Assumption mapping
A Decision Jam that separates what the team knows to be true from what it only believes, then turns the gap into How Might We questions.

- Time
- 60 min
- People
- 4–8 people
Overview
A good starting point for the working group to collectively identify and prioritise the problem statements that anchor the rest of a short discovery run. Built on AJ&Smart’s Decision Jam, the team writes down what it is confident is true about the project, then what it would want to learn from customers, votes on which of those matter most, and reframes the winners as How Might We questions.
The mechanism is the separation. Once “what we know” and “what we want to learn” are on the board in different colours, the open-ended discussion the team was having turns into a research list. Note that this is not the classic two-by-two assumption matrix — the IFS board runs the Decision Jam sequence. Each round prepares the working group’s design thinking in turn, so the session arrives at four to five well-reasoned How Might We statements rather than a single rushed guess.
When to use it.
During framing, once a direction is forming but before deeper concept work starts. Also a good recovery move when a team is going in circles: usually they are disagreeing about an assumption neither has named.
Who to involve.
Four to eight, spanning design, product and engineering. Engineering presence matters here — feasibility beliefs are the ones most often mistaken for facts.
Effort.
Medium prep High engagement Six rounds back to back; the reframing round in particular tends to overrun once the team starts discussing wording at length.
Outcomes.
A set of How Might We questions the team has voted on and can justify, which become the input to Future press release, and a prioritised list of what to test with customers.
Recommendations.
Keep the two sticky colours strict — the moment a “know to be true” note is actually a belief, the exercise stops working, so challenge them. Run each round silently first, so the loudest voice does not set the list. And do not skip round 6: a well-voted question nobody can justify should not become research.
How to run it
Before the session, make sure the product manager or decision-maker has agreed a small set of Jobs to be done (JTBDs), so the session is loosely grounded to a target persona and/or key moments rather than starting from nothing. Share this as a pre-read ahead of time — the outcomes of the pre-framing workshop should inform it — and elect a decision-maker for the session itself before it starts.
Timings are the board’s own.
Create the container
2 minutes Introduce the activity, start with why we are doing this together.
What do we know to be true?
5 minutes Purple stickies. Anything anyone is confident is true about the project. Silent, individual, no discussion yet.
What might we want to learn from customers?
5 minutes Blue stickies. The two colours are the whole point — at a glance you can see how much of the “strategy” is actually established.
Prioritise collectively
3 minutes Everyone gets the same number of votes, placed as stamps, to vote on the most important issues to tackle. Then it lands with the decision-maker: they get three to five votes of their own to make an informed decision from the clustered votes, and whichever they choose moves across to top voted.
Reframe as How Might We
6 minutes Turn each top-voted assumption into a standardised challenge. “Who is the most valuable customer?” becomes “How might we measure the value this product provides to different potential users?”
Why?
10 minutes As an open discussion, ask the group: from a business and a customer perspective, why should we address this? Capture the answers next to each How Might We.


