Well, there is no one answer. It’s based on the role. A competency describes what’s most critical to success in a role, not everything one must do.
Once you’ve built your model, you may discover there are simply too many tasks for a reasonable assessment and resulting development focus. E.g. if you’ve identified more than 40 tasks, the assessment will simply take too long, and you’ll lose the intrinsic motivation you’re trying to create. While there is no hard rule, we recommend 20 – 25.
Think about it logically. If it takes about 1 minute to review behavioral examples and select a value (in a meaningful and thoughtful way), and you have 50 competencies, that’s 50 minutes. Can you say “survey fatigue”? On the other hand, if you select the 20-25 most critical, it should take ~20 minutes… a much more realistic request. Plus, who can focus on that many skills and potential gaps?!
Pick the most critical to success items now, and continue to iterate it to capture changes in strategy, in tools, in technology, and in the environment in which you operate, so you can always focus on the critical tasks and skills for that point in time. It may be that your assessment includes 20 now, and next year, you remove 8 and add 10 new ones.