What needs to be finished first
A current AARRR overview with values per stage exists so Acquisition, Activation, Retention, Referral, and Revenue can be supported with numbers.
A documented Customer Journey for the main user segment exists, from which steps and value moments can be derived.
What needs to be ready before start
Whiteboard or Miro board with circular template (4-6 segments); pens; three colors of sticky notes (activities, friction, amplifiers); printed or tab with current funnel numbers; list of top-3 competitors for comparison perspective.
One facilitator who walks through the loop step by step; one Product or Growth owner; two to three people from Marketing, Sales, Customer Success, or Product Analytics; one scribe for leverage hypotheses.
Current funnel-stage metrics for the last 90 days; NPS or CSAT values if available; known referral sources; list of top-5 support themes; current retention cohorts.
60-120 min
Draw a circle with five segments on the board (Attract, Activate, Use, Succeed, Refer). Make segment metrics visible on a board or separate tab. Define three color codes and label them: green for amplifiers, red for friction, blue for leverage hypotheses.
The one question this method answers
Which friction points are slowing the loop most today, and which two to three levers would make the next cycle more likely?
Marker: Phase
| Step | Duration | Action | Hint |
|---|---|---|---|
1Phase 1: Name loop steps | 15 min | For each segment of the circle, define key activities and transitions briefly. What does the user do, what does the product do, what happens between segments. | If a segment is empty, either product has a blind spot there or the stage does not belong in this flywheel. Name it explicitly rather than skipping. |
2Phase 2: Mark friction | 20 min | Place red notes for friction points in each segment: where users drop off, where wait time occurs, where value proof is missing. Friction notes must be supported by a metric or support quote. | Friction without evidence is intuition. If no source is available, label note as "hypothesis, no evidence" and turn into discovery task. |
3Phase 3: Identify amplifiers | 20 min | Place green notes for amplifiers between segments: what speeds up the next cycle. Examples: successful recommendation trigger; data that shortens onboarding; user groups generating content for others. | Amplifiers work across segment boundaries, not within one segment. If all green notes sit in one segment, loop effect is missing. |
4Phase 4: Derive levers and experiments | 25 min | For each top-3 friction, formulate a leverage hypothesis with expected effect and measurement metric. Prioritize into experiment backlog with owner and target date. | Maximum three levers per quarter into execution. More dilutes attention and makes each lever's impact harder to see. |
What comes out at the end
Flywheel map as diagram (PNG or Miro snapshot) plus Markdown document with segment descriptions, friction list with evidence, amplifier list, and experiment backlog with owner, hypothesis, and success metric.
Create a new version with date for each quarterly review. Mark pre-quarter friction and amplifier notes as "done/open/discarded" instead of deleting them. Archive hypotheses with measured outcome.
Compact working template for Flywheel with context, input, output artifacts, and next step.
What is this method used for?
Which question should be answered at the end?
Which data, observations, or materials are available?
Owner, date, success signal.
Concrete filled scenario, fictional example
Concrete filled scenario, fictional example
Segments and metrics
Friction points
Leverage hypotheses (Q2)
Recognize symptoms and steer against them
Diagram is drawn as a loop, but logic is linear: stages follow one another without feedback into earlier segments.
Explicitly define for each segment boundary how output from this segment becomes input for an earlier segment. If no feedback exists, the model is a funnel, not a flywheel.
Red notes are based on anecdotes ("someone said so") rather than metrics or quotes.
For each friction, name a source: dashboard, support ticket ID, or interview quote. If no source exists, mark it as discovery task and do not escalate into leverage hypothesis.
More than 10 leverage hypotheses at end of workshop, no prioritization cut-off.
Trim to maximum three levers per quarter. Criteria: expected effect on top friction, implementability, measurability. Keep rest in backlog as "later."
One segment has 20+ activity notes, others are nearly empty.
Segment size indicates focus or blind spots. Split an overloaded segment into two or clean it up intentionally. Fill empty segments via interviews or data analysis.
Flywheel diagram hangs in wiki, no one measures lever outcomes or updates figures.
Define fixed measurement cadence for each lever (for example every two weeks). Schedule quarterly review in calendar. Diagram is a living artifact, not wall decoration.
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.