Over on the Adaptive Path blog, Brandon Schauer shows off a couple of nice diagrams, illustrating how new products typically get approached, how they they’re perceived, and how they should be approached.
How not to do it.
What the user sees.
That looks about right.
I’m a little bit divorced from the original context of the diagrams (”how businesses approach delivering value to their customers”?), but I was only thinking about something like this yesterday. What comes first, the interface or the data model? There is no question, that the diagram of user’s perception of a product is spot-on, they could care less how and what you’re storing on the back-end. However, with a complex-enough problem, it’s hard to make informed inroads without exploring the interface and data asynchronously.
The user research and design process will always have the first stroke, but to understand the data and solve problems around the data is to acquire domain-specific knowledge of the problem space. For instance: to meaningfully organize a product spec page for a laptop, you need to understand the data that surrounds a laptop. And if you’re not using a more realized data model, you’re effectively using “lorem ipsum” and a lot can be said about the deficiencies of that strategy.