Skip to main content

Ask them to predict the results first

Stakeholders who say they already knew a finding often didn't. Asking them to write down a prediction first shows them, and makes the research land.

3rd December 2023

Research that should change a business often lands with a shrug. The data is new, the finding is real, and the room says it knew that already.

We saw this early in our careers, again and again. Results we thought were genuinely new left stakeholders underwhelmed, because they felt they understood their audience inside out. We suspected that understanding was thinner than it felt.

So we tried something simple. Before sharing results, we asked stakeholders to predict them: privately, in writing, before the deck opened. The shrug went away. Once people had committed to a guess, they could see where they'd been wrong, and the findings read as news.

There's a name for what was going on. Ethan Mollick, the academic, described it in a post: "Non-experts often think they know best, but if you first ask them to privately write down how something works… they realize they don’t really get it & become more willing to listen to experts." Psychologists call it the illusion of explanatory depth. We think we understand things in far more detail than we do, until we try to explain them.

Audiences invite the illusion. Everyone in a company has met customers, read the reports and watched the numbers. That's enough to feel expert. It's rarely enough to describe the audience accurately or consistently. When we interview stakeholders one at a time, we often find each uses different words for the same audience, and carries a different picture of who's in it. Nobody notices, because nobody has had to write it down.

So the prediction step does two jobs. It makes the research worth more to the people paying for it, because they can see what it taught them. And it shows where a leadership team disagrees about its own audience, which can matter more than any single finding.

It takes five minutes. Those five minutes can decide whether months of research get used.

What to do on Monday

  • Before your next debrief, send stakeholders three or four of the key questions and ask for their predicted answers, privately and in writing.
  • Open the debrief by showing the predictions beside the results. Don't say who guessed what.
  • Ask each member of the leadership team to describe your core audience in three sentences. Compare the answers before you commission anything new.

Related service: Steer

← All points of view