COMMUNICATIONS OF THE ACM April 2000/Vol. 43, No. 4 27

JEA

N-F

RA

NC

OIS

PO

DEV

IN

UUntil recently, the majorERP vendors (SAP, Ora-cle, Peoplesoft, JDEdwards, and Baan)were mainly targetingthe high end of the mar-ket (companies withmore than 1,000employees), but this

market comes close to saturation. Many large com-panies have already adopted ERP systems and areplanning the next step of how to use the installedERP infrastructures as foundations for e-business [1,2]. Most of the small- and medium-sized companiesstill have to make the decision to deploy ERP. Themidsize market is an interesting market, for exam-ple, the number of midsize companies (501,000

employees) in Europe is estimated to exceed100,000. Data from our research shows that withaverage annual IT budgets of more than $500,000,the total European midsize market for IT productsand services surpasses a staggering $50 billion peryear. This market as a whole is very attractive for themajor ERP vendors. However, since the wave ofadoption by midsize companies is in its early stages,little is known about developments and drivers thatform the basis of ERP adoption decisions. The pur-pose of this article is to understand developments inERP adoption within the European mid-market.

Our empirical information is based on a large-scaleEuropean multicountry/multi-industry survey con-ducted in mid-1998. Based on the survey data, wewill address various issues, such as: How did ERPpenetration in the mid-market develop until 1998

Yvonne van Everdingen, Jos van Hillegersberg, and Eric Waarts

EERRPP AADDOOPPTTIIOONNBBYY EEUURROOPPEEAANNMMIIDDSSIIZZEECCOOMMPPAANNIIEESS

Searching for ERP systems offering a perfect fit.

and what can be expected in 2000? Furthermore, weexplore criteria used by European midsize companiesfor investing in ERP systems and for choosing anERP supplier.

ERP Penetration DevelopmentThe survey included questions both on current(mid-1998) adoption of ERP and planned adoptionwithin two years (mid-2000). In 1998over allcountries and industriesalready 27% of the Euro-pean midsize companies had ERP software installedin one or more functional areas. The functional

areas included in the interview were purchase andsales order management, inventory and materialsmanagement, production and assembly, transporta-tion, service and maintenance, marketing and sales,warehouse management, financial accounting, andhuman resource management. Interestingly, aminority (13%) of the adopters used ERP softwarein just one functional area, while most companies(70%) used it in more than three functional areas.Thus, companies clearly aim at using ERPs mainstrength: integrating several functional areas. Thedata further reveals that in nearly all companies, thevarious functional areas are automated using ERPsoftware of a single vendor. Hence, difficulties inintegrating ERP software from different vendorsoften seen in large companies are not yet an issue formidsize companies. This may change, as midsizecompanies will increasingly integrate their supplychains, and thus communicate with suppliers thatare likely to have adopted different ERP solutions.

With respect to expected further penetration of

ERP software our estimations are based on invest-ment indications by the respondents. Among thefirms that did not have ERP software installed in1998, about 40% indicated intention to invest inERP before mid-2000, yielding an expected averagepenetration level of 56% over all countries and indus-tries. If we consider average market prices for ERPsoftware, the total European ERP mid-market can beroughly estimated to exceed $5 billion per year for1999 and 2000. It must be borne in mind that invest-ment intentions are not necessarily equal to actualbehavior, but one may cautiously estimate that during

this year, a majority of all Euro-pean midsize companies willembrace ERP.

Weve shown average penetra-tion percentages across all coun-tries and industries. However,the data also reveals some inter-esting differences between Euro-pean countries and industries.Figure 1 shows the penetrationdevelopments for each country.In 1998, Sweden, Denmark, andThe Netherlands were far aheadwith penetration rates of 45%and higher, while U.K. andSpain were lagging behind withpenetration rates less than 20%.The data also shows the penetra-tion of ERP software is expectedto grow, especially in countriesthat showed low 1998 adoption

figures. Nevertheless, the expected penetration ratesin 2000 still are especially high in the Northern Euro-pean countries: Norway, Finland, Sweden, Denmark,and The Netherlands. This is an interesting phenom-enon, which may be related to certain cultural factors.Cross-cultural research by Hofstede reveals fourdimensions of cultural characteristics of a countrythat can be used to group European countries intothree clusters [3, 5]. One of these clusters includes thefour Scandinavian countries and The Netherlands,which are characterized as having a low level of uncer-tainty avoidance, a high level of individualism, a smallpower distance, and a low level of masculinity. Thiscluster is recognized as the most innovative clusterwith relatively weak resistance to new products and astrong desire for novelty and variety. There is no solidproof of a causal relationship here, but these generalcultural characteristics of the Scandinavian countriesand The Netherlands may have led to a higher levelof ERP trial and adoption. The relatively low ERPpenetration in Great Britain could similarly be

28 April 2000/Vol. 43, No. 4 COMMUNICATIONS OF THE ACM

Figure 1. Penetration development across countries (n=2647).

Percentage of companies

Norway

Sweden

Finland

Denmark

Netherlands

U.K.

Belgium

France

Italy

Spain

0.0

10.0

20.0

30.0

40.0

50.0

60.0

70.0

80.0

90.0Expected ERP Penetration 2000

ERP Penetration 1998

explained by the fact that Great Britain (togetherwith, among others, Ireland), belongs to a cluster thatis described as a fairly risk-averse market. This indi-cates companies from these countries tend to wait toadopt a new product until a large group of other com-panies has already adopted it.

In addition to country differences, the data showsdifferent penetration rates between industries in 1998(see Figure 2). They run from about 20% in theProject industry and the Wholesale industry up tonearly 40% in Discrete and Automotive. The fact thatERP has its roots in manufacturing is likely to explainthese differences. Expected penetration levels for 2000

show, however, that the formerindustries will make up arrearssince many firms in these sectorsare planning to implement ERPin the near future. Interestingly,Electronics is expected to becomethe leader in ERP implementa-tions, with an adoption percent-age of nearly 70%.

Information SystemsSelection CriteriaWhen companies consider buy-ing a new product, the perceivedcharacteristics of the productplay an important role in thefinal decision to start using thatnew product [6]. Thus the char-acteristics of the ERP softwarehave to match the criteria usedby companies to select an infor-mation system. In the survey,respondents were asked to indi-cate the three most importantcriteria for selecting their currentinformation system (lowest cost,user-friendliness, fit with busi-ness procedures, scalability, sup-port, and training). The resultsshow the most important crite-rion used in selecting an infor-mation system is the best fit withcurrent business procedures (seeFigure 3). About one-half of therespondents have mentioned thebest fit in the top three criteriaranking, while more than one-third mentioned it as the singlemost important one. Hence,compatibility with the businessprocedures is the major issue for

companies to decide on a new system. AlthoughERP vendors have given much attention to alloweasy configuration of their packages to match exist-ing business processes, several studies have shownthat configuring and implementing ERP systems canbe costly, and may even require reengineering entirebusiness operations [4], (see also other articles in thisspecial section, for example [7]). Given the largepotential of the midsize market and the limited bud-gets of midsize companies, there is much potentialfor an ERP package that succeeds in meeting thismain selection criterion. Other important selectioncriteria are flexibility, cost and user-friendliness of the

COMMUNICATIONS OF THE ACM April 2000/Vol. 43, No. 4 29

Figure 2. Penetration development across industries (n=2647).

Industry

Percentage of companies

discrete and automotive

electronics

food and beverage

process industry

wholesale

project industry

0.0

10.0

20.0

30.0

40.0

50.0

60.0

70.0Expected ERP Penetration 2000

ERP Penetration 1998

Figure 3. Criteria for information systems selection (n=2401).

Support

Scalability

User-friendliness

Cost

Flexibility

Relative importance

Fit

system, and to a lesser extent scalability and suppliersupport. The importance of criteria is very similaracross all industries, but some differences betweencountries were found. For example, Italian andSwedish companies focus relatively more on user-friendliness, while Spanish companies score the high-est on flexibility. Companies from The Netherlandsjudge the total cost of implementing an informationsystem as important as the fit with current businessprocedures.

ERP Supplier Selection CriteriaIn addition to criteria to select a new system, other cri-teria may be used to select the supplier of a new infor-mation system. The data shows that European midsizecompanies tend to focus on product characteristicsrather than on characteristics of the ERP supplier of theproduct. It makes little difference whether the vendor

is a market leader, an internationaloriented company, or a companywith a superior corporate image.Companies predominantly look atthe functionality and quality ofthe products and services for eval-uating ERP suppliers, which hasbeen found across all lines of busi-ness and all countries (see Figure4). To a somewhat lesser extent,the speed of implementation, thepossibilities of the product forinterfacing with other applica-tions, and the price of productsand services also are importantsupplier selection criteria.

Supplier selection brings us tothe final issue in this article, regard-ing the market positions of the var-ious ERP suppliers. The dataprovides information about the

awareness of the various ERP vendors among the mid-size companies in Europe (circa mid-1998), their shareof the market, and their nomination as preferred sup-plier for future investment. It should be noted, how-ever, that since the time of the survey market positionsmay have changed substantially due to various circum-stances. For instance, Baan has had significant negativepublicity over the past two years, which may enhancethe companys awareness level, but also may have wors-ened their position as preferred future supplier.

At the time of the survey, the penetration of thefive major ERP vendors (SAP, Oracle, Peoplesoft, JDEdwards, and Baan) in the European midsize marketwas rather modest. Obviously, the penetration of ERPat that time was in its early stages. About half of com-panies still preferred their in-house developed, tailor-made information systems to the standardized systemsof the large ERP vendors. Among the adopters of ERP,the combined market share of the big five vendors wasslightly more than 20%. Regarding specific vendors,SAP had both the highest awareness (that is, it was themost often spontaneously mentioned supplier byrespondents) and highest share in the European mid-size market. Roughly 11% of the companies had SAPinstalled. The European-based ERP vendors (SAP,Germany; Baan, The Netherlands) clearly had ahigher level of awareness among the European compa-nies than U.S.-based ERP vendors. SAP and Baanwere mentioned spontaneously by 60% respectively50% of the companies, followed by JD Edwards(33%), and Oracle (28%). Peoplesoft was hardlyknown in 1998 and had a negligible share of the mar-ket. Based on the respondents indications concerning

30 April 2000/Vol. 43, No. 4 COMMUNICATIONS OF THE ACM

Figure 4. Criteria for supplier selection (n=2623).

Importance

Criteria

Functionality product

Quality product

Implementation speed

Interface other systems

Price

Market leadership

Corporate image

International orientation

Figure 5. First considered future ERPsupplier (n=1574).

SAP Baan

Peoplesoft Other

Oracle JD Edwards

39% 25%

12%

18%5%1%

preferred suppliers for future investment in ERP sys-tems (see Figure 5), it can be expected that the com-bined share of the big five vendors will sharply increaseto over 60%. SAP is expected to be the most preferredERP supplier among the European midsize compa-nies; Oracle is expected to strengthen its position.

ConclusionWhile ERP has been adopted by most of the largecompanies, the midsize market is now catching up.Given the large number of midsize companies, thisadoption wave represents a huge number of ERPimplementation projects that have been recently com-pleted, are in progress, or are about to begin. Theseprojects will raise mid-market ERP adoption from ourmeasured 27% in 1998 to the expected 56% this year.Both industries and countries that showed low ERPadoption in 1998 will decrease the gap considerably.

This, of course, constitutes a very interesting mar-ket for ERP vendors to penetrate. A number of issuesemerge from our study that are critical for the devel-opment of this market. From the viewpoint of clients,the fit with current business processes is the mostimportant selection criterion for a new system. At thesame time, companies within the mid-market rate alow price and short implementation times as highlyimportant. Some of the current ERP vendors areattempting to satisfy these requirements by offeringaccelerated implementation methods. However,these methods are usually based on offering a minimalfit, thus ignoring the number-one wish of small andmedium enterprises. Therefore, vendors aiming topenetrate in this market should offer a new generationof ERP systems that is flexible enough to easily han-dle a wide variety of business procedures. Further-more, they should have sufficient interfacingcapabilities in order to reduce the risk of vendor lock-

in and cost of switching. Finally, vendors should takeinto account that the European midsize business mar-ket is not homogeneous. This study shows that signif-icant differences exist in selection criteria amongcountries. However, for ERP vendors succeeding insatisfying these various requirements, the huge mid-size market will be rewarding.

References1. Foley, J. ERP and E-Business: Perfect Together?, www.information-

week.com, (Sept. 13, 1999), 169. 2. Gilbert, A. and J. Sweat, Reinventing ERP, www.informationweek.com,

(Sept. 13, 1999), 1520. 3. Hofstede, G. Cultures consequences: International differences in work

related values. Beverly Hills, Sage, 1997. 4. Jarosh, T.J. AS/400: The high scorer in ERP. Midrange Systems (Apr.

1999), 3637. 5. Kale, S.H. Grouping euroconsumers: A culture-based clustering

approach. J. International Marketing 3, 3 (1995), 3549. 6. Rogers, E.M. Diffusion of Innovations. The Free Press, a division of

Macmillan, 1995. 7. Soh, C, Kien, S.S., Tay-Yap, J. Cultural fits and misfits: Is ERP a uni-

versal solution? Commun. ACM 43, 4 (Apr. 2000).

Yvonne M. van Everdingen (yeverdingen@fbk.eur.nl) is a member of the faculty in the Department of Marketing Management, Rotterdam School of Management, Erasmus University, The Netherlands.Jos van Hillegersberg (jvh@acm.org) is a member of the faculty in the Department of Information and Decision Sciences, Rotterdam School of Management, Erasmus University, TheNetherlands, and a component manager at AEGON Bank.Eric Waarts (ewaarts@fbk.eur.nl) is a member of the faculty inthe Department of Marketing Management, Rotterdam School ofManagement, Erasmus University, The Netherlands.

The survey was conducted at the request of one of the major ERP vendors, but was car-ried out independently of that company. We are grateful to this vendor for their sup-port of this research.

Permission to make digital or hard copies of all or part of this work for personal orclassroom use is granted without fee provided that copies are not made or distributedfor profit or commercial advantage and that copies bear this notice and the full citationon the first page. To copy otherwise, to republish, to post on servers or to redistributeto lists, requires prior specific permission and/or a fee.

2000 ACM 0002-0782/00/0400 $5.00

c

COMMUNICATIONS OF THE ACM April 2000/Vol. 43, No. 4 31

The survey was performed using a prestructuredquestionnaire that covered questions aboutERP adoption decisions, general company policiesconcerning IT, and investment planning. Becausevarious European countries were included in thesurvey, the questionnaire was translated into locallanguages by translation agencies. Subsequently,telephone interviews were performed with IT managers and financial managers responsible for ITdecisions. The interviews were carried out by professional call centers. In total, a substantialsample of 2,647 companies were interviewed from10 European countries (Finland, Sweden, Norway,

Denmark, the Netherlands, Belgium, France, Spain,Italy, and the U.K.) in six industry sectors (discrete and automotive industry, projectindustry, electronics industry, process industry,food and beverage industry, and wholesale industry). The countries and industries wereselected in cooperation with the sponsor of theresearch. For each country/industry combinationan average 45 companies were interviewed. Theywere randomly selected from local industry databases. This procedure ensures that for eachcountry the total sample size exceeds N=200, andfor each industry the sample size is over N=300. c

How the Survey was Conducted