Why so many simple KM loop diagrams are wrong

Here is another post from the archives, this one on the problem with so many simple KM loop cycles.

I have lost count of the number of Knowledge Management lectures, strategies, introductions and other material I have seen, which somewhere along the line have presented a single loop model for Knowledge Management.

Almost all of these have been wrong.

The ones that were wrong, were the ones that look like the upper picture - that start with "Create" (for creating knowledge), then immediately move to "Capture". Try a google image search for "Knowledge Management cycle" and you will see what I mean.

The reason why they are wrong, is that Knowledge does not always need to be captured to be managed. 

You can, in some circumstances,  operate a perfectly good Knowledge Management system while keeping knowledge tacit and uncaptured, and transferring the knowledge through planned and structured conversations between the right people (for example using Peer Assists, Knowledge Handovers, Knowledge Exchanges, Mentoring and Coaching). Therefore a single loop with a "Capture" step is only a partial truth.

In more general terms, Knowledge management can take two routes, generally known as Collect and Connect. In the Collect route, knowledge is documented and becomes content - it is "Captured" (leaving aside for the moment all the arguments around whether knowledge can ever really be captured). In the Connect route, knowledge need never be "captured" so long as it is transferred through conversation.

Therefore all these knowledge cycles, that require "Capture" somewhere along the chain, only refer to the Collect dimension of knowledge management, and miss the Connect dimension.

A single loop is a single line - a one-dimensional thing. Knowledge Management is two dimensional (if not more).  It is both Connect and Collect - it needs a Collect loop, and a Connect loop.

Therefore a more realistic model is the lower one shown here, which reflects the reality of the two dimensions, and expresses them in two cycles.