What needs to be finished first
A list of relevant stakeholders with role and relationship to the problem is available so the right people sit at the table.
A rough written understanding of the problem space exists, at least market, user groups and known constraints.
What needs to be ready before start
Whiteboard or Miro board with four zones (symptoms, affected people, impact, problem statement); colored stickies; timer; pre-briefing as PDF; list of previously failed solution attempts.
One facilitator (neutral, without solution preference); one sponsor with decision authority (present for at least 60 min); 4-8 participants from functions that experience the problem differently; one scribe for verbatim notes.
Reason and trigger for the workshop; relevant data or tickets from the last 90 days; current team hypotheses; known taboos or political sensitivities with the sponsor.
90-120 min
Set up room so everyone can see the wall. Label four zones in advance. State workshop rule: separate symptoms from causes, no solutions before problem definition. Timer visible for every phase.
The one question this method answers
Which concrete problem, for which affected people, with which impact should be worked on next?
Marker: Phase
| Step | Duration | Action | Hint |
|---|---|---|---|
1Phase 1: Collect symptoms | 20 min | Each person silently notes observed symptoms on stickies, one statement per sticky. Then present round-robin and cluster in Zone 1. No discussion about causes. | If stickies already contain causes or solutions, move them to a parking zone. Otherwise the group jumps into solution mode too early. |
2Phase 2: Affected people and impact | 25 min | For each symptom cluster, name concrete affected roles (role, not person) and describe measurable impact (money, time, risk). Document in Zones 2 and 3. | If impact cannot be formulated in numbers or clear consequences, the symptom is probably not a problem but a matter of taste. |
3Phase 3: Draft problem statement | 25 min | Formulate two to three competing problem statements in the format 'who has which difficulty in which situation with which consequence'. Run heatmap voting per statement (3 points per person). | Avoid solution wording in the statement. A good statement does not solve yet; it only describes. If all statements look the same, perspective diversity is missing. |
4Phase 4: Sharpen and assign owner | 20 min | Take the top statement, define terms, document out of scope, capture success criterion for 'problem solved', name owner and next step. | If the sponsor cannot approve the statement, either the wrong person is in the room or the sponsor does not want commitment. Clarify, do not gloss over it. |
What comes out at the end
One-page document with problem statement, symptom list, affected roles with quantified impact, out-of-scope list, success criterion and named owner for the next step.
Date, sponsor and facilitator in the header. Later reframings as new version with clear rationale; old version remains linked and is not overwritten.
Worksheet for problem, target audience, impact, and decision context.
For whom is which problem relevant in which context?
Who experiences the problem, how often, and how do we notice it?
Which data, observations, or examples support the problem?
Which costs, risks, or missed opportunities arise?
What are we deliberately not solving?
What do we need to find out next?
Concrete filled scenario, fictional example
Concrete filled scenario, fictional example
Sponsor: @julia (CPO)
Problem statement: New B2B admins fail during the first 14-day trial to invite their first 5 users and connect them with data, which causes 62% of trials to end without activation.
Symptoms (selection): Trial activation rate 38% (vs. target 60%); 71 support tickets in 30 days with keyword SSO; average time-to-invite 4 days 6 h.
Impact: ARR risk of 380k EUR in the current quarter at current pipeline volume.
Out of scope: Self-service B2C trials; enterprise contracts above 50k EUR (own AE process).
Success criterion: Trial activation 60% by end of Q3, measured through Amplitude funnel trial_activated.
Owner next step: @ben (Product Lead), opportunity interviews by 2026-05-26.
Recognize symptoms and steer against them
Stickies contain feature wishes ("we need an SSO wizard"), not observations.
Move solution stickies to parking. Ask back: which problem would this solution address? Only then return to framing.
Problem statement covers multiple user groups and use cases ("onboarding does not work").
Sponsor forces narrowing: one target group, one situation, one measurable impact. The rest becomes a follow-up workshop.
Workshop ends with a statement nobody can approve with commitment.
Workshop without sponsor is either cancelled or clearly declared as preparation. Otherwise the statement becomes a third opinion next to roadmap and strategy.
Symptoms are based on impression ("many complain") without numbers or tickets.
Require at least one piece of evidence per symptom (ticket ID, metric, quote). Mark unproven symptoms as hypothesis and create validation spike.
Everyone quickly agrees on a harmless formulation, nobody disagrees.
Facilitator forces a second alternative formulation with a different perspective, for example from Sales. Vote statements against each other instead of mixing them.
Statement reads well, but nobody can say when the problem would be solved.
Define a quantitative or behavioral success criterion per statement, with measurement source. Without criterion, the statement is mood, not mandate.
Done signals checkable in under a minute
Go to the profile for purpose, similar methods, and sources or continue to the next method in the catalog.