Time saved—a misleading justification for KM
By Michael E.D. Koenig
A word of warning about "saving the user's time" as a justification for knowledge management. At the KMWorld 2001 Conference and Exposition in Santa Clara in the fall, greatly increased emphasis was placed on structuring and retrieving data (see "The Third Stage of Knowledge Management Emerges," KMWorld, March 2002). There was also, of course, continued emphasis on how to garner support for KM initiatives and how to cost-justify KM expenditures. The principal component of that emphasis was justification via story-telling—the good anecdote or sea story that illustrates an effective example of KM in practice—an effective technique that is getting ever-increasing recognition and that is undoubtedly effective. There was another component, however, about which one has to be a bit more cautious. The advice given by several speakers—typically in the context of how a taxonomically well-structured system can improve system performance—was that improved KM systems can save search time for the user, and that this savings can be a quantifiable justification for KM.
Yes, a well-structured KM system can save user time, but there is an unrecognized danger in using that as a justification for investing in or improving a KM system. It makes sense to save the user's time, but the justification of the KM system ultimately has to be demonstrated by better decisions and improved performance.
The problem with time saving as a justification is that even if the user's time is saved by a KM system, management probably won't see any difference in the user's behavior, and the justification could backfire. Why won't management see a difference?The answer is a surprising one. Over the years, a number of studies of the practices of white-collar professionals have been conducted and their findings are uniform and corroborative. White-collar professional employees spend a consistent 20% to 25% of their time information seeking. A recent study arrived at the same conclusion (Normier, Bernard, report of a 2001 study conducted by Lazard Freres & Co., presentation on "Natural Language Advances," KMWorld Conference, 2001). Workplace technology may have changed, but that 20% to 25% hasn't. The proportion is surprisingly independent of the apparent information intensity of the job domain. Line business managers and administrators spend as much of their time information seeking as do research scientists. What is going on here?
There seems to be a sort of homeostasis, or perhaps more accurately a satisficing mechanism at work. Knowledge workers--whether managers or administrators or researchers--need substantial information input to perform satisfactorily, but when the amount of time devoted to that function approaches 20%, knowledge workers appear to begin to satisfice. They begin to conclude that they have to get on with the rest of their job; that if they have not already done so, they will soon run into diminishing returns in their information seeking; and that it is time to proceed based on the information they have.
The consequence of this phenomenon is that if a KM system allows a user to save time, that time will most likely be diverted into other information seeking behavior and users will still spend the same 20% to 25% of their time information seeking. That is, of course, not bad; both the information need served by the KM system and the information need into which time was diverted may well be better served, with better decisions made and higher productivity resulting. The point, though, is that the bottom line for KM is better decision making and higher productivity. Time saved is a misleading and unconvincing indicator, because you probably won't be able to demonstrate it.
If we make justifications based on time saved to management, they will quite predictably and understandably expect to see that time saved now deployed in "something else" like making sales calls or serving customers. When they see that saved time simply directed to other information search time (if they notice any difference at all, and they probably won't), the behavior will look the same to them, and they will quite understandably feel that they have been misled.The bottom line--don't justify KM by time saved, justify it by sea stories, and where possible by demonstrating better decisions and higher productivity.
Mike Koenig is dean of the College of Information and Computer Science at Long Island University, e-mail michael.koenig@liu.edu.