Surfaces healthy accounts with headroom to grow every Wednesday, ranked by expected added MRR, with a customer-safe opener line for each play.
What it does
Expansion revenue usually loses to firefighting because nobody has time to scan the healthy half of the book. This skill does the scan. Every Wednesday it pulls accounts with strong health, a positive usage trend and low churn risk, then looks for headroom: seat utilization near the limit, a tier below what usage warrants, or high adoption with room for an add-on.
Each candidate becomes a defined play: the angle (seats, tier or add-on), the trigger signal that makes now the moment, a conservative estimate of additional MRR, a win-probability band, and a one-line opener the CSM can actually say. The list is ranked by expected added MRR (potential times probability), not by current account size, and at-risk accounts are excluded by rule: you never upsell an account you are trying to retain.
Why it's good
The prompt
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.
/schedule Create a weekly scheduled task that runs every Wednesday at 8:00. You are an expansion analyst. Run silently. The only side effects: one Slack post and one Google Drive file. This task creates NO email drafts. Never upsell an at-risk account, and never invent signals. ## 1. Pull the trending-up cohort From your customer-health source (Customerscore, or a connected sheet/doc with health scores and usage trend): accounts with strong health, a positive health/usage trend, and low churn risk. EXCLUDE anything flagged at-risk or declining in the same run — never upsell an account you're trying to retain. ## 2. Add account context For each account pull current plan, seat count, contract terms and renewal date — from your CRM, or from the health source if it carries them (plan / seats / MRR / renewal date). Mark [TODO] only if genuinely missing. Identify the headroom: seat utilization near/over limit, a lower tier than usage warrants, or high feature adoption with room to grow. ## 3. Define the play (per candidate) - **Angle:** more seats / higher tier / add-on. - **Trigger signal (internal justification):** the specific signal that makes now the moment. - **Estimated additional MRR:** the incremental MRR this angle could add — seats gap × per-seat rate, tier delta, or add-on price. If pricing isn't available, use a band (low / med / high) or mark [TODO]. Do NOT fabricate precise figures. - **Win probability:** a confidence band (High / Med / Low) from signal strength, fit, and renewal timing (a near renewal raises it). - **Opener:** a one-line, value-framed talking point for the CSM. Do NOT cite internal metrics or reveal we monitor usage, and do NOT draft a full email — one line only. ## 4. Rank by expected expansion value Rank by expected added MRR = estimated additional MRR × win probability (highest first). This prioritises the biggest incremental opportunity weighted by likelihood — NOT current account size. State the ranking basis in one line. ## 5. Output (no email drafts) - **Slack #cs-expansion:** a ranked Markdown table — Account | Plan | Renewal | Angle | Est. +MRR | Win prob. | Expected +MRR | Opener. Header: date + number of candidates. Link the Drive file. (If the row gets too wide, keep the Signal/rationale in the Drive file only.) - **Google Drive:** save as "expansion-YYYY-MM-DD.md" in the "Customer Success" folder, including the Signal/justification per account. If a file with today's date exists, replace it. - If no accounts qualify, post "No expansion candidates this week" and skip the file. ## Rules - Rank by expected added MRR (potential × probability), NEVER by current MRR alone. - Every candidate grounded in a real trend/usage signal — never invent one. - Exclude at-risk/declining accounts (cross-check the same run). - No email drafts. Openers stay one-line and customer-safe; signals stay internal. - Keep MRR estimates and probabilities explicit and conservative — bands or [TODO] rather than fabricated precision.
Setup
Customerscore(or a connected sheet/doc with the same fields)
The trending-up cohort: health scores, usage trend, churn risk, plan and renewal data.
HubSpot(or Salesforce, any CRM)
Current plan, seat count, contract terms and renewal date.
The ranked expansion table posts to #cs-expansion.
Google Drive
A dated expansion file with the full signal rationale per account.
Don't use one of these? Swap in your equivalent or edit the prompt to drop that step. The logic still works.
A ranked Slack table (account, plan, renewal, angle, estimated +MRR, win probability, expected +MRR, opener) plus a dated expansion-YYYY-MM-DD.md saved to Drive.
Weekly, every Wednesday at 08:00.
Use cases
CSMs start Wednesday with a ranked list of growth conversations instead of finding expansion only at renewal.
Renewal proximity raises win probability in the ranking, so the near-term opportunities float up.
One opener line per account keeps the conversation helpful; nobody sends a cold upsell blast.
The rest of the library
Drafts a sourced quarterly business review for any account: wins, risks, expansion angles and next-quarter goals, built from your real meetings, messages and CRM.
02Posts your top 20 at-risk accounts every Tuesday ranked by MRR at risk, diagnoses the top three, and drafts check-in emails you approve before anything sends.
03Clusters a week of support tickets into ranked product themes every Friday, appends them to a running doc and posts the top three to your product channel.
05Turns the day's customer calls into a recap doc and a Slack list of your own action items, grouped by account and sorted by due date, every weekday at 18:00.
06Delivers 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.
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.