How sophisticated should my competency model be?

(Question posted by participant in Accelerating Learning Transfer webinar)

A critical success factor in competency model/capability model development is the ability to have the model written in the language of those who will be using it.  It should NOT be in HR/Learning terms. 

In an earlier post, I pointed to 3 steps to simplify a competency model.  I think that’s the same answer to how sophisticated the competency model should be.

1)      Take a step back and look at the model from a holistic approach.  Is it easy to consume?  By that I mean, do you have to read it, then re-read it to know what it means?  Or is it in easy to read language – the language of the people using it so they can know what great looks like?  Often we see a good capability model written in a way that is very academic.  The person who is writing it looks very smart, but the capability model itself cannot easily be consumed by the end user.  It should read as they speak.  It should state clearly what someone should be able to do. 

2)      Look at the model from a volume check.  We used to see models between 25 and 40 skills.  Now they trend toward 15-25.  (You should keep in mind a manager’s participation when considering volume – something one of our customer managers just reminded me about.)  A capability model should describe what “great” looks like – not be a complete task analysis of everything they do. 

3)      Look at the model as a professional in that role.  Are there things in it that are extraneous?  For example, do they include tasks that people may do, but are not critical to success in that role or may not be done by some at all?  And if so, would those really represent behaviors that high performers would do related to some other task based on something else they do?