activation
Skill0 starsChanged 2 months agoMITView on GitHub
Find the moment that predicts retention and get more people to it. Covers identifying the activation event from data rather than intuition, time-to-value, and why most onboarding work targets the wrong step. Use when new users do not stick, or before redesigning onboarding.
In this repository9 other files it gives its agents.
- capacity-economicsSkill
- channel-fitSkill
- data-trustSkill
- experiment-designSkill
- form-frictionSkill
- funnel-diagnosisSkill
- pricing-decisionsSkill
- retention-diagnosisSkill
- unit-economicsSkill
The file
We split this skill into the rules it gives an agent
2 rules. About 0 are common, 2 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
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
- Distinctive: line 26
Then check causality is plausible: If the action is something only already committed users would do, you have found a symptom.
- Distinctive: line 56
No cohort has enough history to observe retention.
≈ “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.
Files
Activation
Activation is the earliest observable behaviour that predicts long-term retention. It is found in data, not chosen in a workshop.
Finding it, not guessing it
The common failure is picking a plausible-sounding milestone — "completed profile", "invited a teammate" — and optimising for it. Plausible milestones correlate with retention because engaged people do everything, not because that step causes anything.
Method:
- Take cohorts old enough to have observed retention
- For each candidate early action, split users into did and did not
- Compare retention at a fixed later point
- The candidate with the largest separation and enough volume on both sides is your activation event
Then check causality is plausible. If the action is something only already committed users would do, you have found a symptom. The useful activation event is one you can cause through design.
Time to value matters as much as the event
The same event reached on day 1 predicts far better than reached on day 20. Track both whether and how quickly. Shortening time-to-value is often easier than raising completion, and pays more.
Where onboarding work usually goes wrong
- Optimising the step with the biggest drop. Signup-to-first-action always has the largest drop. It is also where the least committed people leave, and many of them should
- Adding education instead of removing steps. A tour explains a form that should be shorter. Removing a field beats explaining it
- Front-loading setup. Anything asked before value is delivered is a tax on people who do not yet know if they want it. Defer everything you can
- One path for everyone. Users arrive with different intent; a single linear onboarding serves the median and loses the tails
The counter-check
Higher activation with lower retention among activated users means the funnel is pushing people through a milestone without delivering the value it represents. Always report both together.
When to refuse
- No cohort has enough history to observe retention. State how long is needed
- Fewer than ~200 users per arm of the comparison
- Asked to validate a pre-chosen activation metric. Test it against the data and say what the data supports
Output
The candidate activation events ranked by retention separation, with volume on each side. The recommended definition, why it is plausibly causal rather than correlated, current completion rate and median time to reach it, and one change aimed at the largest addressable drop before that moment.
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
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.

