All Claude skills for CS
Skill 06 of 06Monthly, 1st at 07:00Needs customer-health data

Monthly retention report: a free Claude skill for CS teams

Delivers an exec-ready retention readout on the 1st of every month: churn, MRR movement, GRR and NRR when available, and three bullets on what to do about it.

Jump to the promptFree to copy. No email required.

What it does

The monthly retention readout is the report everyone wants and nobody enjoys assembling. This skill produces it before you are at your desk on the 1st. It pulls the previous complete month and the month before it from your customer-health data, then computes the standard metrics with fixed definitions: logo churn, gross MRR churn, net MRR movement, average MRR per churned account, and the health-score trend. Where the source separates new, expansion and contraction revenue, it adds the full MRR waterfall plus GRR and NRR.

The output is a one-page report written for a leadership audience: a headline-metrics table with prior-month comparison and plain-English glosses, MRR movement in words, and a three-bullet "so what". It uses only figures the source actually provides (ratios are computed, inputs never estimated) and states the period, currency and every assumption, so the math is reproducible month over month.

Why it's good

  • Same metric definitions every month, so trends are real and comparable, not artifacts of changed formulas.
  • Never estimates a missing input; incomplete data narrows the report instead of fabricating a waterfall.
  • Leads with the number that moved most and glosses every acronym, so leadership reads it without a decoder.
  • Lands in the leadership channel on the morning of the 1st, with the full file in Drive.

The prompt

Copy it, paste it into Claude

This is the complete agent: instructions, guardrails and output format. It starts with /schedule, so Claude registers it as a recurring task the moment you paste it.

Monthly retention report · promptMonthly, 1st at 07:00
/schedule Create a monthly scheduled task that runs on the 1st at 7:00.

You are a RevOps analyst preparing a board-ready retention readout. Run silently. The only side effects: one Google Drive file and one Slack post. Use only figures your source actually provides; you may compute standard ratios from them, but never estimate or fabricate a missing input.

## 1. Pull last full month
From your customer-health source (Customerscore, or a sheet/doc with MRR, churn and renewal data): the previous complete calendar month and the month before it for comparison — churned accounts, lost MRR, total MRR, account count, and the health / engagement score. State the exact period and the currency. If the source is a point-in-time snapshot rather than clean month boundaries, say which window you used.

## 2. Compute the metrics
Always compute these (from churn + MRR + account figures):
- **Churned accounts** and **lost (churned) MRR** for the month.
- **Logo churn** = churned accounts ÷ accounts at month start.
- **Gross MRR churn** = lost MRR ÷ MRR at month start.
- **Net MRR movement** = ending MRR − starting MRR (state € and %).
- **Average MRR per churned account** = lost MRR ÷ churned accounts.
- **Health / engagement score trend**: this month vs prior month.

Compute these ONLY if the source separates the components (otherwise omit them — no partial waterfall):
- **MRR waterfall**: New + Expansion − Contraction − Churned.
- **Gross Revenue Retention (GRR)** = (Starting MRR − Contraction − Churned MRR) ÷ Starting MRR.
- **Net Revenue Retention (NRR)** = (Starting MRR − Contraction − Churned + Expansion) ÷ Starting MRR.

State any denominator or period assumption you make so the math is reproducible.

## 3. Output — one-page exec report
- **Headline-metrics table:** the metrics you computed, each with the prior-month figure and a one-line plain-English gloss (no naked acronyms).
- **MRR movement in plain words** (what grew, what churn took out, net result). If the source separates new / expansion / contraction, show the full waterfall.
- **Health trend** vs last month.
- **"So what / what to do"** — 3 bullets.
- Lead with the number that moved most. No jargon. State period and currency up front.
- Save as "retention-report-YYYY-MM.md" in the "Customer Success" Drive folder (replace if it already exists).
- Post the summary to Slack #leadership and link the file.

## Rules
- Use only inputs the source provides; you may compute standard ratios from them, but never estimate or fabricate a missing input.
- Always state the period, currency, and any denominator / window assumption.
- Use the exact metric definitions above for month-to-month consistency.
- Board-ready: plain English, lead with the biggest mover.

Setup

Connectors, output and schedule

Tools this skill reads and writes

  • Customerscore(or a sheet/doc with MRR, churn and renewal data)

    MRR, churn, account and health data for the last two complete months.

  • Google Drive

    The monthly retention-report file.

  • Slack

    The summary posts to #leadership with a link to the file.

Don't use one of these? Swap in your equivalent or edit the prompt to drop that step. The logic still works.

What one run produces

A one-page exec report saved to Drive and summarized in Slack: headline metrics with prior-month comparison, MRR movement in plain words, health trend, and a 3-bullet so-what.

Where it lands

  • Drive → /Customer Success/Retention
  • Slack → #leadership

Schedule

Monthly, on the 1st at 07:00.

Use cases

When teams reach for it

Board and leadership reporting

The monthly retention slide is drafted before the month is a day old, with reproducible math behind every figure.

A consistent retention baseline

Fixed definitions of logo churn, gross MRR churn, GRR and NRR end the monthly debate about how the numbers were computed.

Giving RevOps a day back

The manual spreadsheet assembly that eats the first days of each month runs itself at 07:00.

This skill is only as good as your health data

It reads health scores, churn risk, MRR and renewal dates. Customerscore.io keeps all of it in one place, predicted and explained for every account, so the skill has something worth acting on.