Rob Fitzpatrick works through a detailed example of idea validation, using an idea which he's currently working on as the example.
You should go and read the whole thing, but here are a couple of pull quotes:
This simple cycle is the core of all startup strategy:
- What are the biggest assumptions and uncertainties baked into our idea?
- What's the fastest, cheapest & easiest way to answer those questions?
- Do it.
- Measure it.
- What are the biggest assumptions and uncertainties still baked into our idea? And so on.
And:
Note that I didn't write a single line of code until I was up to assumptions 4 & 5, by which point I was confident in my problem, value proposition, and customer segment. The total time spent coding to date is still well under 20 hours — a single long weekend or a week of late nights after work.
More on this approach here.