Rogue Librarians

Early in the advent of web-based Integrated Library Systems, (ILS), KPMG invested in a system that the a library director subsequently found to be underperforming and no longer suitable for the needs of the corporate library structure. The system was being used to manage print publications across close to three hundred offices but was not adept at managing access to electronic publications. Annual fees for the library system were expensive and additional training for new users on the system was expensive as well. It was also felt that issues that arose with the system were not addressed in a timely enough manner by this ILS’s customer support services. During this period of time the firm was undergoing substantial weeding of the print libraries, reducing them in size to less than one tenth of their original square footage. Additionally, it should be noted that the electronic publications were also being managed by Tax Knowledge Management and Audit Knowledge Management teams and the efforts to manage electronic information were redundant.

In the summer of 2003 the manager appointed three members from a team of approximately twelve Library Coordinators to investigate alternatives to this system. Although the basic reason for the changeover between the two systems was that the newer system would enable access to electronic publications, the basic reason for this system was the control of print materials. The team investigated the features of several alternative web based systems suitable for corporate libraries amongst themselves and suggested a new automated system in a relatively short period of time. It was announced in the late summer of 2003 that library functions would be migrating to this new ILS and an introductory web based training program was held in October of that year. During this migration and development period the firm would continue using the older system while the sub-team was to work on the migration and the development of the new ILS and the new system was purchased. The sub-team attended multiple training sessions over the next year, worked on data migration from the old system to the new system, and also spent numerous hours customizing the software interface.

The new ILS required that we install user interfaces on the hard drives of the local computers in order to access the web-based system. Unknown to the rest of the Library Coordinators, the sub-team that had made the decision to purchase the new system had not consulted with Information Technology, a process step that was supposed to be included with any software purchase. It was revealed in 2004 that they had chosen an ILS with a user interface that was JAVA-based and incompatible with the Time and Expense reporting system that every U.S. employee was currently using. During the period of time between late 2003 and late 2007 the team expended numerous resources trying to adapt this system for use, despite the fact that it was incompatible with the existing U.S. time and expense software. Solutions included housing the system on a server in Canada, where the JAVA based software wasn’t an issue, and redeveloping the user interface so it wouldn’t interfere with time and expense reporting. The firm continued paying many tens of thousands of dollars annually for each of these two systems, in addition to the redevelopment costs and the additional server charges. In 2007 the firm announced that with the exception of local business news, it would no longer be supplying print serials subscriptions. Despite this, the library team continued funding both systems and continued working on the migration and the development of the newer system. Sometime in 2008 the team reached a point where they were able to abandon the older system and concentrate only on the continued development of the newer ILS. By this point in time the majority of print subscriptions were cancelled and new firm requirements for most offices allocated a maximum of 60 linear feet for print library collections as resources were substantially electronic, (the rest of the monographs from these offices had been discarded).

Over a five year period of time the firm paid for duplicative ILS software programs, one of which was never able to perform adequately enough for a full-scale software roll out.
What occurred in this situation was in part a manifestation of regret avoidance. The original decision to allow these team members to pursue this solution in a relatively unrestricted manner, without clearly understanding the procedures involved in making such a purchase, and furthermore the director signing off on the purchase of this software, was regrettable. Pursuing the inferior solution was clearly a matter of regret avoidance. Bazerman states, “the motivation to minimize the opportunity for regret can lead people to make decisions that are suboptimal with respect to actual outcomes”, (p. 98). The director had an emotional reaction to the drastically inferior outcome, in order to avoid regret for making such an egregious and costly error, this inferior outcome had to be pursued.

Although we can witness a fair amount of “Perceptual Biases” in the aftermath of this decision, where the director ignored information that contradicted the original decision, and also “Judgmental Biases” that necessitated that the director hide the inferior ILS selection, this was clearly and more importantly an example of non-rational escalation of commitment termed “Impression Management”, (Bazerman, pp. 109 – 111). The library director had allowed the team members to make a business decision that had extremely adverse economic effects. The error was perpetuated for five years while the team tried to correct the inferior decision, and exacerbated by continuing to pay for both ILS systems for this same period of time. The rewards for the inferior decision seemed greater than they were, especially in light of the firm’s clear direction for reducing the dependence on print libraries. In actuality, the best decision for the company would have been to “focus on future costs and benefits, ignoring any previous commitments”, (Bazerman, p. 111), by abandoning the decision to change ILS early on. A more rational decision would have been to abandon the newer ILS when it became apparent that it was unsuitable for the firm’s use due to the software conflicts, cancel this contract, and return to the previous ILS.

References

Bazerman, M. H., & Moore, D. A. (2009). Judgment in Managerial Decision Making (7th ed.). Hoboken, NJ: Wiley and Sons.