agentleFS
forsvn-labsforsvn-labs/meta-skills/skills/measure-growth/SKILL.md

measure-growth

Skill14 starsChanged 3 months agoMITView on GitHub

Design growth measurement or learn from marketing results. Use for analytics and tracking plans, KPI trees, campaign measurement, attribution boundaries, experiment readouts, performance reviews, launch retrospectives, cohort or funnel analysis, deciding what to keep or stop, or converting observed results into bounded reusable learning.

The file

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 3 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    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:

    Do not turn correlation into causation.

    line 40
  2. Distinctive:

    Write each durable learning as:

    line 54
  3. Distinctive:

    Use the narrowest permitted data access.

    line 65

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 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

Measure growth outcomes

Connect measurement to a decision. Do not create dashboards without an operator action.

Define the decision system

Specify:

  • business outcome and decision owner;
  • user behavior that represents value;
  • primary signal and why it predicts the outcome;
  • diagnostic signals for reach, attention, comprehension, belief, motivation, friction, and value;
  • guardrails for quality, cost, retention, and harm;
  • segments and observation window;
  • decision date and keep, revise, stop, or scale thresholds.

Define event names, properties, identity rules, source of truth, and QA steps when implementation detail is requested. Distinguish leading, lagging, and diagnostic measures.

Read results carefully

Check:

  • exposure and opportunity volume;
  • baseline and comparable period;
  • audience, channel, device, geography, and customer mix;
  • instrumentation changes and missing data;
  • seasonality, releases, promotions, outages, and competitor movement;
  • downstream quality, revenue, or retention;
  • qualitative objections and customer language.

Do not turn correlation into causation. Prefer a comparison or test that creates different predictions for competing explanations. State uncertainty and accept inconclusive results.

Produce a decision

Return:

  1. result summary with denominator, baseline, window, and confidence;
  2. what changed and what did not;
  3. plausible mechanisms and alternative explanations;
  4. Keep, drop, test decision;
  5. next experiment with one intentional change;
  6. durable learning record.

Write each durable learning as:

  • observation;
  • audience, offer, channel, and time boundary;
  • evidence and confidence;
  • implication;
  • where it must not be generalized.

Promote a learning only from observed behavior or a documented test. Never invent unavailable analytics or silently treat missing observations as zero.

Use the narrowest permitted data access. Keep tracking changes, experiment activation, messages, and external writes behind explicit approval.

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.