This approach has become the core approach used by many start-ups to bootstrap and get their product off the round. Also across the UX and Design industry there is a growing acceptance - like the adoption of Agile project methodologies - that this is the way to do things.
For the last few years, where I've been involved in innovation and new product creation for clients, this has almost become the defacto position:
- Take your new product premise and agree a single hypothesis that the product team agree will test the premise. Keep it simple. Ruthlessly control the variables.
- Determine what is the minimum thing you need to build to test your premise
- Build a test version of product. One that your happy to throw away
- Remember, what your building is for the test - not the final future vision of the product
- Test it with real users i.e. remote online testing, in a lab or even live user testing
- Review your test results - these should be quantitative in nature. If not, your next test must be quantitative
- If you succeeded, great. Carry on to the next test
- If you failed, pivot and rethink the premise.
- If your not sure, you designed the test wrong. Redesign your test, try again or quit
Test hypothesis structure to use in workshops
Case studies to review
- Johnson & Johnson
- Making Product Hunt
- The Lean Startup Approach principles
- The IKEA effect inherent in MVPs - a word of caution summed up well