Questioning · Lesson 2 of 2

Questions that surface assumptions

"What would I expect to see if I were wrong?" is the highest-yield question in this whole section.

6 min read · Free — no account needed to read

Some questions do unusual amounts of work. What would have to be true for this to be a good idea? What would I expect to see if I were wrong? What is the question behind this question?

Applied to yourself, they are the fastest route out of a stuck position — because a stuck position is usually one where an assumption is doing the holding, and assumptions do not survive being written on a line by themselves.

Applied to other people, the same questions are what turn an argument into an enquiry. The difference in tone is entirely in whether you would accept an answer you did not expect.

See it happen

You are convinced a feature will land well. The team is building it.

The read that arrives on its own

Keep going — everyone I have spoken to likes it.

The same situation, with this idea

"What would I expect to see if I were wrong?" Probably: people opening it once and not returning, and support questions about what it is for. Both are already in the data from the beta, and neither had been looked at.

What changed

The question names the evidence in advance, so it cannot be quietly reinterpreted afterwards. Everyone you spoke to liking it was never a test — nothing about it could have come out the other way.

The takeaway

"What would I expect to see if I were wrong?" is the highest-yield question in this whole section.

Easy to take the wrong way

Asking it and accepting an unfalsifiable answer. "I would see it not working" is not an answer; "fewer than a fifth of them come back in week two" is.

Try this

Take a plan you are confident about. Write what you would observe if it were wrong.