I saw Steve Hopkins post today on hypotheses based testing and thought I’d share some ideas we’ve been using at Pollenizer.
A lot of startups we see and work with are getting better at doing customer development, but they sometimes aren’t quite sure what to do with the results. Very rarely do you finish one lap of the testing with an absolute clear answer and a simple view to what needs to be done next. Normally it’s murky, confusing and conflicting.
Important to do before hand is to set some metrics of your expectations. Afterwards you’ll be tempted to justify results in the direction that suits you, which is normally what you have as your best case plan.
Here is how we set them and what we do next;
- x = fantastic results. Do the test again and see if you get the same result.
- y = good, but not great results. Make a minor variation to existing elements to the same market to see if you can improve the result.
- z = bad results. Make a major variation to the elements, priority of elements, target market etc to see if you can get ‘good’ results.
What is minor and major? That’s up to your brilliant instincts as a future billionaire founder. 🙂