agentleFS
45ck45ck/marketing-product-skills/.claude/skills/retention-diagnosis/SKILL.md

retention-diagnosis

Skill3 starsChanged 8 months agoMITView on GitHub

Diagnose retention problems using lifecycle stages, behavior patterns, and practical product/growth interventions.

1 other project uses a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

7 rules. About 0 are common, 7 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈7

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use this skill when the user wants to understand why users or accounts churn, flatten retention curves, or identify behaviors correlated …

    line 9
  2. Distinctive:

    If data is unavailable, define the diagnostic plan and the events needed.

    line 39
  3. Distinctive:

    Look for both drivers and detractors.

    line 45
  4. Distinctive:

    Do not confuse correlation with causation.

    line 46
  5. Distinctive:

    If the user shares a retention curve or cohort chart:

    line 168
  6. Distinctive:

    If the product is B2B:

    line 173
  7. Distinctive:

    consider gross and net revenue retention where relevant

    line 176

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Retention Diagnosis

Use this skill when the user wants to understand why users or accounts churn, flatten retention curves, or identify behaviors correlated with retention.

What this skill must produce

Always produce:

  1. Retention problem statement
  2. Lifecycle framing
  3. Diagnostic hypotheses
  4. Behavioral drivers and detractors
  5. Measurement plan
  6. Intervention ideas
  7. Experiment backlog
  8. Risks and caveats

Inputs to gather

If present, extract:

  • product type
  • retention data by cohort
  • D1/D7/D30 or weekly/monthly retention
  • account retention vs user retention
  • critical event
  • activation event
  • churn reasons
  • product segments / personas
  • pricing tier / ACV / plan differences
  • existing hypotheses

If data is unavailable, define the diagnostic plan and the events needed.

Working rules

  • Separate new-user retention from current-user retention and resurrected-user retention.
  • Tie retention to the critical event, not arbitrary activity.
  • Look for both drivers and detractors.
  • Do not confuse correlation with causation.
  • Identify where product quality issues may distort retention before recommending growth tactics.

Step-by-step method

Step 1: define the retention lens

State whether this is:

  • user retention
  • account retention
  • revenue retention
  • logo retention
  • feature retention

Then define the time interval that matches the product’s natural usage rhythm.

Step 2: frame lifecycle stages

Split the problem into:

  • new users
  • current users
  • dormant / resurrected users

Identify which stage appears most broken.

Step 3: define the critical event

Choose the action that best reflects value. Explain why generic activity is insufficient.

Step 4: generate diagnostic hypotheses

Examples:

  • users do not reach the aha moment
  • activation occurs but habit loop never forms
  • retained accounts lack multi-user adoption
  • feature adoption is shallow
  • poor product quality causes hidden churn
  • wrong segments are entering the funnel
  • plan packaging mismatches true value

Step 5: identify likely drivers

Specify behaviors that may correlate with stronger retention. If relevant, separate:

  • onboarding-phase drivers
  • value-discovery drivers
  • current-user stickiness drivers
  • resurrection triggers

Step 6: identify likely detractors

Specify behaviors or product states that may depress retention:

  • crashes / bugs
  • empty states
  • poor matching / inventory
  • slow time-to-value
  • over-complex workflow
  • weak reminders / triggers
  • poor collaboration adoption
  • pricing mismatch

Step 7: create a measurement plan

Define:

  • cohorts to compare
  • retention intervals
  • event sequences to inspect
  • power-user behaviors to compare
  • segments worth splitting

Step 8: create interventions and experiments

Convert the most likely causes into tests.

Output format

Retention problem statement

  • What seems wrong:
  • Who is affected:
  • Time horizon:
  • Primary business consequence:

Lifecycle framing

StageWhat good looks likeWhat seems broken

Critical event

  • Event:
  • Why it reflects value:
  • Why generic DAU/WAU/MAU may be misleading:

Diagnostic hypotheses

HypothesisWhy plausibleEvidence needed

Behavioral drivers and detractors

Likely drivers

BehaviorLifecycle stageWhy it may matter

Likely detractors

DetractorLifecycle stageWhy it may matter

Measurement plan

  • Cohorts:
  • Intervals:
  • Segment cuts:
  • Key funnel comparisons:
  • Product-quality checks:
  • Needed events/properties:

Intervention ideas

InterventionWhy it could helpExpected impact area

Experiment backlog

ExperimentHypothesisSuccess metricMinimum viable test

Risks and caveats

  • Correlation vs causation:
  • Missing data:
  • Segment distortion:
  • Seasonality / lifecycle mismatch:
  • Product-quality blockers:

Special instructions

If the user shares a retention curve or cohort chart:

  • interpret the shape directly
  • say whether it looks like no product-market fit, weak activation, or weak habit formation
  • be explicit about what can and cannot be inferred

If the product is B2B:

  • assess retention at both user and account levels
  • inspect whether retained accounts have collaboration / seat / workflow depth
  • consider gross and net revenue retention where relevant

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.