Jeffrey J. Johnson JJOHNSON@b202.usu.edu is an assistant professor in the Business Information Systems and Education Department, Utah State University.

---

If you do not like the background color, you can change it by highlighting the color you prefer in the scroll box below.

White Aquamarine Azure Bisque Cadetblue Coral Cornflowerblue Cornsilk Darkorange Darkseagreen Forestgreen Gold Honeydew Indianred Lavender Lemonchiffon Lightgreen Lightpink Lightsteelblue Limegreen Linen Mediumslateblue Mediumspringgreen Mintcream Mistyrose Olivedrab Orange Orangered Palevioletred Papayawhip Peachpuff Rosybrown Seashell Silver Skyblue Snow Tomato Thistle Turquoise Violet Wheat White Whitesmoke Yellow

Introduction

Organizational learning has been one of the buzzwords of the 90s. A learning organization is an organization that is continually expanding its capacity to create its future by joining adaptive learning with generative learning, seeking out and mastering change (Senge, 1990, p14). Systematic thinking, protection from blame, reduced bureaucracies, increased communication, and freedom to experiment promise increased competitive advantage. Information systems that support organizational learning have been developed and tested by industry and researchers in recent years. Vendors of some groupware, internal web, and electronic meeting products claim organizational learning as a benefit of their systems (International Data Corp, 1995). Theoretical and other scholarly treatments of organizational learning emphasize intangible and long-term benefits (Senge, 1990).

Information systems that support organizational memory (OM) and other facets of organizational learning have been developed and tested by researchers in recent years.

Vendors of some groupware products claim OM as a benefit of their systems, for example, Lotus Notes (Summary, 1995). Theoretical and other scholarly treatments of the organizational learning and memory topics often emphasize intangible and long-term benefits. A significant problem for researchers and commercial developers is that tangible gains and immediate benefits of organizational learning are difficult to identify (IBM, 1994). In spite of the dearth of information about short-term "selling points," many organizations do employ IT to support the organizational learning function. But there has been very little research on the influence of information technology on organizational learning (Balasubramanian, 1995).

The purpose of our study was to identify organizations that use IT to support organizational learning and organizational memory, discover how they are using IT, and how they justify the investment. This paper reports on our study.

In recent years OM has enjoyed some popularity as a topic for research and scholarly discussion, particularly with respect to information technology. Huber (1991) writes that expert systems seem ideally suited to a major role in maintaining OM, although Levitt and March (1988) warn against their use. Other researchers have chosen to learn about the concept by developing information systems to support and create various types and aspects of OM. One example is the work by Morrison (1992). She developed a system to support team and organizational memory, targeted specifically at supporting business teams in the context of projects. The system focused on project management, meetings, and decisions and actions that take place between team meetings, emphasizing longitudinal support for group projects. Another interesting information system approach is the "Answer Garden" (Ackerman and Malone, 1990). This system "grows" organizational memory by storing answers to questions posed by users. If the system does not have a satisfactory answer, it sends the question to an expert who then provides the answer, which is stored in the database for future reference. A newer approach uses internet technology and a web browser to create a corporate memory infrastructure (Huynh, Popkin, and Stecker, 1995)

Organizational memory, whatever its form, is generally regarded by all of those who write about it as a resource that can be used to enhance current organizational decision making. In fact, Walsh and Ungson (1991) define organizational memory in terms of its relationship to decision making as "stored information about a decision stimulus and response that, when retrieved, comes to bear on present decisions (p. 61)." El Sawy, et al., (1986) refer to the impact of historic information on managers' understanding of present decision contexts and visions of the future. Morrison (1992) concludes that organizational memory is "a device for integrating organizational knowledge and providing information from the past that can potentially aid present decision making situations (p. 52)." Huber (1990) includes organizational memory in his discussion of technologies to support communication and decision making. Stein and Zwass (1995) present an extensive discussion on methods and means of implementing organizational memory via information technology. A well-developed bibliography on organizational memory is presented by Ackerman (1996.)

We designed this study with a two-pronged approach: First, we identified organizations that use information technology for organizational learning and memory by means of a survey, employing a questionnaire. The main purpose of the survey was to identify organizations that perceive some kind of value or positive return from information technology that supports OM. Second, we contacted respondents who indicated their organizations both use IT for OM and perceive value from the investment, and followed up the questionnaire with site visits and/or face-to-face interviews. This approach was useful in providing a view into the research domain characterized by both breadth and depth.

An important problem in studying organizational use of IT systems for learning and memory is identifying organizations that employ IT for those purposes. We were able to identify several organizations by means of a questionnaire administered during a conference on Management Information Systems. Conference attendees (typically information systems professionals and managers) responded to a one-page questionnaire during the luncheon meeting. The questionnaire simply asked respondents whether their organizations employ information technology to facilitate organizational learning, prompting them with several examples of systems that could be used for such a purpose. It also asked whether the indicated use of IT provided a positive effect or return on investment (ROI). Finally, the questionnaire solicited names and addresses of respondents for future site visits. The questionnaire is reproduced in the Appendix with the number of "Yes," "No," and unanswered responses.

The seminar was geared toward practitioners representing various organizations in business and industry, education, government, and the military. Because no attempt was made to randomize selection of respondents or responses, the survey results may not be statistically generalizable. However, the purpose of the survey was more to identify potential interview candidates than to produce generalizable results. The survey did enable us to identify contacts at organizations where IT is employed to support OM, and was therefore considered successful. Also, we used the results of the survey to help develop our interview questions.

We received sixty usable responses out of 150 questionnaires distributed, for a response rate of 40%. Respondents represented organizations from industry, education, government, and the military in areas of the Western United States.

We asked whether their organizations used IT to support organizational learning in four areas identified by Huber (1991). Roughly half (49.7%) of respondents indicated their organization uses IT for information distribution. About two fifths use IT for either information interpretation (39.6%) or knowledge acquisition (37.1%). Only about one sixth (16.3%) said they use IT to support organizational memory. Figure 1 shows which organizational learning constructs are supported most by IT.

Among those organizations employing IT for organizational memory, less than half (48.7%) reported that their use of IT to facilitate organizational memory has produced a positive effect or return on investment. The respondents reported positive effects more often for the other three learning constructs: Information distribution 69.7%; knowledge acquisition 67.4%, information interpretation 63.6%. Several respondents indicated that the effects of IT use for organizational learning were simply not measured in their organizations. Figure 2 shows which IT supported organizational learning constructs were perceived to have a positive effect or ROI.

As might be expected, E-mail, internal webs, and groupware for information distribution were the most used information technologies. Computer-aided-instruction for information interpretation, decision support systems and executive information systems for knowledge acquisition, and document delivery systems for information distribution were the next most utilized. As the purpose of systems moves from the relatively easy job of information distribution and interpretation to the more difficult jobs of knowledge acquisition and organizational memory the use of information technology systems appears to decrease.

Because the sample was not randomly selected, generalization is difficult. We can say that respondents represented a wide variety of business, government, education, and military organizations. Thus, the survey results were used as a guide in planing our on-site interviews, but not for making inferences about all organizations.

The questionnaires solicited names and addresses of respondents who would be willing to be interviewed in order to help us gain a better understanding of organizational use of IT for organizational learning. Very few respondents provided the information. Follow-up revealed that among those who did provide contact information, not all proved willing or able to accommodate us in on-site visits. Face-to-face interviews constitute a very rich information source, but they are time consuming and expensive. We were able to visit with representatives of three organizations which use IT to enhance organizational memory: a large software producer, a public utility, and a non-profit organization. The author and an assistant participated in the interviews, asking questions and taking careful notes. The interviews consisted of open-ended questions, and lasted between thirty and ninety minutes. After each interview, we individually transcribed our notes as soon as possible. We then compared and combined our records, discussing both our written notes and our mental impressions of the interview.

Our interviews were based on a predefined set of open-ended questions allowing the respondents to describe their own views of organizational learning memory and how their organizations implement it. None of the respondents we interviewed indicated that their organization has implemented IT for the sole purpose of organizational learning. However, they described several existing systems which support day-to-day business operation, and which also contain components that support certain aspects of organizational learning. For example, the non-profit organization maintains and makes available internally a preferred vendor list. This list is maintained electronically and is incorporated into their purchasing system. Employees' experience with certain vendors results in additions, deletions, or updates to the list. When an employee is required to order supplies or materials, the preferred vendor list informs the vendor choice decision. Table 1 summarizes the results of the three interviews, which are described in the following paragraphs.

Table 1  

Interview Summary

The respondent from the software maker was a project leader. He was very enthusiastic about the use and potential of OM. He described how his organization uses a home-grown product-tracking system that stayed active for a given product from inception, through development, through sales and support, until the end of the product's life cycle. The system keeps records, for example, of errors in the product, and the measures taken to resolve those errors. When later product enhancements cause the same or similar errors to (re)surface, the tracking system can provide quick, easy, and valuable information about the history of the problem and its resolution. The system is available to everyone involved, whether they are software developers, product managers, or customer service representatives.

An additional application of OM formerly used by the software vendor (before they were bought out by a larger company) involved an application that might be described as an extended telephone directory. This database application was available to all employees and included capabilities for lookup by name, department, manager, function, project, and more. The respondent described examples of sales opportunities that were easily exploited because any employee could quickly refer a potential customer to the right person. By using the system, employees could quickly and easily find another employee with a specific set of skills or

experience, regardless of personal acquaintance or lack thereof.

When asked about how his organization measured the benefits of these systems or justified their continued use, the respondent answered in positive but vague assertions. The systems made employees' jobs easier; employees used the systems regularly; he couldn't imagine how difficult his job would be without these systems. Yet, he could think of no specific measurements the organization had taken to prove the value of these systems. The systems' value was apparently tacitly assumed.

Two respondents participated in our interview at the public utility: a department manager and a line supervisor (both in payroll accounting.) They reported no systems in place exclusively for purposes of organizational memory. Rather, OM is built in to their (newer) systems (which are predominately mainframe, batch-oriented, and developed in-house.) For example, every field on their new customer service system screens has a help feature that describes that field's purpose. Rather than ask a colleague, or laboriously look up an entry in a manual, the user simply enters a question mark, and the explanation is presented online.

Another implementation of OM is in the checklists, implemented on electronic spreadsheets, that are used by department personnel. These contain basic accounting checks and balances, plus evolutionary changes in policy that have arisen from problems over the last 25 years, including both automated and manual procedures. Use of the spreadsheets ensures that policies and procedures are consistently followed, regardless of the individualwho may be executing the instructions.

The respondents perceive the actual programming code, which has evolved over years of use and revision, to be a form of OM. That is, the program code contains rules and policies, as well as accumulated wisdom from the organization's experience. Still, according to the respondents, most OM is in people's minds and in policies. Some OM exists as a combination of people and IT (e.g. the spreadsheets.) Hence, training is an important component of OM.

The benefits of the public utility organization's memory-enhancing information systems were generally not measured. Immediate benefits, the respondents said, include savings of the distribution costs of policy documents. Longer term benefits include easier access to policy documentation. They also believed the automated systems' memory components help them avoid repeating mistakes.

Continued use of their systems is justified by continually questioning whether the systems add value (do they make the job easier? do they help? or do they cause more trouble than they are worth?) This is apparently a judgment by consensus, not by any empirical measures. The only hard measurements reported were cost comparisons for adding the online help feature to some of the older systems versus the new systems. The help component turned out to be a popular feature in the new systems, so they investigated the possibility of adding a similar feature to their older systems. Their analysis concluded that the cost of retro-fitting the old systems was unjustifiably high.

The respondent for the non-profit organization was a managing director (vice president) of materials management (he told us to consider him the "supply sergeant" of the organization.) He reported that their automated OM, like that at the public utility, was principally built-in to several functional systems. These systems support functions such as materials acquisition, printing, manufacturing (clothing,) distribution, and fleet management. They have no systems that were built specifically for OM.

Their printing system, for example, tracks costs, compares actuals to estimates, and performs analyses of the variance. The systems is also used to create job estimates. This enables past experience to inform present production decisions, regardless of which specific employee might be planning the print job. To support purchasing, they have a system that tracks vendor performance (timeliness, quality of product, etc.) This system allows them to maintain a "preferred vendor" list. Vendors earn and keep a place on the list by their performance. The list is available to the entire organization, thus giving every department the benefit of other departments' experience with this vendor or that vendor.

When we asked the respondent about the benefits of OM in their automated systems, he responded with three goals, which are pertinent to most organizations, whether commercial or non-profit: cost effectiveness, high product (or service) quality, and timeliness. The OM components of their systems allow for continual cost reduction and fewer mistakes (returns, customer complaints.)

The respondent said justification of many costs in a non-profit organization is a value judgment, however they do some justification of their systems in terms of cost reduction and ROI. The OM components in their systems contribute toward doing more work with the same people, and toward meeting the three goals mentioned above. Measurement of their value is, again, mostly a judgment call, but they use some economic analysis based on salaries and other costs.

The respondent noted that some employees may resist the idea of automating certain knowledge (their own expertise) for fear of job security. He also noted that all estimating software should be considered organizational memory because it uses the organization's past experience to make predictions for current use. A different example of OM is random warehousing. This system keeps track of bin locations that are assigned randomly for more efficient warehouse space allocation. The "memory" of where everything is belongs to the organization, and contributes to the goals of cost effectiveness, timeliness and quality. This organizational memory is different from other types of OM because it is not simply a repository for knowledge that exists in a collection of people's heads.

The use of IT to facilitate organizational learning seems to be at least somewhat established in organizations, but organizational memory systems are the apparent exception. The results from the questionnaire suggest that the use of IT to facilitate information distribution is the most established, with the use of IT to facilitate information distribution and knowledge acquisition closely following. The questionnaire results suggest that use of IT to facilitate organizational memory is the least-exploited source of organizational learning. This implies the possibility that IT-facilitated organizational memory is a potential source of competitive advantage for many organizations. Since organizational memory is apparently not often exploited, there may be an opportunity for those who are willing to take advantage.

The results of the questionnaire suggest that investments in IT to facilitate information distribution, the most mature technology of the four areas, is adequate at about 70%. However, this leaves plenty of room for improvement. Investments in IT to facilitate knowledge acquisition at about 67% is also adequate but interestingly the technology is relatively new. Investments in IT to facilitate information interpretation comprised of even newer technologies received even fewer positive effect responses, at about 64%. Investments in IT to facilitate organizational memory with the newest technologies were the least likely to be perceived positively with less than half (49%) of the responses reporting a positive effect or return on investment. Perhaps technologies which are newer, less diffused in the organization, and requiring large training investments are perceived as having a less positive effect than those that are mature, highly diffused, and requiring little training investments.

The respondents we interviewed gave us the impression that OM, at least for some people, is perceived as a resource to be exploited at the operational level — not necessarily at the tactical or strategic level. We expected to find examples of OM systems at the higher (tactical and strategic) decision-making levels. Instead, our respondents described systems that focused on day-to-day operations. Justification for systems with OM at this level is perhaps easier, or at least more concrete, than for higher levels. Also, requirements definition is easier for systems at the operational level than for higher levels.

Organizational memory was perceived, in two organizations, as a by-product of automated systems that were created to support business functions. In these organizations, there was apparently no effort or desire to create an OM system specifically for the purpose of memory. In some cases, OM may be merely a serendipitous side effect of automation. In this sense, the logic can be extended to assert that everything/anything that is automated is OM. If this is so, organizational memory may be nothing special. It may simply be a phrase that constitutes another way to describe all automated systems. However, careful consideration convinces otherwise. The respondents in this study interact with many automated programs and systems, but they described in the interviews only those systems that had obvious OM components. The respondents perceived clear distinctions between systems that had had OM features and those that did not. The systems they described (that is, the OM components of those systems) seemed to function mainly as mechanisms of policy enforcement.

Finally, we note that most of the systems described to us were in place to support specific business functions — not only OM. Justification for the OM features in these systems was stated in terms of supporting the business function involved — not in vague references to supporting decision-making with information from the past. In previous consideration of OM and systems to support OM, we had considered only systems built specifically and exclusively for corporate memory purposes. The results of this study suggest that while OM is still probably the least likely part of organizational learning to be implemented with IT, it may be embedded in a broad spectrum of systems in more places than we had previously realized.

This study was not intended to be generalizable to a particular population of businesses or organizations. Rather, the intent was to identify how an investment in information technology to support organizational memory might be justified. This is more a theoretical effort than an empirical one. The results suggest that the model of OM as a strategic resource might be premature. A more realistic model might portray OM as an operational resource, perhaps evolving to support higher levels of decision making as systems change over time. However, this study included very little of the emerging Internet-based business systems which support the new paradigm of completely electronic businesses. For now, we can only theorize that the value of information technology from an OM perspective, lies in supporting operations.

In all three interviews, there was very clearly a preference for operational justification. That is, the value of the organizational memory components in existing systems was described in terms of how they support operations. Organizational memory is a value-added feature of some information systems. This implies two things: first, researchers and developers seeking to convince management of the potential value of an OM system might do well to focus on features that enhance daily operations of functional systems. They should apply this focus both in their development efforts and in "selling" the idea to management. Second, managers should recognize the OM benefits they already enjoy by virtue of existing automated systems, and demand greater OM benefits in new systems. If OM is embedded in current systems that were developed without specific effort to include OM features, perhaps a little effort in new development can greatly enhance the value added by new systems.

Emerging technologies (e.g., data mining) are helping managers realize that a wealth of knowledge exists in their current databases. Perhaps we will also discover that the information systems themselves (programs) contain knowledge that is more valuable to organizations than previously realized. Business seems to moving toward more dependence on information systems (Intranets, extranets, interactive web sites). That OM will exist in these systems seems a natural consequence. Whether mangers will take advantage of the potential value remains to be seen.

Conclusion

This study consisted of two parts. First our survey identified the extent to which IT is used to support organizational learning in various organizations. The survey highlighted organizational memory as the component of organizational learning that is least supported by IT in organizations. Information systems that facilitate organizational memory are also least likely to be perceived as having a positive effect or return on investment. The second part of this study identified some benefits or gains associated with organizational memory systems as perceived by respondents in those organizations which reported positive effects.

The OM systems described were actually business information systems with OM components, and the benefits of OM were most often described in terms associated with specific business functions. The systems described supported operational-level decisions and tasks. The systems were apparently perceived mostly as policy-enforcing systems, their OM components serving to maintain conformity and reduce errors. While there are examples of information systems, designed and implemented for the sole purpose of supporting OM [2; 7] the respondents reported no use of such systems.

The results included nothing about OM as a tactical and strategic resource. Possibly, the potential for information systems technology to facilitate OM in serving these higher decision-making levels exists, but the benefits may be more difficult to quantify. Subsequent study will be required to identify and measure those benefits.

[1] Ackerman, M. S. & Malone, T. W. 1990. Answer Garden: A tool for growing organizational memory. Proceedings of ACM Conference on Office Information Systems, 1990: 31-39.

[2] Conklin, E.J. (1976) Designing Organizational Memory: Preserving Intellectual Assets in a Knowledge Economy. http://www.cmsi.com/business/info/pubs/desom/body.htm.

[3] El Sawy, O. A., Gomes, G. M., & Gonzales, M. V. 1986. Preserving institutional memory: The management of history as an organizational resource. Best Papers Poceedings, 46th Annual Meeting of the Academy of Management, Chicago.

[4] Huber, G. P. 1990. A theory of the effects of advanced information technologies on organizational design, intelligence, and decision making. Academy of Management Review, 15.

[5] Huber, G.P. (1991), Organizational learning: the contributing processes and the literatures. Organization Science, 2(1) 88-115.

[6] Huynh, M., Popkin, L. and Stecker, M. (1995) Constructing a corporate memory infrastructure from Internet Discovery Technologies. http://www.ncsa.uiuc.edu/SDG/IT94/Proceedings/CorInfSys/huynh/cmi.html

[7] Laabs, J.J. (1993) Electronic Campus Captures Apple's Corporate Memory. Personnel Journal v.72, pp 104-105+.

[8] Levitt, B. and March, J.G. 1988. Organizational learning. Annual Review of Sociology, 14, 319- 40.

[9] Morrison, J. 1992. Development and evaluation of a system to support team and organizational memory. Doctoral Dissertation, University of Arizona.

[10] Summary of the Financial Impact of Lotus Notes on Business (1995) White paper, http://www.lotus.com/notesdoc/220e.htm

[11] Walsh, J. P., & Ungson, G. R. 1991. Organizational memory. Academy of Management Review, v16, pp:57-91.