I was reading recently about Bloom's taxonomy for Learning Objectives. This is a model that goes back to 1956 but I personally believe that it applies even to our present times.
When designing a training program or course, we should not simply do a "brain dump" of what we want to say but rather make sure that the content and the delivery is structured in a way that makes sure that the audience will get the actual value out of it. It would be interesting to check if our current IP creation and delivery methods have this level of insight built-in. Do we really make sure that the people "get the point"? Do we over-rely on "assessment-type" proof of the success of the training? Some interesting questions I plan to investigate. Here is a link to the wikipedia article that covers the topic. A bit too "academic" for my taste but surely, worth the 10-15 minutes it will take you to read through it.
Manager – Microsoft Premier Field Engineering