agentsclimarketplace

Churn save plan

Skill quivly/skills/customer-engineering/churn-save-plan

Production-ready Agent Skills for Customer Engineering, Post-Sales, and Customer Success teams. Quivly Skills is a curated open-source collection of reusable skills that give AI agents deep expertise in customer engineering workflows. Every skill follows the official Agent Skills specification.

Install
npx -y skills add quivly/skills --skill churn-save-plan

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

  • 4 stars4 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

Deep-dives a red account and produces a structured save plan — root causes ranked, save plays matched to causes, and a realistic assessment of whether the account is savable. Use when an account turns red, gives churn signals, or someone says "we're going to lose X".

The file declares its own license as MIT. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.

SKILL.md

3.6 KB, as published. Nobody here has run it

Churn Save Plan

You build the save plan for an account in real danger — honest about root causes, honest about odds, specific about plays.

Core principle: no play without a named root cause — discounts don't fix value gaps, and generic plays waste the runway.

Workflow

  1. Establish the full risk picture: health components (get-health-score), active signals (get-insights), ticket history (search-tickets), usage trend (get-usage), recent call tone (search-calls), relationship map (get-contacts).
  2. Identify root causes — usually 1-2 dominate. Rank them:
    • Value gap — never reached the outcome they bought for
    • Product gap — a needed capability doesn't exist or underperforms
    • Trust damage — support failures, missed commitments, outage scars
    • Relationship loss — champion gone, sponsor disengaged, single thread cut
    • External — budget cuts, M&A, strategy shift (least addressable by us)
  3. Assess savability honestly: is there time before the notice deadline, an engaged human on their side, and a cause we can actually fix?
  4. Build the plan: each play matched to a root cause, with owner, deadline, and a "proof point" — the observable event that tells us it's working.

Output Format

Save plan: {Customer} — ARR, renewal/notice date, savability: Good odds / Uphill / Long shot

Root causes — ranked, each with evidence

The plan — table: play, targets which cause, owner, deadline, proof point

Ask list — internal help needed: exec sponsor, product commitment, commercial flexibility

Tripwire — the date/event at which we stop investing in the save and shift to graceful-exit + win-back planning

Stop conditions

  • Two re-engagement attempts with no response → stop sending. The next move is an exec-to-exec touch or the tripwire decision, not a third email.
  • The tripwire date or event passes → stop the save. Shift to graceful exit and win-back planning.
  • No engaged customer contact AND no exec path in → the plan is not viable as written; say so and escalate the relationship question before proposing plays.

Self-check

  • The plan has five plays → you don't know the root cause yet. Go back to diagnosis.
  • A discount appears before a named cause → you're buying time, not saving the account.

Guidelines

  • One play per root cause beats five generic plays. Discounts don't fix value gaps.
  • A save plan without an engaged customer contact is a fantasy — if no one on their side will meet, that IS the first play.
  • Set the tripwire on day one. Zombie save plans consume the team's best hours.
  • "Long shot" is a legitimate verdict; say it and size the effort accordingly.

Related skills: diagnosis inputs → health-drop-diagnosis, usage-drop-investigation, ticket-pulse; exec entering the account → escalation-brief.

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.