agentleFS
ejwhite7ejwhite7/product-release-checklist/skills/launch-measure/SKILL.md

launch-measure

Skill0 starsChanged 35 days agoMITView on GitHub

Plan and read feature adoption measurement for a launch in the release checklist studio. Covers the activation definition, events, baselines, targets, guardrails, holdout, and the T+7/T+30/T+90 readouts. Use for "Plan measurement" or "Write readout" requests, or for "set launch goals", "tracking plan", "how is the launch doing", or "feature adoption".

The file

We split this skill into the rules it gives an agent

2 rules. About 0 are common, 0 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Write the basis in item metric's note, labelled as a benchmark until there's real history.

    line 19
  2. Very specific:

    If an analytics MCP is connected (PostHog, Amplitude, Mixpanel, BigQuery, Supabase…), query the trailing 4 weeks of the behaviour this …

    line 23

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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

Launch Measure

Works on measure in the launch JSON. The Measure tab shows the funnel (eligible → exposed → discovered → first use → activated → retained) with targets and actuals.

Plan (before T-14)

  1. activation. Define it as the smallest repeatable action that delivers the value, within window. Reject "opened the feature". Offer 2–3 options in chat and pick the best.
  2. primary. Phrase it as activated users ÷ eligible users by T+30.
  3. events. Rename the placeholder <slug>_core_action. Keep entry_point on the feature-open event, because it's how adoption gets attributed to channels.
  4. Targets (stages.*.t7, t30).
    • Base them on launches/benchmarks.md (in the export folder) from past launches.
    • With no history, start from 20–40% of eligible users discovering the feature by T+30 for a well-promoted T1, and 30–50% of those activating.
    • Write the basis in item metric's note, labelled as a benchmark until there's real history.
  5. guardrails and holdout. Size the holdout and note in chat that an analyst should check it's big enough for the effect you expect.

Baseline

If an analytics MCP is connected (PostHog, Amplitude, Mixpanel, BigQuery, Supabase…), query the trailing 4 weeks of the behaviour this feature replaces, and the eligible-user count, into stages.*.baseline and eligibleCount. If none is connected, write the exact queries or insight definitions into item baseline's note.

Readouts (T+7, T+30, T+90)

Pull the actuals into stages.*.a7 / a30 / a90 and channels. Never invent a number: leave the cell blank and put the missing query in the notes. Then write readouts.<n>:

  • T+7: exposure and discovery against target. Name the biggest drop-off stage and the fix for it.
  • T+30:
    • activation against target
    • holdout lift (treated activation rate minus holdout activation rate, with a confidence interval)
    • cost per activation by channel
    • guardrails and support themes
  • T+90: retention and revenue of adopters vs matched non-adopters. Say plainly that it's correlational unless it comes from the holdout.

Set verdict, recommendation and at, and keep notes to the decisions and the reasons for them.

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.