesses,” Communications of the ACM (47:3), pp. 83-87.

Davenport, T. H. 1993. Process Innovation–Reengineering Workthrough Information Technology, Cambridge, MA: HarvardBusiness School Press.

Davenport, T. H. 1998. “Putting the Enterprise into the EnterpriseSystem,” Harvard Business Review, July-August, pp. 121-131.

Davenport, T. H., and Harris, J. G. 2007. Competing on Analytics: The New Science of Winning, Cambridge, MA: Harvard BusinessSchool Press.

Davenport, T. H., Harris, J. G., and Cantrell, S. 2002. The Returnof Enterprise Systems: The Director’s Cut, Wellesley, MA: Accenture Institute for High Performance Business.

Davenport, T. H., Harris, J., and Cantrell, S. 2004. “EnterpriseSystems and On-Going Process Change,” Business ProcessManagement Journal (10:1), pp. 16-26.

Davis, G. B. 1974. Management Information Systems: ConceptualFoundations, Structure, and Development, New York: McGrawHill.

Davis, G. B., and Olson, M. H. 1985. Management InformationSystems, Conceptual Foundations, Structure, and Development(2nd ed.), New York: McGraw-Hill.

Deloitte Consulting. 1998. ERP’s Second Wave: Maximizing theValue of ERP-Enabled Processes, New York: DeloitteConsulting.

DeLone, W. H., and McLean, E. R. 1992. “Information SystemsSuccess: The Quest for the Dependent Variable,” InformationSystems Research (3:1), pp. 60-95.

DeLone, W. H., and McLean, E. R. 2003. “The DeLone andMcLean Model of Information Systems Success: A Ten-YearUpdate,” Journal of Management Information Systems, (19:4),pp. 9-30.

DeMarco, T. 1978. Structured Analysis and System Specification,New York: Yourdon Press.

Financial Executives International & CSC. 2006. 2006 AnnualReport: Technology Issues for Financial Executives, Committeeon Finance & Information Technology of Financial ExecutivesInternational, Financial Executives Research Foundation, andComputer Sciences Corporation, Florham Park, NJ (http://www.csc.com/solutions/managementconsulting/knowledgelibrary/uploads/2335_1.pdf; accessed August 2009).

Finney, S., and Corbett, M. 2007. “ERP Implementation: A Com-pilation and Analysis of Critical Success Factors,” BusinessProcess Management Journal (13:3), pp. 329-347.

Gable, G. G., Sedera, D., and Chan T. 2003. “Enterprise SystemsSuccess: A Measurement Model,” in Proceedings of the 24th

International Conference on Information Systems, S. T. March,A. Massey, and J. I. DeGross (eds.), Seattle, WA, December 14-17, pp. 576-591.

Gane C., and Sarson, T. 1979. Structured Systems Analysis: Toolsand Techniques, Englewood Cliffs, NJ: Prentice-Hall.

Garvin, D. A. 1988. Managing Quality: A Strategic and Competi-tive Edge, New York: The Free Press.

Gartner. 2008. “Gartner EXP Worldwide Survey of 1,500 CIOsShows 85 Percent of CIOs Expect ‘Significant Change’ OverNext Three Years,” Gartner Newsroom (http://www.gartner.com/it/page.jsp?id=587309; accessed August 2009).

Gattiker, T. F., and Goodhue, D. L. 2005. “What Happens AfterERP Implementation: Understanding the Impact of Inter-dependence and Differentiation on Plant-Level Outcomes,” MISQuarterly (29:3), pp. 559-585.

Gefen, D., and Ragowsky, A. 2005. “A Multi-Level Approach toMeasuring the Benefits of an ERP System in ManufacturingFirms,” Information Systems Management (22:1), pp. 18-25.

Grover, V., Jeong, S. R., and Segars, A. H. 1996. “InformationSystems Effectiveness: The Construct Space and Patterns ofApplication,” Information and Management (31:4), pp. 177-191.

Hammer, M. 1996. Beyond Reengineering, New York: HarperCollins.

Hammer, M. 2007. “The Process Audit,” Harvard BusinessReview, April, pp. 111-123.

Hammer, M., and Champy, J. 1993. Reengineering the Cor-poration: A Manifesto for Business Revolution, New York: Harper Business.



Harley, B., Wright, C., Hall, R., and Dery, K. 2006. “ManagementReactions to Technological Change: The Example of Enterprise Resource Planning,” The Journal of Applied Behavioral Science(42:1), pp. 58-75.

Harris, J. G., and Davenport, T. H. 2006. New Growth fromEnterprise Systems: Achieving High Performance throughDistinctive Capabilities, Wellesley, MA: Accenture Institute forHigh Performing Business.

Hesterman, C., Anderson, R. P., and Pang, C. 2009. “MagicQuadrant for Midmarket and Tier 2-Oriented ERP for Product-Centric Companies,” Gartner Research (available online ath t t p : / / w w w . g a r t n e r . c o m / t e c h n o l o g y / m e d i a -products/reprints/microsoft/vol4/article12/article12.html)

Hofstede, G. 2001. Culture’s Consequences: Comparing Values,Behaviors, Institutions, and Organizations Across Nations (2nd

ed.), London: Sage Publications. Holland, C., and Light, B. 1999. “A Critical Success Factors Model

for ERP Implementation,” IEEE Software (16:3), pp. 30-36. Hong, K-K, and Kim, Y-G. 2002. “The Critical Success Factors for

ERP Implementation: An Organizational Fit Perspective,”Information and Management (40:1), pp. 25-40.

Ives, B., Olson, M. H., and Baroudi, J. J. 1983. “The Measurementof User Information Satisfaction,” Communications of the ACM(26:10), pp. 785-793.

Joshi, K. 1991. “A Model of Users’ Perspective on Change: TheCase of Information Systems Technology Implementation,” MISQuarterly (15:2), pp. 229-242.

Joshi, K. 1992. “A Casual Path Model of the Overall User Attitudestoward the MIS Function: The Case of User Information Satis-faction,” Information and Management (22:2), pp. 77-88.

Keen, P. G. W. 1981. “Information Systems and OrganizationalChange,” Communications of the ACM (24:1), pp. 24-33.

Keen, P. G. W., and Scott Morton, M. S. 1978. Decision SupportSystems: An Organizational Perspective, Reading, MA:Addison Wesley.

Keil, M., and Tiwana, A. 2006. “Relative Importance of EvaluationCriteria for Enterprise Systems: A Conjoint Study,” InformationSystems Journal (16:2), pp. 272-262.

Keller, G., and Teufel, T. 1998. SAP R/3 Process-Oriented Imple-mentation, Boston, MA: Addison-Wesley Longman Publishing.

Ko, D-G., Kirsch, L. J., and King, W. R. 2005. “Antecedents ofKnowledge Transfer from Consultants to Clients in EnterpriseSystem Implementations,” MIS Quarterly (29:1), pp. 59-85.

Kohli, R. 2007. “Innovating to Create IT-Based New BusinessOpportunities at United Parcel Service,” MIS Quarterly Executive(6:4), pp. 199-210.

Kohli, R., and Grover, V. 2008. “Business Value of IT: An Essayon Expanding Research Directions to Keep up with the Times,”Journal of the AIS (9:1), pp. 23-39.

Kotter, J. P. 1996. Leading Change, Cambridge, MA: HarvardBusiness School Press.

Lapointe, L., and Rivard, S. 2005. “A Multilevel Model of Resis-tance to Information Technology Implementation,” MISQuarterly (29:3), pp. 461-491.

Lee, J. Siau, K, and Hong, S. 2003. “Enterprise Integration withERP and EAI,” Communications of the ACM (46, 2), pp. 54-60.

Lee, J. C., and Myers, M. D. 2004. “Dominant Actors, PoliticalAgendas, and Strategic Shifts Over Time: A Critical Ethnog-

raphy of an Enterprise Systems Implementation,” Journal ofStrategic Information Systems (13:4), pp. 355-374.

Liang, H., Saraf, N., Hu, Q., and Xue, Y. 2007. “Assimilation ofEnterprise Systems: The Effect of Institutional Pressures and theMediating Role of Top Management,” MIS Quarterly (31:1), pp.59-87.

Light, B. 2005. “Going Beyond ‘Misfit’ as a Reason for ERPPackage Customization,” Computers in Industry (56:6), pp.606-619.

Linthicum, D. S. 2000. Enterprise Application Integration, UpperSaddle River, NJ: Addison-Wesley.

Liu, A. Z., and Seddon, P. B. 2009. “Understanding How ProjectCritical Success Factors Affect Organizational Benefits fromEnterprise Systems,” Business Process Management Journal(15:5), pp. 716-743.

Luo, W. H., and Strong, D. M. 2004. “A Framework for EvaluatingERP Implementation Choices,” IEEE Transactions on Engi-neering Management (51:3), pp. 322-333.

Mabert, V. A., Soni, A., and Venkataraman, M. A. 2003a.“Enterprise Resource Planning: Managing the ImplementationProcess,” European Journal of Operational Research (146:2), pp.302-314.

Mabert, V. A., Soni, A., and Venkataraman, M. A. 2003b. “TheImpact of Organization Size on Enterprise Resource Planning(ERP) Implementations in the US Manufacturing Sector,” Omega(31:3), pp. 235-246.

Markus, M. L. 2004. “Technochange Management: Using IT toDrive Organizational Change,” Journal of Information Tech-nology (19:1), pp. 4-20.

Markus, M. L., Axline, S., Petrie, D., and Tanis, C. 2000. “Learning from Adopters’ Experiences with ERP: ProblemsEncountered and Success Achieved,” Journal of InformationTechnology (15:4), pp. 245-265.

Markus, M. L., and Benjamin, R. I. 1996. “Change Agentry: TheNext IS Frontier,” MIS Quarterly (20:4), pp. 385-407.

Markus, M. L., and Tanis, C. 2000. “The Enterprise SystemsExperience—from Adoption to Success,” in Framing theDomains of IT Research: Glimpsing the Future through the Past,R. W. Zmud (ed.), Cincinnati, OH: Pinnaflex EducationalResources, pp. 173-207.

McAfee, A. 2002. “The Impact of Enterprise Information Tech-nology Adoption on Operational Performance: An EmpiricalInvestigation,” Production and Operations Management (11:1),pp. 33-53.

Mohr, L. B. 1982. Explaining Organizational Behavior, SanFrancisco: Jossey-Bass Publishers.

Moon, Y. B. 2007. “Enterprise Resource Planning (ERP): AReview of the Literature,” International Journal of Managementand Enterprise Development (4:3), pp. 235-263.

Motwani, J., Mirchandani, D., Madan, M., and Gunasekaran, A.2002. “Successful Implementation of ERP Projects: Evidencefrom Two Case Studies,” International Journal of ProductionEconomics (75:1), pp. 83-94.

Nah, F. H., Lau, J. L-S., and Kuang, J. H. 2001. “Critical Factorsfor Successful Implementation of Enterprise Systems,” BusinessProcess Management Journal (7:3), pp. 285-296.

OASIS. 2007. Web Services Business Process Execution LanguageVersion 2.0,11 April (http://docs.oasis-open.org/wsbpel/2.0/wsbpel-v2.0.pdf; accessed August 2009)



Object Management Group. 2007. “Unified Modeling Language(UML), Version 2.1.1,” Object Management Group, Needham,MA (available online at http://www.omg.org/technology/documents/formal/uml.htm; accessed August 2009).

Parr, A., and Shanks, G. 2000. “A Model of ERP ProjectImplementation,” Journal of Information Technology (15:4), pp.289-303.

Parr, A., Shanks, G., and Darke, P. 1999. “Identification ofNecessary Factors for Successful Implementation of ERPSystems,” in New Information Technologies in OrganisationalProcesses, O. Ngwenyama, L. D. Introna, M. D. Myers, and J. I.DeGross (eds.), Boston, MA: Kluwer Academic Publishers, pp.99-119.

Purvis, R. L., Sambamurthy, V., and Zmud, R. W. 2001. “TheAssimilation of Knowledge Platforms in Organizations: AnEmpirical Investigation,” Organization Science (12:2), pp.117-135.

Ragin, C. C. 1987. The Comparative Method: Moving BeyondQualitative and Quantitative Strategies, Berkeley, CA: Univer-sity of California Press.

Rai, A., Sang, S. S., and Welker, R. B. 2002. “Assessing theValidity of IS Success Models: An Empirical Test and Theo-retical Analysis,” Information Systems Research (13:1), pp.50-69.

Rajagopal, P. 2002. “An Innovation-Diffusion View of Implemen-tation of Enterprise Resource Planning (ERP) Systems andDevelopment of a Research Model,” Information andManagement (40:2), pp. 87-114.

Ramiller, N. C., Swanson E. B., and Wang, P. 2008. “ResearchDirections in Information Systems: Toward an InstitutionalEcology,” Journal of the Association for Information Systems(9:1), pp. 1-22.

Ranganathan, C., and Brown, C. V. 2006. “ERP Investments andthe Market Value of Firms: Toward an Understanding ofInfluential ERP Project Variables,” Information SystemsResearch (17:2), pp 145-161.

Reilly, K. 2005. “AMR Research Reports Oracle-PeopleSoftMerger Could Result in a $5.5 Billion Applications Business,”AMR Research, January 18 (available online at http://www.amrresearch.com/content/view.aspx?compURI=tcm:7-33464&title=AMR+Research+Reports+Oracle-PeopleSoft+Merger+Could+Result+in+a+$5.5+Billion+Applications+Business;accessed August 2009).

Robey, D., Ross, J. W., and Boudreau, M-C. 2002. “Learning toImplement Enterprise Systems: An Exploratory Study of theDialectics of Change,” Journal of MIS (19:1), pp. 17-46. Rosemann, M., Vessey, I., and Weber, R. 2004. “Alignment in

Enterprise Systems Implementations: The Role of OntologicalDistance,” in Proceedings 25th International Conference onInformation Systems, R. Agarwal, L. Kirsch, and J. I. DeGross(eds.), Washington DC, December 12-15, pp. 439-448.

Ross, J. W., and Vitale, M. 2000. “The ERP Revolution: SurvivingVersus Thriving,” Information Systems Frontiers (2:2), pp.233-241.

Ross, J. W., Weill, P., and Robertson, D. C. 2006. EnterpriseArchitecture as Strategy, Creating a Foundation for BusinessExecution, Cambridge, MA: Harvard Business School Press.

Samson, D., and Terziovski, M. 1999. “The Relationship betweenTotal Quality Management Practices and Operational Per-

formance,” Journal of Operations Management (17:4), pp.393-409.

SAP. 2006. “Caterpillar Logistics Services Inc.” (http://www.sap.com/usa/industries/automotive/pdf/CS_Caterpillar_Logistics_Services.pdf; accessed June 2008).

SAP. 2008. “Evolving to a Business Process Platform with SAPSolutions, Sapphire, Orlando USA” (available online at http://www.sap.com/community/showdetail.epx?itemID=13327;accessed August 2009).

Sarker, S., and Lee, A. S. 2003. “Using a Case Study to Test theRole of Three Key Social Enablers in ERP Implementation,”Information and Management (40:8), pp. 813-829.

Scheer, A. W. 1994. Business Process Engineering: ReferenceModels for Industrial Enterprises, New York: Springer Verlag.

Scheer, A. W., and Haberman, F. 2000. “Making ERP a Success,Using Process Models to Achieve Positive Results,” Com-munications of the ACM (43:4), pp. 57-62.

Scott, J. E. 1999. “The FoxMeyer Drugs’ Bankruptcy: Was it aFailure of ERP?,” in Proceedings of the Fifth AmericasConference on Information Systems, Milwaukee, WI, pp. 223-225(available online at http://aisel.aisnet.org/amcis1999/80).

Scott, J. E., and Vessey, I. 2002. “Managing Risks in EnterpriseSystems Implementations,” Communications of the ACM (45:4),pp. 74-81.

Sedera, D., and Gable, G. 2004. “A Factor and Structural EquationAnalysis of the Enterprise Systems Success MeasurementModel,” in Proceedings of the 25th International Conference onInformation Systems, R. Agarwal, L. Kirsch, and J. I. DeGross(eds.), Washington, DC, December 12-15, pp. 449-464.

Shang, S. 2001. Maximizing Benefits from Enterprise Systems,unpublished Ph.D. Dissertation, The University of Melbourne,Australia.

Shang, S., and Seddon, P. B. 2002. “A Comprehensive Frameworkfor Assessing and Managing the Benefits of Enterprise Systems:The Business Manager’s Perspective,” Information SystemsJournal (12:4), pp. 271-299.

Sia, S. K., and Soh, C. 2002. “Severity Assessment of ERP-Organization Misalignment: Honing on in Ontological Structureand Context Specificity,” in Proceedings of the 23rd InternationalConference on Information Systems, R. D. Galliers, L. Applegate,and J. I. DeGross (eds.), Barcelona, Spain, December 15-18, pp.723-729 (available online at http://aisel.aisnet.org/icis2002/70).

Sia, S. K., and Soh, C. 2007. “An Assessment of Package-Organization Misalignment: Institutional and Ontological Struc-tures,” European Journal of Information Systems (16:5), pp.568-583.

Siegel, S., and Castellan, N. J. 1988. Nonparametric Statistics forthe Behavioral Sciences (2nd ed.), New York: McGraw-Hill.

Skok, W., and Legge, M. 2001. “Evaluating Enterprise ResourcePlanning (ERP) Systems Using an Interpretive Approach,”Proceedings of the 2001 ACM SIGCPR Conference on ComputerPersonnel Research, April, New York: ACM Press, pp. 189-197.

Soffer, P., Golany, B., and Dori, D. 2003. “ERP Modeling: AComprehensive Approach,” Information Systems (28:6), pp.673-690.

Soh, C., Kien, S. S., and Tay-Yap, J. 2000. “Cultural Fits andMisfits: Is ERP a Universal Solution?,” Communications of theACM (43:4), pp. 47-51.



Soh, C., and Sia, S. K. 2005. “The Challenges of Implementing‘Vanilla’ Versions of Enterprise Systems,” MIS QuarterlyExecutive (4:3), pp. 373-384.

Soh, C., Sia, S. K., Boh, W. F., and Tang, M. 2003. “Misalign-ments in ERP Implementation: A Dialectic Perspective,”International Journal of Human-Computer Interaction (16:1), pp.81-100.

Somers, T. M., and Nelson, K. 2001. “The Impact of CriticalSuccess Factors across the Stages of Enterprise ResourcePlanning Implementations,” in Proceedings of the 34th HawaiiInternational Conference on System Sciences, Los Alamitos, CA: IEEE Press.

Somers, T. M., and Nelson, K. G. 2003. “The Impact of Strategyand Integration Mechanisms on Enterprise System Value: Empirical Evidence from Manufacturing Firms,” EuropeanJournal of Operational Research (146:2), pp. 315-338.

Somers, T. M., and Nelson, K. G. 2004. “A Taxonomy of Playersand Activities across the ERP Project Life Cycle,” Informationand Management (41:3), pp. 257-278.

Staehr, L. 2007. “Assessing Business Benefits from ERP Systems: An Improved ERP Benefits Framework,” in Proceedings of the28th International Conference on Information Systems, Montreal,Canada (available online at http://aisel.aisnet.org/icis2007/36).

Staehr, L., Shanks, G., and Seddon, P. 2002. “Understanding theBusiness Benefits of ERP Systems,” in Proceedings of the 8th

Americas Conference on Information Systems, R. Ramsower andJ. Windsor (eds.), Dallas, TX, August 9-11, pp. 899-906.

Staehr, L., Shanks, G., and Seddon, P. B. 2006. “Understanding theBusiness Consequences of ERP Use,” in Proceedings of the 14th

European Conference on Information Systems, J. Ljungberg andM. Andersson (eds.), Göteborg, Sweden, June 12-14.

Strong, D. M., and Volkoff, O. 2004. “A Roadmap for EnterpriseSystem Implementation,” Computer (37:6), pp. 22-29.

Sumner, M. 2000. “Risk Factors in Enterprise-Wide/ERP Pro-jects,” Journal of Information Technology (15:4), pp. 317-327.

Tallon, P., Kraemer, K., and Gurbaxani, V. 2000. “Executives’Perceptions of the Business Value of Information Technology: A Process Orientated Approach,” Journal of ManagementInformation Systems (16:4), pp. 145-173.

Taylor, S., and Todd, P. 1995. “Understanding Information Tech-nology Usage: A Test of Competing Models,” InformationSystems Research (6:2), pp. 144-176.

Umble, E. J., Haft, R. R., and Umble, M. M. 2003. “EnterpriseResource Planning: Implementation Procedures and CriticalSuccess Factors,” European Journal of Operational Research(146:2), pp. 241-257.

Volitich, D. 2008. IBM Cognos 8 Business Intelligence: TheOfficial Guide, New York: McGraw-Hill.

Volkoff, O., Elmes, M. B., and Strong, D. 2004. “Enterprise Sys-tems, Knowledge Transfer and Power Users,” Journal ofStrategic Information Systems (13:4), pp. 279-304.

Wagner, E. L., and Newell, S. 2007. “Exploring the Importance ofParticipation in the Post-Implementation Period of an ES Project:A Neglected Area,” Journal of the AIS (8:10), pp. 508-524.

Walton, M. 1986. The Deming Management Method, New York:Perigee Books.

Webster, J., and Watson, R. T. 2002. “Analyzing the Past to Pre-pare for the Future: Writing a Literature Review,” MIS Quarterly(26:2), pp. xiii-xxiii.

Wei, H. L., Wang, E. T. G., and Ju, P. H. 2005. “UnderstandingMisalignment and Cascading Change of ERP Implementation: A Stage View of Process Analysis,” European Journal ofInformation Systems (14:4), pp. 324-334.

Wixom, B. H., and Watson, H. J. 2001. “An Empirical Investi-gation of the Factors Affecting Data Warehousing Success,” MISQuarterly (25:1), pp. 17-41.

Wu, J-H, and Wang, Y-M. 2007. “Measuring ERP Success: TheKey-Users’ Viewpoint of the ERP to Produce a Viable IS in theOrganization,” Computers in Human Behavior (23:3), pp.1582-1596.

Yang, S., and Seddon, P. B. 2004. “Benefits and Key ProjectSuccess Factors from Enterprise Systems Implementations: Lessons from Sapphire 2003,” in Proceedings of the AustralasianConference on Information Systems (available online athttp://aisel.aisnet.org/acis2004/27).

Ziskasen, J. 2008. “The Business Drivers and Value of Upgradingto SAP ERP 6.0 at Kraft Foods, Sapphire, Orlando USA”(http://www.sap.com/community/showdetail.epx?itemID=11736;accessed August 2009).

Zuboff, S. 1988. In the Age of the Smart Machine: The Future ofWork and Power, New York: Basic Books.

About the Authors

Peter B. Seddon is an associate professor in the Department ofInformation Systems at the University of Melbourne, Australia. Histeaching and research focus on helping people and organizationsgain greater benefits from their use of information technology. Hismajor research interests are in the areas of evaluating informationsystems success, packaged enterprise application software, ITmanagement, IT outsourcing, business intelligence, and accountinginformation systems.

Cheryl Calvert is an Enterprise Systems Project Manager,Corporate Business Systems, Monash University, Australia. Herresearch interests include enterprise system change management,education, and learning; evaluation of enterprise systems education;and finance-related work on stock market seasonality. Cheryl hasauthored a number of books on SAP training that are used atuniversities in Australia and internationally. She is a project leaderfor major SAP business improvement projects at Monash University.

Song Yang is a Ph.D. student at the University of Melbourne. Herresearch interests include understanding social consequences ofmobile phone use and factors affecting benefits from enterprisesystem.


Seddon et al./Key Factors Affecting Organizational Benefits—Appendices

RESEARCH ARTICLE

A MULTI-PROJECT MODEL OF KEY FACTORS AFFECTINGORGANIZATIONAL BENEFITS FROM ENTERPRISE SYSTEMS

By: Peter B. SeddonDepartment of Information SystemsUniversity of MelbourneVictoria [email protected]

Cheryl CalvertCorporate Business SystemsMonash UniversityVictoria [email protected]

Song YangDepartment of Information SystemsUniversity of MelbourneVictoria [email protected]

Appendix A

Full List of Organizations and Speaker Roles

2003

2005

Organization Name Presenter Bus

ines

s M

anag

er(s

ee N

ote

for

mea

ning

of “

1”)

CIO

Hea

d, E

S o

r P

roje

ct M

anag

er

On

the

podi

um m

ore

than

once

in 2

003

or 2

005

On

the

podi

um in

bot

h20

03 a

nd 2

005

Sam

e to

pic

disc

usse

din

bot

h se

ssio

ns?

1 2Wire Senior Director, IT 11 Abbot Laboratories Business Manager 11 Addmore Personnel + Bookham

+TallardPresident and VPs 1 1

1 Adidas CIO of Adidas Salomon 11 Advanced Energy Director of planning 1

MIS Quarterly Vol. 34 No. 2, Seddon et al., Appendices/June 2010 A1


2003

2005


ines

s M

anag

er(s

ee N

ote

for

mea

ning

of “

1”)

CIO

Hea

d, E

S o

r P

roje

ct M

anag

er

On

the

podi

um m

ore

than

once

in 2

003

or 2

005

On

the

podi

um in

bot

h20

03 a

nd 2

005

Sam

e to

pic

disc

usse

din

bot

h se

ssio

ns?

1 Air Products & Chemicals Director – SAP HR Project 11 Air Products & Chemicals Director, Business process ERP

program1 1 1 No

1 Alabama Gas Corp., EnergenCorp

Vice President, Vice President & CIO 1 1

1 Allied Irish Bank Group CIO 11 American Army Project manager, Logistics IS 11 Armstrong World Industries VP Logistics & IT for Building

Products1

1 Artisan Entertainment CFO, CIO and CEO 11 AstraZeneca Executive director, SAP project 11 AT&T bus services Process controller for revenue 1

1 AUDI AG Head of CRM Applications 11 Auto Industry Action Group General Motors Loaned Executive 1

1 Avaya Senior Manager, Supply ChainPlanning

1

1 AZ Electronic Materials CFO 11 B & Q Plc Director of Commercial Systems 11 Banco Itaú General Manager 1

1 Bank of Canada HR director + ERP services mgr 11 Barclays Bank Director, Finance Projects 11 BASF Project Leader 1

1 Blount + Fusion UV +Greenheck

VP, IS + Sales Support + VI, IS 1 1

1 Bosch North Corporation Project Lead 11 Bosch Rexroth Corporation VP and CIO, Director Business

Applications1 0 No

1 Bristol-Myers Squibb Co. Program Director Informatics 11 Brookshire Grocery Co. Director Financial Accounting 11 Brother International Corp. CIO 1

1 Brother International Corp. President & CEO, and CIO 1 1 1 Yes1 Capita + Cincinnati Insurance both Project managers 11 Cardinal Health Senior Project Leader, SD 11 Caterpillar Group President, Caterpillar logistics 11 Celent CEO 11 CHEP Enterprise Architect 1

1 ChevronTexaco Manger, Global SAP Strategy 11 City of Cape Town Director of ERP Business

Transformation 1

1 CN Rail Director of Business Solutions 11 Coca Cola Enterprises Inc. Manager, e-Procurement 1

1 Colgate Palmolive Director, Global IT 11 Computer Sciences Corp. Vice President, Bus Development

plus Director, Global Outsourcing2

1 Computer Sciences Corp. VP, Business Development 1 1 Yes1 ConAgra Foods Inc. VP Enterprise System implementation 1

1 ConAgra Foods Inc. Director, Business Practices 1 1 Yes1 ConocoPhillips Project manager 11 Du Pont Mgr Business Planning 1

A2 MIS Quarterly Vol. 34 No. 2, Seddon et al., Appendices/June 2010


2003

2005


ines

s M

anag

er(s

ee N

ote

for

mea

ning

of “

1”)

CIO

Hea

d, E

S o

r P

roje

ct M

anag

er

On

the

podi

um m

ore

than

once

in 2

003

or 2

005

On

the

podi

um in

bot

h20

03 a

nd 2

005

Sam

e to

pic

disc

usse

din

bot

h se

ssio

ns?

1 Energen VP&CIO 11 Five North American state and

local government orgsall five are project managers 1

1 Florida Crystals Corporation Vice President & CIO 11 GE Consumer and Industrial Content mgr leader 11 Gen-Probe Senior director, IS 1

1 Graybar VP and CIO 11 Graybar VP and CIO 1 1 Yes1 Great West Life and Annuity Project mgr 1

1 GTECH Corp.+ Novo Nordisk +IDC

Director, of Fin. Planning & Analysis 1

1 GTECH Corporation Director, Corp financial planning &anal.

1 1 No

1 Halliburton Company Director, ERP Center of Excellence1 Hawaiian Tropic EVP & CFO,VP, Kentucky Division 11 Hershey Foods Director of applications 1

1 HP Director Shared Services and SAPCOE

1

1 Indigo Books CTO 11 Infineon Technology Vice President, IT Alignment; plus

partner from Accenture1

1 International Paper Director, HR Operations 11 J. Crew Senior Vice President & CIO 11 Johnson & Johnson Senior Director, Pharmaceutical

Research and Development, J&JPharmaceuticals

1

1 Kaeser + Robotics Inc Delivery Bus Systems Leader 11 Kimberley Clark VP Sen Marketing Office + IT

manager1 1

1 KLA-Tencor Senior Information TechnologyDirector, Applications

1

1 KLA-Tencor Senior Director Sales systems 1 1 No1 Lennox + Komatsu Director of IT + Manager of ERP

Development1 1

1 Lions Gate Entertainment CIO 11 Lockheed Martin ERP project mgr 1

1 Lockheed Martin Product Manager 1 1 No1 L’Oreal Project leader 11 LSI logic Director, SCM 11 Lyondell Chemical Co Project mgr 1

1 Marathon Ashland PetroleumLLC

Project Manager, & TechnologyManager

1

1 Marathon Oil + Wellogix President, Global procurement + VPOperations

1 1 No

1 Mar-Mac Wire, Inc. CIO & CFO 11 MassMutual Financial Group Vice President, Corporate Services 11 McCormick & Company CIO & V.P. Global Business Solutions 11 MCI/WorldCom, Inc.+

IBM/Telefonica Snr Director, Strategic ITDevelopment

1

1 Millennium Chemicals Director, eBusiness, 1



2003

2005


ines

s M

anag

er(s

ee N

ote

for

mea

ning

of “

1”)

CIO

Hea

d, E

S o

r P

roje

ct M

anag

er

On

the

podi

um m

ore

than

once

in 2

003

or 2

005

On

the

podi

um in

bot

h20

03 a

nd 2

005

Sam

e to

pic

disc

usse

din

bot

h se

ssio

ns?

1 Morrison homes VP&CIO 11 Mott + Johnson + Conair +

Brown-FormanFour Directors, Supply Chain 1 1

1 Nike Discussed Netweaver 11 Norske Canada Project Director & VP Supply Chain &

IT1

1 Nortel SAP Program Mgr 11 North Carolina Dept of

Transport Project Manager & Funct. TeamManager

1

1 OfficeMax VP – Direct Marketing, VP – CRM 11 Ondeo Nalco Global Director, Mfg Volume Strategy 11 Ontario Electricity Manager, Planning & Perf. Mgt

Systems 1

1 Pacific Coast Feather Company CIO & Director of Business IS 11 Phillips + Deloitte Senior Bus Info Mgr 1

1 Procter & Gamble Employee services area, 11 Procter & Gamble Global Director, Supply Network

Operations1 1 No

1 Purdue Pharmaceuticals Exec Director, Information Officer 11 Rohm & Haas e-Transformation Director 11 Rohm & Haas Company eBusiness Technology Manager 1 1 No1 Royal Dutch Shell Group: CIO Team 1

1 Schenker AG Head of IT Management Logistics 11 SI Corporation Project manager 1

1 Sony Europe General Manager, Finance & HRSystems

1

1 Tastykake Director, Enterprise Apps 11 Tesoro VP and CIO 1

1 Tetra Pak SA + Adobe SystemsInc.

Head of Global IM Support & VP IS 1

1 Texas Instruments Lead solution architect 11 Texas Instruments Director Procurement Systems 1 1 No

1 Titanium Metals Mgr Bus Support and Apps 11 Tom Davenport + Wells Fargo+

Bank One +First Chicagofour project managers 1

1 Toyota Material Handling,U.S.A.

Vice President 1

1 Trivirix international Vice President of InformationTechnology

1

1 Tyson Foods Project mgr 11 Ulta Senior VP Sales+Sen VP IS 1 11 Uni of Kentucky + Baylor

CollegeProf and project director and ProjectManager

1

1 US Army Project mgrs (one from CSC) 11 US Customs + Treasury Business requirements director 11 US Navy Project manager 11 US Pipe Project manager, technology 1

1 VF Services, Inc. VP supply chain 11 Visteon Corp (Part of Ford till Global director, Applic. Strategy 1



2003

2005


ines

s M

anag

er(s

ee N

ote

for

mea

ning

of “

1”)

CIO

Hea

d, E

S o

r P

roje

ct M

anag

er

On

the

podi

um m

ore

than

once

in 2

003

or 2

005

On

the

podi

um in

bot

h20

03 a

nd 2

005

Sam

e to

pic

disc

usse

din

bot

h se

ssio

ns?

2000)1 Washington Post VP of Operations 1

1 Waters Corp. + Villeroy & BochAG

Director of Marketing Services & CIOand Project leader

1 1 1

1 Waters Corporation Director of PLM 1 1 No1 Wells Fargo Bank + Corp

Properties GpProject mgr 1

1 Whirlpool + Pacific Cycle Director, eBusiness Nth America +Director, IS

1

1 William Wrigley Mgr Global development SAP 11 Wolf Inc CFO 11 Wolverine Director, Internet marketing + Senior

IT director1 1

1 Wrigley Project lead 160 70 48 31 64 5 8 4 Yes’s

130 142 34% 22% 45%

Notes:1. A “1” indicates that the characteristic of interest, identified by the column heading, was present or applies in this presentation.2. A “+” indicates speakers from more than one organization (e.g., a panel discussion).



Appendix B

Examples of Strength of Evidence Judgments of Positive Causal Relationshipsbetween Benefit Drivers in the OBES Model and Organizational Benefits fromES Use from the Perspective of Senior Management

Factor SoE Example Explanation of SoE Classification

1. Functionalfit

1

(Case 59, slide 10) Copyright © Texas Instruments; used by permission.

In this slide, the speaker from Texas Instruments(TI), a U.S. $19 billion semiconductor company,explains how TI was using SAP’s enterprise buyerprofessional (EBP) product, interfaced to SAP’sR/3 ERP product, to enable over 1,300 users infive countries to use eProcurement. The func-tionality highlighted EPB’s ability to use “punchout” catalogs (e.g., online catalogs maintained bySuppliers A and B, rather than an internal catalogmaintained by TI). Since the benefits of thisfunctionality were not clearly spelled out, thestrength of this evidence that functional fit leads tobenefits was judged to be limited (i.e., 1).

2

(Case 28, slide 10) Copyright © CSC; used by permission.

In this slide, the speaker from Computer SciencesCorporation (CSC), a U.S. $13.6 billion out-sourcing firm, explains why her firm chose to gofor an early release of SAP’s CRM product. Presumably, the listed functionality will producevalue for CSC. The strength of this evidence thatfunctional fit leads to benefits was judged to bemoderate (i.e., 2).

3 “And any of you guys there in the audience that do apparelor footwear understand that people come in sizes, clothescome in sizes and software doesn’t understand sizes. Justdoes not. Everybody wears clothes, you’d think they’dunderstand it, but it is very difficult to find a package that issuitable for footwear and apparel....So, after searching, weselected SAP, and I think we made the right decisionbecause—look at them. They are in it for the long haul. They are a major player, and we worked with Reebok,because Reebok at the same time was doing somethingsimilar. They were out there searching for a system, theyselected SAP about the same time we did. And so we got

In this quotation, the speaker from VF ServicesInc, a U.S. $ 5 billion per annum manufacturerwhose brands include Lee, Wrangler, Vanity Fair,and North Face, found that functional fit in SAP’sbasic retail ERP solution was so poor that theyworked with Reebock and SAP to develop an ap-parel and footwear solution (AFS) version ofSAP’s ERP software, tailored to the needs of theclothing and apparel industry (which needs tokeep track of garments of the same style in manysizes and colors). Presumably, this additionalfunctionality leads to greater benefits. The

Data Mart Leverage existing data mart Supplier reporting portal Total spend/analytics

Standard PO/GR controlreconciliation

Standard SAP financial reports If a PR needs to be placed with EBP

supplier, then supplier sees same output/processes

Leverage existing archiving Leverage existing tax logic

Purchase Order

Purchase Requisition

SAP Reports

SAP R/3 Catalog

Shopping Cart

Texas Instrument’s SRM eProcurement Solution

Text Order

Internal Catalog

Supplier A

Supplier B

EBP Bugseye

Punch-out to Supplier Catalog

© C

SC

/ S

AP

AG

200

3,

CR

M in

th

e S

erv

ice

s S

ect

or:

Pre

pa

ratio

n a

nd A

lign

me

nt, L

yn B

urc

hfie

ld,

Ro

n R

icke

tts

/ 1

0

CSC Proprietary

Scope: Clearly Defined Based on Objectives

• Contact Management• Lead Management• Opportunity Management (including TAS Opportunity Planning Methodology)• Client Data• Client Planning• Sales Cycle Management• Reporting and Analysis• Access via the Internet• Mobile Usage• Work Flow Management• Linkages to Research/Information Repositories• Security• Marketing/Campaign Management




together, and between the two of us, we helped designAFS. So if there is a lot of mess in there, I guess you canblame VF and Reebok. But we tried to do our best” (Case55, transcript p.2).

strength of this evidence that functional fit leads tobenefits was judged to be strong (i.e., 3).

2. Over-comingorganizationalinertia

1 “People must be prepared for change across the organi-zation. They need support to become IAS literate in time. Stakeholder management is key to ensure desired behaviorchange....Training staff is the most significant challenge inconverting to IAS” (Case 9, slide 13).

In this quotation, the director of Finance Projectsfor Barclays Bank plc, a major UK bank, discus-ses the bank’s planned move to internationalaccounting standards (IAS). These two sen-tences extracted from slide 13 imply that changemanagement and training are major determinantsof organizational benefits from the ES-enableduse of IAS. The strength of this evidence thatovercoming organizational inertia leads to benefitswas judged to be limited (i.e., 1).

2 “My experience has shown that one of the number onedetriments to any post merger or post acquisition success ischange management. And our goal was to completelyeliminate that as an issue. And you will see that themerecurring throughout this presentation. So to that end, thefirst decision that was made was to utilize SAP—mySAP,actually—as the core application solution” (Case 33,transcript, p. 4).

In this quotation, the VP and CIO from FloridaCrystals, a U.S. $1 billion sugar producer andrefiner, specifically states that he believes thatchange management is a major determinant oforganizational benefits in this ES project. Thestrength of this evidence that overcoming organi-zational inertia leads to benefits was judged to bemoderate (i.e., 2).

3 “So if you look at a couple of the key-success factors…these are actually ranked in order. I have to say, thatexecutive sponsorship and leadership is number one. Thisthing…could have died a thousand deaths. Every timesomething happened, a tool didn’t work right, forecastingwas a little bit too complicated for the makers—‘Oh jeez, thedata is wrong’—that became a reason to kill the project. That’s how tough that was. So by having our executivesponsors there, CFO or CIO, we had our “executive supplychain”! Those guys really helped keep pushing this thingforward” (Case 54, transcript p. 7).

In this quotation, the director of ManufacturingVolume Strategy for Ondeo Nalco, a U.S. $2.6billion per annum water-treatment companyoperating in 126 countries around the worldexplains mechanisms for overcoming resistanceto change. The strength of this evidence thatovercoming organizational inertia (achievedthrough executive support) leads to benefits wasjudged to be strong (i.e., 3).

3 “Education: we spent a minimum of 20 hours on face-to-face training with an individual that would be just, let’s say,a plant operator, who would enter data from the floor, to 60or 70 hours for the more complex roles of a customer ser-vice representative entering orders and tracking shipmentsand so forth, to a supply-demand planning individual....Wealso had e-learning that was put out for our folks, so thatthey could on their breaks and free time go in and educatethemselves at their leisure” (Case 31, transcript, pp. 3-4).

In this quotation, the director of Global e-Transformation for Rohm and Haas, a U.S. $6billion manufacturer of coatings and adhesives,explains his firm’s efforts with respect to changemanagement and training. Since this expenditureon training presumes that training produces bene-fits, the strength of this evidence that overcomingorganizational inertia (through training) leads tobenefits was judged to be strong (i.e., 3).

3. Integration Please see Table 3 in the body of the paper for examples ofstrength of evidence where integration was judged to havecaused benefits.




4. Processoptimization

1

(Case 42, slide 22) Copyright © Lockheed Martin; used by permission.

In this slide, the speaker from Lockheed Martin, aUS$27B aerospace company, explains plans toimprove processes using SAP’s portal product. The nine and seven-screen processes on the rightare to be replaced by single portal screens. Sincethis new process had not yet actually producedany benefits, the strength of this evidence thatprocess improvement leads to benefits wasjudged to be limited (i.e., 1).

2 “Streamlining the purchasing organization. What we didwas the purchasing organization is now much morefocused. We reduced the staff from there and we have nowmore information from the divisions, as I told you. So wehave a smaller staff, doing a better job with better results.... When someone, any place in the bank, asks for something,he goes to the shopping cart; he puts it into the shoppingcart. It comes to our headquarters. It runs, it checks ifthere is a contract for that. If there is a contract, the con-tract is selected and the purchasing order is put on thesupplier immediately and it goes to the branch that is askingfor that. The same for services, so we have a more consis-tent process and it is very quick” (Case 49, transcript p. 5).

In this quotation, the speaker from Banco Itaú, a43,000- employee Brazilian bank, explains howtheir purchasing processes have been improvedusing SAP’s supplier relationship management(SRM) system. Strength of this evidence thatprocess improvement leads to benefits wasjudged to be moderate (i.e., 2).

3

(Case 24, slide 38) Copyright © Villeroy & Boch; used by permission.

In this slide, the speaker from Villeroy & Boch a€1 billion per annum German manufacturer ofhome interior products, explains how theircustomer service processes have been improvedusing SAP’s CRM system. As shown in thediagram, turnaround time dropped from 2 weeksin the upper half of the slide to 2 days in the lowerhalf using the CRM system. The strength of thisevidence that ES-based process improvementleads to benefits was judged to be strong (i.e., 3).

ROI: Old and New Process for Customer Service

-- Process redesign: to be faster with documents at the customer

BeforeCRM

WithCRM

consumer

installer

architect

Sales service

Mar-keting

Sales rep

address

service

docu-ment

service

consumer

installer

architect

Sales rep

Special-ized

servicecenter

docu-ment

serviceCRM Print server

consumer

consumer

2 days

2 weeks

Internet

phone

phone

Mail

Fax

letter

Internet

PhoneFax

Mail

Mail

FaxPaper




5. ImprovedAccess toInformation

1 “But probably the most interesting thing that happenedwasn’t necessarily planned, and that was all of a sudden wehad visibility. And I think this is probably the key. And whatI mean by visibility is, all of a sudden you could see exactlyin the order to cash process where something was beingheld up. So we had metrics now that we couldn’t even thinkup before. We now know that an order is held up becauseit’s in credit lock and you can tell how many days it’s beenheld up there” (Case 52, transcript p. 3).

In this quotation, the speaker from ChevronTexaco, a U.S. $100 billion plus oil companyexplains his firm’s use of SAP’s APO and R/3 IS-Oil solutions to increase visibility of the order-to-cash process. Because the amount of informationdoes not seem to have increased massively, thestrength of this evidence that improved access toinformation (through use of APO and R/3) leads tobenefits was judged to be limited (i.e., 1).

2 “It wasn’t easy, we had six months stabilization, particularlyin the business warehouse and reporting capabilities, thatwas probably the most complex piece....It took a lot longerthan we anticipated, and we did have performance issues,and with the system, we did have issues with our ownpeople in terms of performance who could not work on thenew platform, as much as the training was done in antici-pation.…The use of BW has been phenomenal for us interms of having information, different cubes, to do every-thing from investment analysis. We have about 300 reportsthat we use in BW right now with all our financial reportingfor our entities and the stand of the 80 entities, our balancesheets and income statements, and the kinds of analysisreports are all done very quickly and very easily. We knowthe data is good—so it’s been outrageously beneficial forus” (Case 23, transcript pp. 7-8).

In this quotation, the presenter from MassMutual,a Fortune 100 U.S. insurance company, explainsthat despite initial problems during stabilization,the use of SAP’s data warehouse (called Busi-ness Warehouse or BW) has been “outrageouslybeneficial.” Because of the initial problems, thestrength of this evidence that improved access toinformation using an ES leads to benefits wasjudged to be moderate (i.e., 2). (There were manysimilar cases where users of SAP’s BW productreported much better access to information; seethe following example.)

3 “The base foundation of everything is, in fact, the businesswarehouse. Business warehouse is the most critical appli-cation that we have. The company turns on data, but more,turns on information. Every application we have feeds BW. Purchase-to-pay-to-reporting, that was the scope of effortfor the project, from master data all the way through to pointof sales....Our decision process comes out of BW. I cannotsay enough about it. We are on [version] 3.0....The bestapplication within J.Crew is the business warehouse” (Case 4, transcript p. 2).“The most accurate, timely, actionable data that the com-pany has seen in years. Scott Rosen, CFO” (Case 4, slide15).

Here, the presenter from J.Crew, a U.S. $750million retail fashion chain, explains benefits fromSAP’s data warehouse. The strength of thisevidence that improved access to informationusing an ES leads to benefits was judged to bestrong (i.e., 3).

(Incidentally, there is a benefits-from-integrationstory in this quotation, too, that was coded SoE =2.)




6. On-goingimprovementprojects

1

(Case 43, slide 15) Copyright © Marathon Ashland; used by permission.

In this slide from Sapphire 2003, the presenterfrom Marathon Ashland Petroleum, presents his“wish list” of future projects following implementa-tion of SAP’s portal product for 12,000 users. This slide was treated as evidence that on-goingimprovement programs lead to benefits. Thestrength of this evidence was judged to be weak(i.e., 1).

2 “Phase two, internal sales force functional enhancementthat is what we’re dealing with as of today. That is coveringsales planning and forecasting. That’s based on SAPportals and SAP CRM as well. Order management formobile sales, so our sales forces are also able to createorders offline at the customer. Later on that will be repli-cated to our back bone, to our back office ERP system, andalso to our BW system.... So we’re integrating all of thesethree systems into one view to the sales force that theydon’t have to deal with different systems. It’s just oneapproach for them” (Case 45, transcript p. 5).

In this quotation, the speaker from Bosch Rexroth,a €3.6 billion German engineering company,explains his firm’s plans for phased implementa-tion of SAP’s CRM, portals, supply chain manage-ment, and data warehouse. Since implementationof these systems is expected to lead to greaterbenefits the strength of this evidence that on-going improvement programs lead to benefits wasjudged to be moderate (i.e., 2).

3 “And now we are going to roll that thing out over the rest ofGraybar’s geographic business units at least over next year,rolling it out to each one of these business units. And afterall that’s done, we are going to grade up at 4.60 and we willbe lucky to going through that upgrade until the turn of2004/2005. It’s never over. You never will see an end andstand up and say it’s completely finished. Again, and that isone of the reasons why you need to have a viable, healthypartner, because you are never finished. You get to the endof your implementation and you end up in an upgradestage. You get to the end of your implementation andsomebody wants a new functionality…so you are neverdone. This is an on-going, never finishable, never finishedkind of a war that we are living in. And this is a great job todo when you like doing that” (Case 34, transcript, p. 5).

In this quotation, the VP and CIO from GraybarInc., a U.S. $4 billion electrical distributor,explains why his firm’s implementation of SAP’ssuite of software (ERP, CRM, APO, BW) is not theend of the journey. Presumably, the on-goingimprovements discussed will lead to more bene-fits. The strength of this discussion as evidencethat on-going improvement programs lead tobenefits was judged to be strong (i.e., 3).

• Eliminate paycheck

• Direct deposit with debit card

• Labor utilization

• Leverage Business Warehouse

• Direct Interface with background check

• Leverage industry standards

• Benefit providers

• New opportunities

• Recruiting interface

Future Benefits



Appendix C

Examples of Strength of Evidence Judgments Concerning the Need for SuccessfulGo Live in order to Achieve Organizational Benefits from ES Use from thePerspective of Senior Management

SoE Example Explanation of SoE classification

1 Audience: “How frequently would you suggest meeting with

executive sponsors during an implementation?”

Presenter: “I will give you an example of what we’ve done. I don’t

have all the expertise in different implementations that people have

done. We actually have a very specific project structure…so we have

weekly meetings at the lower level of the implementation but we also

have weekly meetings with the leadership across all of our projects.

And actually we are having, I believe, every other week read outs to

the CEO staff on the status of the project and any specific escalations

or integration points that we are working—things that we know are the

critical success factors of the project. So it actually occurs fairly

frequently in our projects right now.” (Case 16, transcript p. 7)

In this transcript extract, the Senior Manager, Supply

Chain Planning, Avaya Inc., a U.S. $5 billion supplier of

telecommunications equipment and services, responds to

a question after his formal presentation about his organiza-

tion’s commitment to a project to reduce inventory by 45%

using SAP’s APO planning software. He makes no explicit

statement that successful go live is necessary to achieve

benefits, but the pursuit of that goal is implied by the clear

management interest in the project. For comparison with

the SoE scores for the six OBES hypotheses, the strength

of this example as evidence that Successful go live is

necessary to achieve benefits was judged to be limited

(i.e., 1).

2

(Case 40, slide 11) Copyright © Eveready; used by permission.

In this slide, the VP of the Kentucky Division of Hawaiian

Tropic a manufacturer of sun-care creams with a

“dominant position in U.S. and Canadian markets” (slide

4), outlines steps in the implementation of SAP’s All-in-

One ERP-style system for small businesses. This and

adjacent slides 10 and 12 (not reproduced here) provide

evidence of Hawaiian Tropic’s interest in having their

project deliver a working system. For comparison with the

SoE scores for the six OBES hypotheses, the strength of

the evidence in these three slides that Successful go live

is necessary to achieve benefits was judged to be

moderate (i.e., 2).

(The diagram in this slide is from SAP’s ASAP project

methodology. Many presentations contain ASAP

diagrams similar to this one.)

3 “From my personal standpoint it was one of the most challenging

years of my professional career. This project was a race, every

minute of it. I think the race started for me about the middle of

November of 2001 and it was a push. There never was a time when

there wasn’t an impending deadline every week or the week later.

We were racing every minute. It was tough. The team did absolutely

awesome, but it was very, very tough. It was one of the most difficult

things I’ve ever done” (Case 22, transcript, p. 5).

In this quotation, the Director of Manufacturing Volume

Strategy for Norske Canada, a 3,700-employee Canadian

paper manufacturer, explains his team’s commitment to

the project. Since this “awesome” effort was presumably

undertaken to achieve a successful go live and so achieve

benefits—why else would they work so hard?—the

strength of this evidence that Successful go live is neces-

sary to achieve benefits was judged to be strong (i.e., 3).

m yS A P A ll-in -O n e Im plem entatio n: W h ere W as th e T im e S p en t?

A ctivities : B u sin ess

P rocess M appin g w ith S A P B est P ractices

A ctiv ities : G o- L ive S upp ort In tern al S upp ort

2 w ee ks

5 w e eks

1 6 w ee ks

5 w ee ks

A ctiv ities : E nd- user

Tra in ing

A ctiv ities : S A P fun ction al

kno w led ge transfer B usiness cases tes ting

cyc les / In tegra tio n tes tin g

A ctiv ities : M aster data

c lean- u p D ata

C onversion M ap ping

S u m m ary To ta l T im e S p en t:• M aster D ata C lean -u p /D ata C on vers io n : 25%• B u sin ess P ro cess M ap p in g w ith m yS A P A ll -in -on e B est P ractices : 10%• B u sin ess P ro cesses /In teg ratio n T estin g : 45%• S u p er-u sers /E n d -u sers Tra in in g : 20%


Copyright of MIS Quarterly is the property of MIS Quarterly & The Society for Information Management and

its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's

express written permission. However, users may print, download, or email articles for individual use.

a multi-project model of key factors affecting organizational benefits from enterprise systems;...

Documents

sap products

enterprise systems es

sap netweaver

sued sap

products of sap ag

sap group products

sap business objects

organizational processes