agentleFS
Sign inStart free
kyle-olmsteadkyle-olmstead/cs-claude-engine/skills/retention-report/SKILL.md

retention-report

Skill1 starsChanged 23 days agoApache-2.0View on GitHub

This skill should be used when the user says "what's our NRR", "GRR this quarter", "net revenue retention by segment", "logo vs dollar churn", "retention report", "churn concentration", "expansion pipeline velocity", "ARR per CSM", "retention cost", "LTV to CAC", "QBR completion rate", or asks any question about lagging retention or CS efficiency metrics. It computes the engine's retention and capacity metrics from the warehouse marts when connected, otherwise from CRM/billing records, using the exact formulas in the metrics taxonomy, and presents them with cohort definitions, caveats, and the decomposition (expansion, contraction, churn, new logos) so the numbers can be defended.

The file

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 1 is distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    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:

    Run the functions with Python when available (write the input JSON and a five-line script into cs-engine/runs/<today>/).

    line 44
  2. Very specific:

    Never annualize a monthly NRR with ^12.

    line 56
  3. Distinctive:

    Do not invent cost, margin, or CAC inputs.

    line 58

≈ “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

Retention & efficiency report

Compute the lagging and capacity metrics from ${CLAUDE_PLUGIN_ROOT}/docs/02_METRICS_TAXONOMY.md Parts A and D with the exact formulas there. Read the relevant section before computing so the cohort rules (start cohort = ARR > 0 at t0; new logos excluded from NRR/GRR; ramps tagged) are applied, and state them in the answer.

Source order

  1. ~~data warehouse with dbt/models/marts/mart_retention_nrr_grr.sql built → query the mart for the period, window (monthly / ttm), and segment. This is authoritative; use it as-is.
  2. ~~billing subscription history → build two ARR snapshots (t0, t1) per account with segment, write them as JSON, and run:
    python3 - <<'EOF'
# see src/models/metrics_calculator.py::cohort_retention_by_segment for the exact call
EOF
    (Write a short script to the working folder that loads the snapshots into ARRSnapshot objects and prints cohort_retention_by_segment(...); keep the script with the run's outputs.)
  3. ~~crm ARR property history or closed-won/closed-lost renewal opportunities → same as 2 with reduced confidence; say that CRM-derived ARR is not a billing-grade number.

Metrics and functions

AskFunction (src/models/metrics_calculator.py)Needs
NRR, GRR, logo churn, dollar churn, net dollar churn, churn concentrationcohort_retention_by_segment(t0, t1, ...)ARR snapshots at t0 and t1
Expansion pipeline velocity, coverageexpansion_pipeline_velocity(opps, as_of, remaining_quota)~~crm opportunities of type expansion/upsell/cross_sell, open + closed in 180d
ARR per CSM, balance index, over-capacity flagsarr_per_csm(accounts, segment_targets)accounts with owner, ARR, segment, health band
CRC, CRC per $ retained, CRC ratiocustomer_retention_cost(total_cs_cost, customers, grr, start_arr)CS cost from the user (ask once; never guess headcount cost)
LTV:CACltv_cac(arpa, gross_margin, grr_ttm, cac)gross margin and CAC from the user (ask once)
QBR/EBR completion velocity, ARR coverageqbr_completion_velocity(qbrs)due/completed dates from ~~crm or ~~calendar, ARR per account

Run the functions with Python when available (write the input JSON and a five-line script into cs-engine/runs/<today>/); otherwise compute by hand and show the arithmetic.

Output

  1. Headline: the metric, the window, the segment, the number, one line each.
  2. Decomposition table: start ARR, expansion, contraction, churn, end ARR, new-logo ARR (excluded), logos in cohort, churned logos.
  3. Benchmarks from docs/02_METRICS_TAXONOMY.md Part E for the segment, with one sentence of interpretation (top quartile / median / below).
  4. Caveats that apply: ramp-driven expansion included or excluded, FX, merged accounts, sample size for win rates (< 20 closed opps → low confidence), CRM-derived ARR.
  5. Offer: save as a document in ~~documents, or chart the trend inline when more than one period is available.

Rules

  • Never annualize a monthly NRR with ^12.
  • GRR is capped at 100% by construction; if the data produces more, the cohort rule was broken: recompute.
  • Do not invent cost, margin, or CAC inputs; ask once and remember the answer in the stack profile for this user.

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.

Check your setup, free