As you may or may not know, for Visual Studio "Orcas" planning we’re using a new process based on the concepts of scenarios, value propositions, experiences, and tasks. This system relies on the idea of a scenario – think of them as a very long, end-to-end keynote demo that we’d show to someone to show them how Visual Studio helps them do their job. A scenario has a set of value propositions – think of those as the moments in the demo when the audience claps spontaneously (we’d hope) – and a value proposition is made up of a set of experiences (collections of features that enable a part of the value proposition). Eventually we get down to the point where we’re actually talking about features.
The cool thing about planning this way is that we’re able to see how each feature rolls up to a specific scenario. We also know that if we cut certain features, the value proposition it lives in starts to fall apart and we can have a better discussion about whether, if we have to cut some of these features, the entire value proposition and all its features shouldn’t just be cut.
It’s a way to avoid making peanut butter, as Brad Abrams calls it.