We've Moved!

Think Clearly has a new home! Click here to see our latest posts.

*If there’s older content you’d like to catch up on, you can browse right where you are, until Friday, April 5th.
We know it takes a minute to get used to change!

KM Programs: Implement, Improve, and Iterate

Posted by Stan Garfield on 3/29/2018
Stan Garfield
Find me on:


Knowledge management programs can use a wide variety of people, process, and technology components. It’s important for KM program leaders to gain direct experience with as many of these components as possible, to evaluate their possible application, and to lead the way in implementing new ones to fill current and future needs.

Classic software development projects included lengthy time allocations for analysis, design, and development before users ever had a chance to try out the results. Given that it is difficult to know exactly what features users want and how they should actually work, the “finished product” would often be unsatisfactory to the users for whom it was developed, despite the fact that it met their specifications.

Give it a try

A more useful approach involves rapidly developing a prototype, letting users try it out, capturing their feedback, and then improving it accordingly. My colleague at both Washington University and St. Louis University, Dave Bridger, used to say that a software program should be measured not by how closely it met its design specifications, but rather by how useful it was during its development. Programmers who rapidly develop, try and refine save time, continuously involve users, and produce better results in fewer cycles.

Ban the Big Bang

Knowledge Management programs and intranet systems often make the same mistakes as software development projects. Lengthy designs or redesigns are followed by big launches and then by users disliking or ignoring the touted offerings. I call this the “big bang” approach, such as when a new or revised website is unveiled after six months of development, only to miss the mark as judged by its intended audience. What are the users supposed to do during the time prior to launch? It’s much better to quickly launch a simple site serving up the most important content (as defined by the users) and then continue to improve and add more content on an ongoing basis. This results in a site that is both immediately useful and perceived as being continuously improved.

As soon as you have a potentially good idea for a people, process, or technology innovation, try it out. Start by discussing it with a group of trusted colleagues, fellow members of a community of practice, or insightful friends and family. Mock up a simple picture, screenshot, or process flow. Encourage candid comments and suggestions—and incorporate as much of this feedback as possible in your initial design.

Solicit Feedback

Implement your idea directly, through a colleague, or through a team good at development. Do this sooner, rather than later. Publicize your initial implementation through a relevant community of practice, your social network, and your work team. Solicit feedback for improving functionality, usability, and effectiveness. Then quickly make improvements and repeat the cycle. Continue this process indefinitely, with longer cycle times as functionality better aligns with user requirements.

New KM tools are being developed and made available with increasing frequency. For the ones with great personal appeal, relevance to your organization’s needs, or the potential to be widely adopted, it’s a good idea to get out in front of future demand and try them out. This embodies learning by doing, leading by example, and modeling desired behavior.


Topics: Knowledge Management, KM, Strategy