The droid that runs renewals & churn watch

Renewals that never sneak up.

A droid that watches every account for the signals that precede a churn: slipping usage, a renewal date creeping up, a support thread left hanging. It scores the risk, drafts the right outreach, and hands a CSM the red accounts with the full story — so renewals are worked weeks early, not the week they lapse.

Does: Scores churn risk, works renewals
Where: B2B SaaS, ~10-person customer team
Reaches people on: Slack, email
Works inside: HubSpot, Stripe
Runs: Daily + ahead of every renewal

01The situation

The accounts most likely to churn are the ones you never hear from.

Renewals had a way of arriving as a surprise. An account would go quiet, usage would slide, and the team would notice the week the contract was up — far too late to do anything but scramble. The accounts most at risk were the ones nobody had heard from, which is exactly why they were invisible.

Health was reviewed once a quarter, in a spreadsheet, by which point the story was already written. What the team needed was a continuous early-warning system — something watching every account every day for the signals that come weeks before a churn, not after.

02How it works

How the droid took it on.

Rather than review health once a quarter, the team handed the watch over. Every morning the droid scores every account, works the amber ones and the upcoming renewals, and hands a CSM the red accounts the moment they turn.

TASK#274Renewal & churn watchstanding

trigger: Every morning at 7:00, a full account health scan
also: Each renewal at 60 days out, plus sudden usage drops
scope: Every active account · usage, billing, activity
runs as: A contained droid action per scan, renewal, or signal
memory: A health history per account, so it sees trends not snapshots

Set up once, in plain language: “watch every account for churn signals, start renewals early, and pull a CSM in the moment one turns red.” The droid turned that into a standing job, the early-warning system a small customer team can't run by hand.

FIG. 1The durable task, exactly as it exists in the runtime.

Every morning trips the same loop:

01Scan fires7:00am, daily
02Droid wakesJust for the scan
03Reads usage + CRM + memoryBilling, activity, renewal dates
04Scores every accountGreen / amber / red
05Works the riskDrafts outreach, preps renewals
06How risky?Amber → draft a check-in · Renewal due → start early · Red → hand to a CSM
07Posts to #customerRenewal & risk digest

FIG. 2The loop each run walks.

03Ongoing handling

How it ran, account after account.

Here’s a day of account-watching as it actually unfolded, down to how it reached for each app. Only the one red account ever needed a CSM.

Isometric customer-health scene: account cards with health gauges, a renewal calendar, and a contract with a retention arrow

See it in action

One day, account by account.

The morning scan, the upcoming renewals, and a CSM's question land on the left. Watch the droid pick up each one and work it across billing and the CRM, handing off to a human only for the account that's genuinely at risk.

The result

What changed, by the numbers.

Weeks of warningrisk surfaced while it's still fixable

Renewals earlyworked 60 days out, not the week of

~1 in 9accounts that needed a CSM, with the rest watched quietly

Signals, not vibesusage, tickets and sentiment checked per account

An illustrative workflow built from real product mechanics. The connected apps, the schedules and the run handling are how a droid actually works; the companies, the people and the figures are composites rather than a customer's numbers.

Try this workflow

Run this workflow yourself.

Copy the brief below and paste it into Unify. It’ll walk you through the prerequisites, connect what it needs, and stand the workflow up with you.

Workflow brief

I run customer success at a B2B SaaS company, and renewals keep arriving as a surprise. An account goes quiet, usage slides, and we notice the week the contract's up, which is too late for anything but a scramble. The accounts most at risk are the ones we haven't heard from, which is exactly why they're invisible. We review health once a quarter in a spreadsheet, by which point the story's already written.

Be our churn early-warning system. Apps/channels: HubSpot (accounts, renewal dates, activity, owners), Stripe (usage and billing signals), Gmail (renewal and check-in outreach), Slack #customer (CSM hand-offs and the daily digest).

Run a full account-health scan every morning, start every renewal 60 days out, and react to sudden usage drops. On each scan:
1. Pull usage and billing for every account and cross it with CRM activity, then score each green / amber / red.
2. For amber accounts, draft a low-key check-in for the owner to send.
3. For renewals at 60 days, prep a value recap and draft the renewal opener.
4. Post a renewal-and-risk digest to #customer.

When a CSM asks, give a pre-call health snapshot with talking points. Use judgment on escalation: handle amber and routine renewals yourself, but when an account turns red, whether that's a champion gone or usage flat near renewal, hand it to its CSM with the full history and a suggested save play; don't try to save it with an automated email. Keep a health history per account so you see trends, not snapshots.

What would you take off your desk?

Tell us the job that never gets done before close. We'll wire it up on a call and you can watch it work.

Try now, free Book a walkthrough