agentsclimarketplace

Run cohort analysis

Skill alexe-ev/product-plugins/data-analytics/skills/run-cohort-analysis

Skill library for AI agents — 15 product domains, 121 skills. Tells the agent what to ask, how to reason, and what to output.

Install
npx -y skills add alexe-ev/product-plugins --skill run-cohort-analysis

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Structure and interpret a cohort analysis to understand retention, engagement, or behavior patterns over time. Use this skill when a team needs to understand how different user groups behave across their lifecycle.

SKILL.md

2.8 KB, as published. Nobody here has run it

Run Cohort Analysis

Purpose

Help teams design, run, and interpret cohort analyses that reveal retention patterns, lifecycle behaviors, and the impact of product or marketing changes on different user groups.

Skill type

Conceptual skill with calculation-aware components

Use this skill when

  • Retention trends need to be understood beyond aggregate numbers
  • The impact of a product change on different user groups needs to be evaluated
  • Acquisition cohorts need to be compared across channels or time periods
  • Behavioral differences between user segments need to be quantified

Do not use this skill when

  • The goal is a single-point metric (use detect-performance-signals)
  • The goal is an A/B experiment analysis (use analyze-experiment-results)

Required inputs

  • Cohort definition (what event groups users: signup date, first purchase, acquisition channel)
  • Metric to track per cohort (retention rate, revenue, feature usage)
  • Time period and data availability

Optional inputs

  • Breakdown dimensions (segment, plan, geography)
  • Benchmark or comparison cohorts
  • Product changes or events to annotate

Upstream context

Works best when:

  • Event tracking is instrumented
  • Retention metric is defined

Downstream handoff

Output can feed:

  • analyze-churn-retention (cohort patterns → churn root causes)
  • detect-performance-signals (cohort changes surface signals)
  • formulate-experiment-hypothesis (cohort insight → experiment idea)

Instructions

  1. Define the cohort grouping: time-based (weekly/monthly signup), behavior-based, or acquisition source.
  2. Define the metric and measurement window.
  3. Set up the cohort table: rows = cohorts, columns = time periods.
  4. Identify the retention curve shape: fast decay, stable plateau, improving trend.
  5. Compare cohorts to identify improvement or regression over time.
  6. Annotate cohort boundaries with relevant product or market events.
  7. Extract 2–3 key insights from the analysis.

Output

Provide:

  • Cohort definition and metric
  • Cohort table structure (describe if actual data not available)
  • Retention curve analysis: shape, plateau level, decay rate
  • Cohort-to-cohort comparison: improving / stable / worsening
  • Event annotations
  • Key insights (2–3 specific findings)
  • Recommended next steps

Risks / caveats

  • Cohort sample sizes must be large enough for meaningful comparison — flag small cohorts
  • Time-based cohorts can conflate seasonal effects with product quality changes
  • Don't act on a single cohort anomaly without checking for data quality issues

Gives 0 of the 12 instructions most data analysis skills give

Counted across 286 of the 286 authors here whose files we hold, read 2026-08-06

  • use excel formulas instead of hardcoded calculated valuesin 35 of 286, across 7 files
  • match existing template conventions when modifying filesin 35 of 286, across 7 files
  • document sources for all hardcoded valuesin 35 of 286, across 7 files
  • write minimal concise python codein 35 of 286, across 7 files
  • place all assumptions in separate assumption cellsin 32 of 286, across 5 files
  • apply industry-standard color coding to financial modelsin 31 of 286, across 5 files
  • format years as text stringsin 30 of 286, across 3 files
  • recalculate formulas using recalc.py after modificationsin 30 of 286, across 3 files
  • format negative numbers using parenthesesin 30 of 286, across 3 files
  • fix all identified formula errors before finishingin 27 of 286, across 1 file
  • use colorblind-safe palettesin 19 of 286, across 12 files
  • Name tests after the prevented bugin 13 of 286, across 8 files

Said here and by no other author read

  • define the cohort grouping
  • define the metric and measurement window
  • set up rows as cohorts and columns as time periods
  • compare cohorts to identify trends
  • annotate cohort boundaries with relevant events
  • extract two to three key insights

Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once.

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.