methodatlas
RunsheetInnovation

Brainwriting

ComplexityLow
Time15-30 min
Participants3-12
FormatWorkshop + async
MaturityEstablished
01

Prerequisite

What needs to be finished first

Complete firstPrompt definition

A clear brainwriting question or How-Might-We statement exists and is understandable in 1-2 sentences.

Without: Without a clear prompt, participants generate ideas for different problems and the output is not clusterable.
Complete firstShared contextnot in catalog

All participants understand the domain and the problem situation, ideally through a prior briefing.

Without: Without context, ideas rest on inconsistent assumptions and cannot be compared.
02

Preparation

What needs to be ready before start

Materials

6-3-5 template (6 people × 3 ideas × 5 rounds) on A4 sheets or digital (Miro, FigJam, Google Docs); timer; visible prompt; quiet setup.

People / roles

One facilitator to hold the rounds and the timer; 4-12 participants from different functions; ideally mixed experience levels (anti-hierarchy effect).

Pre-read

Prompt; context briefing (2-3 min or pre-material); examples of good idea phrasing (full sentence, not keyword); anti-examples (too vague, too finished).

Time needed

15-30 min writing phase, plus 10-15 min condensation

Setup

One sheet per person with 3 fields × 5 rounds (classic 6-3-5 format) or one slot per round digitally. Prompt visible at the top. Set the timer to 5 min per round.

03

Core question

The one question this method answers

What range of ideas can we generate when each person first writes quietly and builds on others' ideas?

04

Flow

Marker: Phase

StepDurationActionHint
1Phase 1: Briefing and rules
5 minRead the prompt and recap the context. Explain the format: everyone writes 3 ideas per round, passes the sheet along, and builds on predecessor ideas. Silence during writing.If someone does not understand the format, the rotation stalls. Show a format demo with an example sheet.
2Phase 2: Round 1 - first ideas
5 minEach person writes 3 ideas alone on the sheet or digitally. Silence. After 5 minutes, pass the sheet along (clockwise, or switch the digital slot).The first round yields obvious ideas. That is fine; they become the base for building in later rounds.
3Phase 3: Rounds 2-5 - build on ideas
20 minEach follow-up round lasts 5 minutes. Participants read the 3 predecessor ideas and write 3 new ones that build on or combine them. Silence, no talking.Building does not mean copying. Variation, combination, and counter-proposals are allowed. Copying verbatim misses the point.
4Phase 4: Collect and review
10 minPut all sheets (or slots) on the wall or board. Silent reading phase. Consolidate duplicates and form clusters (analogous to affinity mapping).With 6 people, 90 ideas are created (6 × 3 × 5). Volume is normal; clustering reduces it to 5-10 themes.
5Phase 5: Convergence
10-15 minDot voting on clusters or top ideas. 3-5 dots per person. Mark the top 5 for follow-up. Name an owner per top idea.Convergence is mandatory. Without a selection, the output remains a pile with no effect.
05

Artifact

What comes out at the end

Form

Brainwriting collection (photo, scan, or digital export) with all 6-3-5 sheets, consolidated theme clusters, and a top shortlist with owner and next step.

Versioning / ownership

Create a separate snapshot per session with date and prompt. Store the condensed shortlist in the research or discovery repository. If the same prompt is used again, keep the previous version as a reference.

Tool alternatives
  • Physical 6-3-5 format on A4 sheets
  • Miro with a 6-3-5 template
  • Google Docs with a table (one column per person)
  • FigJam with slot frames
  • Notion with database rows per round

workshop-agenda-markdown.md

Compact agenda template for method sequences with timebox, purpose, and output.

Workshop Agenda

Goal: ... Participants: ... Duration: ...

TimeMethodPurposeOutput
00:00Check-inSet focusShared start
00:10How Might WeIdeation questionsHMW list
00:35BrainwritingGenerate ideasIdea pool
01:10Dot VotingPrepare selectionPrioritized ideas
01:30StoryboardingMake the concept tangibleStoryboard

Close-out

Decision, owner, next test.

06

Example output

Concrete filled scenario, fictional example

brainwriting-beispiel.md

Concrete filled scenario, fictional example

Brainwriting — onboarding aha moment (CW 19/2026)

Prompt: "What could give a solo tax advisor a clear aha moment within the first 30 minutes with our software?"

Setup: 6 people × 3 ideas × 5 rounds = 90 ideas, 25 min writing phase

Example flow of one sheet

Round 1 (Anna):

  • Pre-filled demo client with real documents
  • DATEV export button immediately visible
  • 5-click first entry path

Round 2 (Ben builds on it):

  • Demo client with stories (e.g. "VAT dispute from the last quarter")
  • DATEV export with preview ("This is what it looks like in DATEV")
  • 5-click first entry path + success confetti at the end

Round 3 (Lisa builds on it):

  • Demo story as a tutorial with "You are the tax advisor now - step 1: ..."
  • Live DATEV preview while entering data
  • Success confetti + an immediate sharing element ("Show this to your client")

...and so on

Consolidated clusters (10 themes)

  1. Pre-filled realistic demo data (16 ideas)
  2. DATEV export front and center (14 ideas)
  3. Guided 5-click first entry (12 ideas)
  4. Migration magic (10 ideas)
  5. Personalized first-question answer (8 ideas)
  6. Aha moment trigger (animation, confetti) (8 ideas)
  7. Live coach element (7 ideas)
  8. Client sharing as reward (6 ideas)
  9. Industry-specific examples (5 ideas)
  10. Outliers strewn across the board (4 ideas)

Top 3 after voting (15 dots distributed)

  1. Pre-filled demo data (8 dots) - owner @lisa, prototype CW 22
  2. DATEV export prominent (5 dots) - owner @anna, solution sprint CW 21
  3. Guided 5-click first entry (4 dots) - owner @ben, wireframes CW 22
07

Pitfalls

Recognize symptoms and steer against them

Trap

Talking while writing

Symptom

Participants discuss quietly, silence breaks, and bias reduction disappears.

What to do

The facilitator interrupts politely. If needed, separate rooms (including separate digital channels). Silence is non-negotiable.

Trap

Ideas are too keyword-like

Symptom

Keywords without context ('AI', 'better', 'faster') mean the next person cannot build on them.

What to do

Rule: one full sentence per idea. Show examples before starting. Flag short ideas during review and let the author expand them.

Trap

Forgetting to build on ideas

Symptom

Each round produces completely independent ideas and the building effect is lost.

What to do

Repeat the format rule explicitly. If forgotten, reading the previous 3 ideas is a mandatory step before writing.

Trap

Too many rounds

Symptom

After round 5, energy drops and the last rounds produce junk ideas.

What to do

Do not force classic 6-3-5 to five rounds. If energy drops, stop after 3-4 rounds.

Trap

No convergence

Symptom

90 ideas are generated, nobody chooses, and everything goes into the wiki.

What to do

A convergence phase is mandatory. Voting or a decider decision in the same session. Without a shortlist, the method was just busywork.

08

Stop criteria

Done signals checkable in under a minute

The prompt is not clear or refers to a problem that is too vague.
Participants have no shared context, so ideas would be arbitrary.
Silence cannot be enforced and group dynamics take over.
Fewer than 4 people, so the building effect is too small.
The stakeholder expects an immediate decision, but brainwriting is ideation, not decision making.
There is no time for convergence, so the output remains unused.

Finished the runsheet?

Go to the profile for purpose, similar methods, and sources or continue to the next method in the catalog.