I really enjoyed this article but the argument that personas and user stories are only based on assumptions as opposite to Job to be Done is bogus. All 3 methods must be based on research to be successful. (https://medium.com/radical-ux/why-your-personas-dont-work-for-you-75c1abfcec94#.bawux995e)

In the example, Job to be Done statements don’t appear until after interviews and analysis of their results.

If Clay Christensen had not observed customers he could have come with the following Job to be Done statement: When I finish lunch, I want to eat something sweet so I can satisfy my need of indulgence. There, a Job to be Done statement based solely on (wrong) assumptions.

I’m left convinced that the Job to be Done framework offers a very structured way of thinking about the motivations to use a product but only if it is based on previous research.