Spot at-risk customers early and get the next best action for each
Starts every week or when a CRM export arrives. Ends with a health tier, the revenue at risk, a play with steps and a customer email draft.
Bring the step you are stuck on: 30 minutes, a straight answer and the next step. If you would rather have it built, a reply within 5 working days.
The route
7 steps: 2 rule steps, 2 AI steps and 1 person step
Copy, share, discuss
- Copy
Available once this workflow passes its checks.
- Share
todkett.com/automations/customer-support/customer-health-score/
- Discuss
Opens when the first workflow passes its checks. No GitHub account? Ask about this by form.
How the library, the tracks and the talk work: the community page.
At a glance
- Starts when
- A weekly schedule, or a CRM or analytics export posted to one webhook.
- Input
- Account data from a CRM or analytics export.
- Output
- A health tier, the revenue at risk, a play with steps and a customer email draft.
- AI
- Uses AIReading notes, the small capped adjustment, the play and the email draft.
- Rule checks
- 5 checks around the AI
- Tools
- Webhook from a CRM or analytics export, OpenRouter, Google Sheets
- Last stress test
- No recorded run yet
01Steps
The steps
7 steps: 2 rule steps, 2 AI steps and 1 person step. Your first renewals are coming and nobody owns customer success. Usage, tickets and notes live in different tools, so churn shows up at renewal time.
01Trigger
Account export
Account data from a CRM or analytics export.
02Rule step
Fixed health score
Score health out of 100 with fixed weights: seats, usage trend, support, NPS, champion, last review meeting and renewal runway.
03AI step
AI reads the notes
AI reads the account manager's notes for risks and expansion signals and suggests an adjustment, with a reason.
04Rule step
Cap and overrides
Code caps the adjustment at 15 points either way and applies the overrides for risky renewals.
05AI step
AI writes the play
AI writes a play with steps and a customer email draft.
06Person step
Account manager reviews
The account manager reviews the play and sends the email draft.
07Output
Tier, play and draft
Tier, revenue at risk, the play and the email draft.
Account manager
02Where AI is used
Where AI is used, and what stays with you
AI reads, judges or drafts. Fixed rules check, decide and route. A person approves anything that goes out.
AI step
Steps 03, 05Reading notes, the small capped adjustment, the play and the email draft.
Notes hold signals the numbers miss. The score itself is fixed arithmetic.
What the model reads
- The account manager's notes, for risks and expansion signals.
What it never sees or decides
- The health score: fixed arithmetic from seats, usage, support, NPS, champion, reviews and renewal runway.
- More than 15 points either way: code caps any adjustment.
- A risky renewal shown as healthy: overrides in code prevent it.
- The customer's inbox: the account manager sends the email.
Model calls: Not counted from the workflow yet. It is counted before this page goes live.
Fixed rules
They check, decide and route. 2 rule steps.
- 02Fixed health score
- 04Cap and overrides
A person
- 06The account manager reviews the play and sends the email draft.
03Checks
Checks around the AI
What runs before the model, and what checks its answer after, in plain words. Build these into your version too.
- 01The score is fixed arithmetic.
- 02The AI adjustment is capped at 15 points either way.
- 03An open P1 within 60 days of renewal, or a champion gone within 90 days, can never show as healthy.
- 04Expansion is flagged by rule: 90% or more of seats used and usage growing.
- 05At most 100 accounts per run.
04The 8 checks
Stress test
The 8 checks we run on every workflow, and you can run too. A check marked Not needed says why.
- Status
- Not yet0 of 7 that apply have a recorded pass
- Last run
- No recorded run yet
- Model review
- Still open
Checks marked Not yet are still open. 2 of them were seen in the workflow design on 23 Oct 2026; a check shows Passed only once its test run is recorded.
ST7Human sign-off
Status: Not yet
Seen in the workflow design on 23 Oct 2026. The recorded test run is still open.
ST8Rules around the AI
Status: Not yet
Seen in the workflow design on 23 Oct 2026. The recorded test run is still open.
0 passed · 1 not needed · 7 not yet
What we testWe show what we tested, never the test data. The example below is made up.
05Test replay
How to test it
Before you trust a workflow, try to break it. The sample record runs the route step by step. Each break test below gets its own replay once its recorded run exists.
- Record
- ACCT-0074
- Data
- Sample run, made-up data
01Account export
Fernway account: 90% of seats used, usage growing
Done02Fixed health score
Score computed with fixed weights
Done03AI reads the notes
Notes read; an adjustment suggested, with a reason
Done04Cap and overrides
Adjustment capped; open P1 within 60 days of renewal: not healthy
Done05AI writes the play
Play with steps and a customer email drafted
Done06Account manager reviews
Held for the account manager
Waiting for a person07Tier, play and draft
Tier, revenue at risk, play and draft; expansion flagged by rule
Account managerArrived
ACCT-0074 reached the last station: Account manager. Tier, revenue at risk, play and draft; expansion flagged by rule.
Sample run on made-up data. No stress-test run is recorded yet, so this plays the example, not a test. Nothing runs live.
The break tests for this workflow
- ST3Not needed
Same input sent twice
Returns a result only: writes and sends nothing.
- ST2Not yet
Instructions hidden in the input
No recorded run to replay yet.
- ST4Not yet
A connected service fails
No recorded run to replay yet.
- ST8Not yet
The AI returns a broken answer
Seen in the workflow design. No recorded run to replay yet.
06Example
The example, in and out
Made-up data, written to show the steps; no recorded run stands behind it yet. Fernway is a made-up small software company, and its address ends in .example, so it cannot belong to a real business.
InputExample
Example on made-up data: a Fernway customer account with 90% of seats used, usage growing, an open P1 ticket and renewal in 45 days.
OutputExample
Fixed health score computed. Any AI adjustment capped at 15 points. Override applied: an open P1 within 60 days of renewal, so the account cannot show as healthy. Expansion flagged by rule. A play and a customer email draft for the account manager.
07Limits
When not to automate this
- It does not predict churn as a fact. The score is fixed arithmetic plus a small capped adjustment.
- It needs usage, ticket and renewal data together in one export.
08What you need
What you need to run it
| Tool | Access |
|---|---|
| Your CRM or analytics tool | Post an account export to one webhook, or allow a weekly read |
| OpenRouter | One API key with a credit limit, for the notes and drafting steps |
| Google Sheets | Optional: write a test log |
- Where it runs
- In n8n, self-hosted or n8n cloud.
- Model calls
- Not counted from the workflow yet. It is counted before this page goes live.
- Running costs
- Yours, paid to each service directly. Scope in writing first. A quote before any build. No prices on the site yet.
How to adapt it
Changes we have made or tested on this workflow.
- 01Run it weekly on a schedule, or post a CRM export to its webhook.
What changed
- ContentFirst version, written from the 23 Oct 2026 catalog.
09FAQ
Questions people ask about it
Does AI decide the health score?
No. The score is fixed arithmetic out of 100 from seats, usage trend, support, NPS, champion, the last review meeting and renewal runway. AI may move it by at most 15 points, with a reason.
Can a risky renewal show as healthy?
No. An open P1 within 60 days of renewal, or a champion gone within 90 days, can never show as healthy.
Does it email customers?
No. It drafts the customer email for the account manager to review and send.
10Related
Where to go next
Use cases
6 steps
Operations
Turn raw weekly exports into a clean KPI workbook with fact-checked insightsAI in one step only: writing the weekly notes
7 steps
Can you automate it?
Stuck on your version of this workflow?
Bring the step that does not work. In 30 minutes you get a straight answer and the next step. If you would rather have it built, tell us your tools and the rules you already follow. Scope in writing first. A quote before any build. No prices on the site yet.