Form
Workshop record with date, participants, selected top problem, HMW reframe, top solutions, chosen experiment with owner, date, and success metric plus follow-up date.
RunsheetFacilitation
Decision Jam
ComplexityLow
Time60-90 min
Participants4-10
FormatWorkshop
MaturityEstablished
01
Preparation
What needs to be ready before start
Materials
Whiteboard or Miro with areas for problem collection, voting, HMW reframe, solution ideas, and decision; stickies; timer; action list; shared pre-read.
People / roles
One facilitator; 4-10 participants; one decider (often from the team, for example PM or team lead); one scribe.
Pre-read
Context: what is currently occupying the team; known tensions; time pressure or impetus; note that the format targets fast convergence.
Time needed
60-90 min
Setup
Create wall sections. Rule: each phase strictly timeboxed, no preemptive discussion. Clarify decision authority in advance.
02
Core question
The one question this method answers
Which problem is most important right now, and which concrete first step changes it in the next 14 days?
03
Flow
Marker: Minute
| Step | Duration | Action | Hint |
|---|---|---|---|
10-10 min | 10 min | Silent work: write problems or frustrations silently on stickies, max 5 per person. Keep phrasing short. | Silent work prevents dominance. If someone speaks immediately, they influence others. No discussion in this phase. |
210-20 min | 10 min | Round-robin share, cluster, voting (3 dots per person). Identify top problem. | Voting is silent and simultaneous. The top problem becomes the focus of remaining time. Make a clear decision, not consensus seeking. |
320-30 min | 10 min | Reframe top problem as a "How Might We" question. Generate several HMW variants, then select one. | HMW is the method core. "How could we reduce X so Y happens?". If reframing is too narrow, solution space shrinks. |
430-50 min | 20 min | Silent solution idea generation for HMW: Crazy 8s or brainwriting, minimum 5 ideas per person. Then share round-robin. | Quantity before quality. Stopping after 5 minutes means insufficient exploration. Write even absurd ideas; they often increase creativity. |
550-70 min | 20 min | Voting (3 dots per person). Select top solution. Discuss what the smallest experiment step is in 14 days. | Run an experiment instead of a program. Not "we solve X," but "we test Y and learn." |
670-90 min | 20 min | Draft experiment: what, who, when, and what indicates success or failure. Maximum 2-3 experiments. | Owner must be present and approve. Success criterion must be measurable or observable within 14 days. |
04
Artifact
What comes out at the end
Versioning / ownership
One entry per Decision Jam. If multiple jams on the same topic, link them and integrate experiment outcomes.
Tool alternatives
- Miro with Decision Jam template
- FigJam with columns
- Whiteboard with photo export
- Notion page with sections
decision-jam-working-template.md
Compact working template for Decision Jam with problem framing, solution ideas, voting, and next action.
Decision Jam Working Template
Goal
Turn a stuck decision into a clear choice through framing, divergence, and convergence.
Context
What decision is stuck, and what outcome should the group reach?
Input
- Decision question:
- Constraints:
- Stakeholders:
- Evidence or observations:
Working area
- Problem framing:
- How might we questions:
- Solution ideas:
- Criteria:
- Voting or selection:
Output artifacts
- Shortlisted options:
- Decision note:
- Next action:
Open questions
- ...
Decision / next step
Owner, date, and success signal.
05
Example output
Concrete filled scenario, fictional example
decision-jam-beispiel.md
Concrete filled scenario, fictional example
Decision Jam — Team Backend, "What is currently slowing us down the most?" (24.05.2026, 75 min, 6 Engineers)
Top problems (voted):
- PR review takes 1-3 days (12 dots)
- Unclear Definition of Done (8 dots)
- Database migration pain points (5 dots)
Top problem selected: PR review time.
HMW: How can we complete PR reviews within 4 working hours without reducing quality?
Solution ideas (excerpt):
- Review quota for 1 PR per day per engineer
- Pair review as default for complex PRs
- Fixed 11:00-12:00 PR-review slot
- Enforce smaller PRs (max. 300 LOC)
- Bot reminder after 4 hours
- Async review template with checklist
Top solution (voted): Fixed 11:00-12:00 PR review slot + reminder bot after 4 hours.
Experiment (2 weeks from 27.05.):
- Daily 11:00-12:00 review slot, everyone closes reviews. Owner: @anna (Engineering Lead). Success: 80% of PRs open less than 4h.
- Slack bot that pings when PR has been open for more than 4h. Owner: @ben, configured by 27.05.
Review: 10.06.2026 in next retro.
06
Pitfalls
Recognize symptoms and steer against them
Trap
Collection phase becomes discussion
Symptom
Initial participants speak too long while others lose their own problems.
What to do
Keep silent work absolute. Only share after 10 min; max 15 s per sticky.
Trap
HMW is too narrow
Symptom
HMW already contains a solution ("How could we introduce Tool X"), shrinking solution space.
What to do
HMW should stay open to multiple solution paths. If phrased too narrowly, return to problem level.
Trap
Experiment too large
Symptom
"Implement solution in six weeks," and the format loses Decision Jam character.
What to do
Choose experiment in 14 days. Prefer learning before solving. Larger work should go to a follow-up sprint with findings.
Trap
Decision becomes consensus seeking
Symptom
Voting result gets questioned, top problem is over-discussed, energy dissipates.
What to do
Make voting result binding. If someone wants to change top problem, they need a prior vote. Decider can protect the voting result.
Trap
Owner absent
Symptom
Experiment has an owner who was not present in workshop and did not know details.
What to do
Only owners in the room receive actions. If someone is missing, appoint an alternative person or reschedule.
07
Stop criteria
Done signals checkable in under a minute
Problems are strategic and need depth; Decision Jam is too shallow and a Consensus Workshop or RAPID is more suitable.
Group is larger than 10, format becomes difficult to moderate.
Psychological safety is missing, and collection phase gets filtered by participants.
Decision authority is not in the room and experiment owner cannot commit.
Not enough time (under 60 min), phases become rushed.
Top problem is outside team mandate and experiment is not feasible.
Finished the runsheet?
Go to the profile for purpose, similar methods, and sources or continue to the next method in the catalog.