The theories behind hypotheses

This is the first chapter of my upcoming book on Product management. If you want to advance order a copy, contact me on LinkedIn.

Product managers often use the word hypothesis, which is fabulous because ideas need validation. But beyond a single testable idea are theories that explain underlying patterns in your customers and market. If you believe in the need to test hypotheses, then you should be eager to refine or even reject the more encompassing theory underlying it. As new information emerges, you may find an alternate theory that better describes, explains, and predicts.

If hypotheses are a cornerstone of product management and are in service of theories, it pays to learn what people who’ve thought deeply about theories have to say. Popper’s falsifiability criterion says you can’t create a theory that cannot be disproven. A theory must be logical and cannot be drawn from nothing. A theory may start broad, generating various hypotheses that, by deductive reasoning, can be used to test and maybe narrow it down. It also can be expanded based on new information. Finally, a theory should talk about what is and not what isn’t. In that spirit, you should formulate hypotheses to confirm or disconfirm the theory.

When evaluating two prospective theories to explain the same things, favor the one with fewer assumptions (Occam’s razor). Evaluate the theory under extreme conditions to find its limit and even try to expose its hidden flaws (reductio ad absurdum). You also can strengthen a theory via indirect proof, where it’s shown that opposite results can’t be true.

Be aware that even repeatedly proven theories may rest on shifting sand. Thomas Kuhn’s The Structure of Scientific Revolutions makes a strong case for paradigm shifts in which there is never a rational, linearly progressing path of understanding. And so your supposedly rational experiments reflect an ever-evolving understanding of your product, users, and market.