Copyright National Association of Purchasing Management, Incorporated Winter 1994Anecdotal evidence suggests that electronic data interchange (EDI) is being used increasingly by companies, particularly to facilitate purchasing transactions. As this trend continues, one of the issues facing managers is understanding what effect adoption of the new technology will have on the organization as a whole, and on the purchasing department in particular. The results of this study, based on a survey of 122 purchasing managers, indicate that the impact of EDI is moderated by the volume of transactions using EDI. Its adoption requires somewhat different skills and training for buyers, and it also changes some aspects of a firm's relationship with its suppliers. The implications of these findings for purchasing executives are discussed.
INTRODUCTION
Electronic data interchange (EDI) needs little introduction today to most of the business community. The basic idea behind EDI is to replace the paper documents associated with routine business transactions--purchase orders, acceptances, and invoices--with electronic communication of such documents in a standardized format. EDI, therefore, is the movement of business documents electronically between or within firms in a structured, machine-retrievable, data format that permits data to be transferred, without re-keying, from a business application in one location to a business application in another location. In a fully developed EDI environment, a computer can use directly the data sent by other computers in electronic form.
Some researchers predict that by modifying an organization's structure and work processes, EDI can improve the way the firm does business, and subsequently it will become an essential technology for the firm. Some argue that the economics of EDI are so compelling that no firm today can ignore the competitive opportunity.(1)
Different firms have implemented EDI using different degrees of automation. Even in its simplest form, however, EDI offers some obvious advantages. Experience indicates that effective use of EDI can reduce inventory levels, improve cash flow and streamline a company's operations, and increase the speed and accuracy of information transmission.(2) EDI is also believed to help create closer ties between buyers and suppliers. Quick response to customers creates a real advantage for EDI users.(3) Under certain manufacturing conditions that require frequent deliveries, EDI can help reduce costs and monitor the transactions.(4)
Potential shortcomings of EDI include the difficulty of establishing an audit trail, absence of signatures on documents, and the unavailability of hard copies of documents.(5) Other potential problems may include the lack of data integrity and a perceived loss of control due to supplier initiation of some transactions.(6)
One of the functional areas that typically has a direct involvement with EDI is purchasing. Most EDI studies have evaluated the impact, benefits, and potential impediments of its adoption on the purchasing organization or its impact on the buyer-supplier relationship.(7) A 1985 study of EDI in purchasing predicted that "most purchasing departments of 1990 will look different from the typical purchasing departments of 1985"--because of the use of electronic communication of standard documents.(8) Another study predicted that EDI would improve the productivity of purchasing but would not significantly change the purchasing process--that is, the manner in which purchasing decisions are made and how the procedures are performed.(9)
While there have been several additional studies dealing with the impact of EDI on purchasing, it is useful at this time to assess the nature of this impact now that sufficient time has elapsed after initial EDI adoption by many firms during the mid-1980s. It is necessary to understand the subsequent impact of EDI on purchasing, along with the changes created in purchasing policies and procedures, to facilitate the development of future EDI operations. In view of the potential importance of EDI, this study was conducted to investigate the changes caused by EDI in purchasing policies and procedures and in the organization in general.
RESEARCH METHODOLOGY
The sample for this study was drawn from a partial membership list of the National Association of Purchasing Management (NAPM).* This list, broken down by SIC codes, contained approximately 5,500 names, with 1,463 from chemical and allied products (SIC Code 28), 748 from health services (SIC Code 80), 241 from transportation (SIC Codes 40, 42, 47), and 121 from retailing (SIC Codes 53 and 54). A proportionate sample was drawn from each SIC Code on the list, producing a total sample size of 2,000 purchasing professionals.
All members of the sample were sent a cover letter explaining the aim of the study, along with the questionnaire and a business reply envelope. A total of 318 responses were received; of the 318, 19 responses were incomplete and therefore were discarded. Of the remaining 299 responses, 122 were from EDI-users; these respondents form the basis for the study reported in this article.
As can be seen in Table I, (Table I omitted) the respondents represent a wide range of industries, including food and beverages, printing, publishing and allied industries, chemical and allied products, transportation, automotive, wholesale, general merchandise stores, banking and insurance, and business and health services. However, since proportionate sampling was used, the final sample contained a large representation of purchasing managers from two industries, which accounted for almost 40 percent of the names on the list--chemicals and allied products and health services--and relatively few from industries like transportation and retailing. The size of the companies, also shown in Table I, varies widely. About 24 percent of the respondents employ less than 50 employees, while another 23 percent employ more than two thousand employees. The revenue for 1991 was less than $50 million for about 33 percent of the respondents, while for another 27 percent it was more than $1 billion. All relevant demographic data for the respondents are summarized in Table I.
Data were collected using a two-part questionnaire. The first part contained questions related to broad aspects of information technology and was to be completed by all respondents. The second part included questions that were designed to study the impact of EDI purchasing environments; this section was to be completed only by respondents from companies that were EDI users.
The items used to measure the impact of EDI on purchasing were developed on the basis of discussions with consultants and on a review of current literature. The initial version of the questionnaire was first evaluated by a small group of independent consultants and researchers to check for its consistency, practicality, completeness, and validity. Changes were made based on these discussions. Subsequently, the revised questionnaire was pilot tested on a small number of purchasing executives to assess its clarity and readability. The results of these pilot tests were incorporated into the final version of the questionnaire.
Organizational characteristics such as number of employees, annual revenue, and so on were measured directly. All questions used to assess the impact of EDI were measured on a 5-point Likert scale (1=strongly disagree; 3=neutral; 5=strongly agree). These questions required respondents to indicate what changes in the purchasing function (e.g., ordering procedure, supplier relations, performance monitoring, payment terms, etc.) and what changes in the overall organization (e.g., power structure and information flows) had occurred as a result of the use of EDI operations.
Three different analyses were conducted. First, t-tests were used to assess the statistical significance of any reported impact of EDI adoption. For each question, mean responses were compared with the midpoint of the 5-point Likert scale, the value 3, to see if respondents' agreement or disagreement with the statement was significantly different from the mean of the scale. Second, the sample of EDI users was split into two groups based on whether the respondents encouraged their suppliers to use EDI (group 1) or did not encourage their suppliers to use EDI (group 2). This analysis was conducted to investigate whether the impact of EDI adoption was different depending on whether the organization encouraged or did not encourage its suppliers to adopt EDI.(10)
In the final analysis, the impact of EDI transaction volume was assessed in order to determine whether organizations with a higher EDI usage were impacted to a greater extent by its adoption. Transaction volume was measured by (1) the percentage of suppliers using EDI, and (2) the percentage of purchasing transactions. handled by EDI. In each case, the lowest third of the sample (group 1) and the highest third (group 2) were categorized as low and high EDI users, respectively. This split resulted in group 1 firms having 5 percent or fewer of the suppliers and transactions utilizing EDI and group 2 firms having 25 percent or more of their suppliers and their purchasing transactions using EDI. Hypothesis tests were conducted to determine whether perceptions of the impact of EDI adoption were different between the two groups in each set; this was done by comparing their mean scores for significant differences. In addition, the impact of various organizational characteristics (e.g., age of the EDI system, industry type, and so on) on EDI-related purchasing and organizational changes was analyzed.
RESULTS
Results of the study are presented in the following eight sections and are summarized in Tables II through VI. (Tables II through VI omitted)
CURRENT STATE OF EDI USE AND INVESTMENTS
The extent of EDI usage (measured by the percentage of purchasing transactions using EDI and the percentage of active suppliers using EDI) and the maturity of the technology in the company (measured by the length of EDI usage in the company) may significantly influence the outcome of EDI usage. In this sample, only 40 percent of the respondents had implemented EDI prior to 1990. Consequently, the remaining 60 percent had implemented EDI during the last three years. Such recent adoption of EDI by a large number of respondents simply confirms a growing acceptance of the technology in many industries. However, the extent of EDI usage in purchasing operations still is not very high. About 50 percent of the respondents use EDI for less than 10 percent of their total purchasing transactions; this is reasonably consistent with earlier estimates.(11) In spite of such low usage, though, 5 percent of the respondents use EDI for virtually all of their purchasing transactions.
The extent of EDI usage in a firm can also be judged by the percentage of active suppliers using EDI. In this study, 55 percent of the respondents have EDI links with 10 percent (or less) of their suppliers. Only 8 percent have EDI links with more than 50 percent of their suppliers. Therefore, there seems to be enormous potential for the growth of EDI even among current users.
It is generally believed that buyers encourage, and in some cases even force, their suppliers to adopt EDI. While the study found support for this view in that more than 50 percent of the respondents have either strongly or very strongly encouraged their suppliers to adopt EDI, it also found that about one-third of the suppliers had either strongly or very strongly encouraged the buying organization to adopt EDI.
IMPACT ON PURCHASING POLICIES AND PROCEDURES
The major thrust of EDI is to transmit electronically standardized business documents in such a manner that they can be acted upon with minimal human intervention. Such an environment requires a high degree of integration between the related functions of an organization. It may also require changes in operating procedures to facilitate such automation, and it may cause some changes in the job functions of the personnel involved. Procedural changes in some purchasing departments may involve changes in ordering and order approval, supplier selection, and performance monitoring.
In an automated environment, computer skills may become more important for selected personnel. Additionally, in some cases automation may require less frequent personal contact with suppliers which in turn may require different types of negotiation and interpersonal skills. The use of standardized formats may reduce flexibility in doing a job, and the electronic media clearly will require different audit and control procedures.
Seventeen items were developed to assess the impact of EDI on a purchasing department's policies, procedures, and people. These factors represent ordering approval and release procedures, staffing needs, job flexibility, and control-related issues. However, EDI-induced changes may be moderated by certain independent factors, such as type of organization (manufacturing or service), maturity of the system, the person championing EDI adoption, the extent to which the firm encouraged its suppliers to be actively involved with EDI, and the volume of transactions handled by EDI. The impact of these moderating factors is discussed later in the article.
Table II reports respondents' perceptions of the changes EDI adoption has caused in their purchasing policies and procedures. For each of the 17 items, mean scores were compared with the midpoint of the measuring scale (3=neutral) to identify those statements with which respondents agreed. The statistically significant changes that occurred as a result of EDI adoption are discussed in the following paragraphs.
It appears that companies prefer to use suppliers with EDI linkage capabilities--but, interestingly, such preferences do not significantly change supplier monitoring or automatic reordering activities. Buying organizations also require EDI orders to have paper trails. This suggests they desire human control even when an automated environment is preferred. Respondents also indicate strongly that EDI has not changed their payment terms or procedures, thus indicating that EDI is not yet fully integrated with an electronic funds transfer system.
The automated environment of EDI does not appear to reduce the perceived importance of negotiation skills for buyers. More important, it does not reduce the number of buyers required to do the job-or the number of people required in the purchasing department. This is somewhat surprising; because the automated environment of EDI generally is thought to reduce the burden of routine ordering for buyers, it is often assumed that EDI will reduce the number of buyers in the department. This study found that such reductions in purchasing personnel clearly have not materialized.
The need for computer skills among some buyers in an EDI environment may explain the expressed need for changes in the buyer training program under EDI, although the rigid format of EDI apparently does not change the operational flexibility of the purchasing department. Interestingly, buyers are not unhappy about less frequent personal contact with supplier representatives in an EDI environment, which some believed might occur.
IMPACT ON THE ORGANIZATION
The use of an EDI system clearly has the potential to impact many aspects of an organization. This section reports the organizational changes caused by EDI as perceived by purchasing executives. This perspective, when combined with results from other studies, should expand the understanding of EDI's potential impact throughout an entire organization.
Nine potential organizational changes thought to occur as a result of EDI adoption were identified and purchasing managers' responses to them measured. These results are presented in Table III (see p. 36). Implementation of an EDI system appears to act as a catalyst in moving firms toward increased automation. Such automation typically requires closer cooperation among the various departments and eventually should promote integration of various functions in an organization. EDI also has tended to modify the information flow in a typical organization. And, clearly, EDI is viewed as a tool for achieving an advantage over competing firms, a finding that is consistent with the results of a previous study.(12)
SIGNIFICANCE OF THE AGE OF AN EDI SYSTEM
As a system matures, many of the problems associated with the early stages of its implementation eventually are resolved. Therefore, one might expect that the companies that have been using EDI for a longer period of time will have had a different set of experiences than those firms that have only recently adopted a system.
Since EDI is a relatively new field, however, a majority of the respondents had implemented EDI only within the last five years. To study the impact of age, then, users where EDI had been in place for more than three years were compared with those with newer systems. Surprisingly, the results showed no significant differences between the two groups on any of the purchasing or organizational factors studied. This suggests that either the impact of EDI does not change significantly with time, or that most of the companies are still in the fairly early stages of adoption and the EDI system as a whole has not yet matured enough to reveal any differences.
INDUSTRY TYPE AND EDI IMPACT
To determine whether the impact of EDI usage varied across industries, respondents were segregated into two groups--manufacturing firms and service firms. When comparing the mean responses of these two groups it was found that EDI had stimulated more reorganization in the service sector firms than in the manufacturing firms. However, even for the service firms, the changes were not statistically significant. In addition, EDI had also modified the organizational power structure somewhat in the service sector firms, although not significantly so from a statistical point of view. So, in summary, the impact of EDI on purchasing activities and on the overall organization is no different between firms in the manufacturing and the service industries.
INFLUENCE OF EDI PROPONENTS IN THE COMPANY
It was thought that the proponents and policy developers for EDI might also influence the specific usage of EDI in a company. In this study, 55 percent of the respondents indicated that their EDI policy, as related to purchasing activities, was developed by purchasing department personnel--and in the remaining 45 percent of the firms the purchasing EDI policy was developed by top management or by information system specialists. Here again, there was no evidence of any significant differences between companies in which EDI adoption was championed by the purchasing department and those in which it was championed by nonpurchasing EDI proponents. This suggests that the operating impact of EDI does not depend on the people who were responsible for developing policies regarding its adoption and use.
INFLUENCE OF SUPPLIER ENCOURAGEMENT
Contrary to the popular belief that it is the buying organization that generally requires suppliers to adopt EDI, this study found that often suppliers also encouraged buyers to develop EDI capabilities. To assess whether the source of encouragement had any influence on the impact of EDI adoption, mean responses of respondents that were encouraged by their suppliers to adopt EDI were compared with the responses of those firms that were not. Additionally, respondents that encouraged their suppliers to adopt EDI were compared with those firms that did not. The results of the study found no significant differences in the impact of EDI adoption between the respondents that were encouraged by their suppliers to adopt EDI and those that were not. However, some significant differences were found between the respondents that encouraged their suppliers to adopt EDI (group 1) and those that did not (group 2).
As can be seen in Table n! group 1 companies had a significantly higher preference for using EDI suppliers than did group firms. They also reordered automatically more often and they placed orders more frequently. Similarly, group 1 companies required fewer signatures in the ordering process and were comfortable with the paperless environment of EDI. Whereas most of the companies in the sample made some significant changes in their buyer training program to accommodate EDI, such changes were significantly greater among the companies that encouraged their suppliers to adopt EDI. Similarly, while many companies indicated a movement toward more automation, this trend is stronger for group 1 companies. Clearly, then, there are some significant differences between the firms that encourage their suppliers to develop EDI capabilities and those firms that do not do so.
ROLE OF EDI TRANSACTION VOLUME
The results shown in Table V (see p. 38) suggest that transaction volume (as measured by the percentage of suppliers using EDI) plays a significant role in moderating the impact of EDI. It appears that EDI has necessitated greater changes in buyers' training programs and greater changes in payment terms for high volume users (group 2) than is the case for low volume users (group 1). In addition, lower volume users believe that computer systems have become more complicated and more difficult to understand; they are more concerned about the systems' vulnerability to inappropriate usage; and they strongly perceive EDI to be a tool that can be used to create strategic advantage.
As can be seen from the data in Table VI (p. 39), EDI has a differential impact on high volume users, as compared with low volume users, when measured by the percentage of transactions using EDI. Relatively speaking, high volume users (group 2) enjoy the benefits commonly associated with EDI usage to a greater extent than do low volume users. Typically they do not require as much approval of routine orders; they clearly prefer to work with EDI linked suppliers; they use more standardized product labeling; and they place smaller orders more frequently with EDI suppliers. Although all firms studied state that EDI usage does not reduce buying and other purchasing personnel requirements significantly, high usage firms have more positive experiences in this regard than do low usage firms. The same thing appears to be true with respect to negotiating skill requirements for buyers.
CONCLUSION
This study found that utilization of an EDI system is associated with some changes in purchasing policies and procedures, and that the impact is more strongly perceived by high volume EDI users. EDI has resulted in some changes in training programs for buyers, as computer skills for certain types of buyers have become more important. High transaction volume EDI users report that the EDI-induced automation can result in some reduction in the number of buyers and in the staffing needs of the purchasing department, although most firms have yet to achieve this goal. At the same time, the high volume users seem less concerned about the complexity and security issues sometimes associated with EDI adoption.
Somewhat surprisingly, however, while both high volume and low volume users see EDI as a tool for achieving a strategic advantage over competitors, as a group low volume users perceive this benefit more strongly. This suggests that when EDI becomes the norm rather than the exception, its impact is more clearly felt. This implies that to benefit fully from EDI, an organization should increase both the percentage of its suppliers using EDI and the percentage of its purchasing transactions conducted via EDI.
A 1985 study reported that most purchasing departments would dislike having an MIS department developing EDI policies for communicating with suppliers, and that electronic environments could lead to buyer dissatisfaction as a result of less frequent personal contacts with supplier personnel. The present study found that the department developing the EDI policy for purchasing transactions has no impact on the changes created by EDI in purchasing activities, and that buyers were not unhappy about less frequent personal contact with suppliers, when this in fact occurred. However, the findings of this study confirm the predictions of the previous study that buyers would function cautiously in an electronic environment and would not reduce monitoring and control of supplier performance. This study found that EDI orders still require a paper audit trail, and that most reorders still require conventional authorization.
The study found that the firms in which routine orders no longer required regular approval were also firms in which supplier monitoring was reduced, repetitive ordering did not require conventional authorization, ordering and scheduling activities were integrated, and the paperless environment of EDI was not perceived as a problem. These types of companies seem to have taken a lead in truly implementing an automated integrated environment.
The study also found that in some companies a reduction in the number of buyers used and a reduction in the level of supplier performance monitoring appeared to go together. Therefore, a reduction in supplier monitoring, and not simply in EDI implementation, in all probability was responsible for reducing the staffing needs. EDI tends to promote a long-term buyer-supplier relationship that can lead to increased mutual trust. Consequently, in the long run, use of EDI may help in reducing the number of personnel required in the purchasing department.
Earlier studies predicted that by 1990 a major portion of purchasing transactions would be conducted electronically. Although such a prediction has failed to materialize, the use of EDI is definitely increasing. And to fully enjoy its benefits, firms should enhance most buyers' basic understanding of computer applications. In reality, an electronic environment does not reduce the operational flexibility of a purchasing department, nor does it dehumanize the purchasing process. An integration of EDI with appropriate functions such as planning, scheduling, and accounting should provide an added competitive advantage for an organization. Therefore, in order to realize the full benefits of EDI implementation, firms should strive to achieve greater integration of appropriate functional areas, and expand their use of supplier certification programs that reduce supplier follow-up and inspection activity requirements.
While this study reveals some of the purchasing-related changes associated with EDI use, additional research is needed to shed light on the impact of EDI utilization on the effectiveness of purchasing performance. In addition, future studies should investigate EDI advantages and disadvantages on an industry-specific basis. While the presence of a large number of respondents from several industries in this study may have biased the results somewhat, they nevertheless provide a valuable starting point for further research into the impact of EDI on the purchasing function.
*The partial membership list was provided by the Center for Advanced Purchasing Studies.
REFERENCES
1. Peter Keen, Shaping the Future (Cambridge, Mass.: Harvard Business School Press, 1990).
2. B. Dearing, "The Strategic Benefits of EDI", Journal of Business Strategy, vol. 11 (January-February 1990), pp. 4-6; Robert Monczka and J. Carter, "Implementing Electronic Data Interchange," Journal of Purchasing and Materials Management, vol. 25, no. 1 (1989), pp. 26-33; J.V. Hansen and N.C. Hill, "Control and Audit of Electronic Data Interchange," MIS Quarterly, December 1989, pp. 403-13.
3. J.J. Mohr, "Computerized Communication in Interorganizational Relationships: Its Impact on Structure, Conduct, and Performance," AMA Summer Educators Proceedings, 1990.
4. S. Banerjee, D.Y. Golhar, "EDI Implementation in JIT and Non-JIT Manufacturing Firms: A Comparative Study" International Journal of Operations and Production Management, forthcoming.
5. Hansen and Hill, op. cit.
6. W. Powers, EDI Control and Audit Issues (The Electronic Data Interchange Association, Alexandria, TDCC/EDIA Monograph, 1989).
7. R. O'Callaghan, P.J. Kaufmann, and B.R. Konsynski, "Adoption Correlates and Share Effects of Electronic Data Interchange Systems in Marketing Channels," Journal of Marketing, vol. 56 (April 1992), pp. 45-56.
8. B.J. LaLonde and M.A. Emmelhainz, "Electronic Purchase Order Interchange," Journal of Purchasing and Materials Management, Fall 1985, pp. 2-9.
9. M.A. Emmelhainz, "Electronic Data Interchange: Does It Change The Purchasing Process?" Journal of Purchasing and Materials Management, Winter 1987, pp. 2-8; M.A. Emmelhainz, Electronic Data Interchange: A Total Management Guide (New York: Van Nostrand Reinhold, 1990).
10. EDIA, EDI in North America: Status of Usage and Technology, (Electronic Data Interchange Association, 1989).
11. D.A. Hough, "Editorial Comments," EDI World, March 1991, p. 4.
12. R.I. Benjamin, D.W. Long, and M.S. Scott-Morton, "Electronic Data Interchange: How Much Competitive Advantage?" Long Range Planning, vol. 23, no. 1(1990), pp. 29-40.
Ven Sriram is an Associate Professor at Morgan State University in Baltimore. He received his Ph.D. degree from the University of Maryland, College Park. Dr. Sriram's research interests include international marketing, the impact of information technology on marketing and marketing strategy.
Snehamay Banerjee is an Assistant Professor at Drexel University. He holds a Ph.D. degree from the University of Maryland, College Park. Dr. Banerjee's research interests include knowledge-based systems and their uses in management science, model management, and electronic data interchange.