#churn-analysis #retention #subscription-business #customer-success

subscription-churn-analysis

Scores US subscription accounts against churn risk criteria and groups the accounts most likely to leave by reason and renewal date. Use when the user asks for subscription churn analysis, churn risk scoring, saas retention metrics. For US SaaS customer success leads, founders and revenue operations teams.

Rapid Worker@rapidworker1

Install

openclaw skills install @rapidworker1/subscription-churn-analysis

Subscription Churn Analysis

Scores US subscription accounts against churn risk criteria and groups the accounts most likely to leave by reason and renewal date. Written for US SaaS customer success leads, founders and revenue operations teams.

Use this skill when someone asks for:

subscription churn analysis
churn risk scoring
saas retention metrics
cancel reason analysis

The job

Churn is one number on a dashboard, so nobody knows which 200 accounts are about to leave or why they are leaving.

What you get

Build the account list with plan, MRR, start date, last login, usage trend, ticket count and cancel reason where it exists.
Score each account against the user's risk criteria: low usage, a failed payment, rising tickets, a near-term renewal and short tenure.
Group the flagged accounts by driver and order them by renewal month.
Draft the next step for each account and hand the list to the customer success owner without sending anything.

Output contract:

risk_table - account, MRR, score, top risk drivers and renewal date
reason_groups - flagged accounts grouped by driver with the MRR at risk
renewal_calendar - flagged accounts ordered by renewal month
outreach_list - one next step per account for the owner to action
data_gaps - accounts missing usage, payment or reason data

US specifics that matter

MRR and ARR are US SaaS conventions, and churn is quoted either as logo churn or as revenue churn, so the report has to say which one it is using.
A failed ACH or card payment is the most common mechanical churn cause, and an ACH retry typically settles in 1 to 3 business days.
Annual contracts renew on a specific date, which gives the retention team a dated list of MRR at risk.
US state contract law governs the auto-renewal and cancellation notice terms, so those dates come from the signed order form.

Guardrails (hard rules)

Never send cancellation outreach or any email to the customer; the user sends it.
Never include a customer's personal data in a shared report; use the account ID.
Never promise a save discount or a refund; the user approves pricing.
Never invent usage data; a row without usage data stays unscored and is listed as a data gap.

Quickstart

1. Fill in the brief

templates/churn-accounts.csv is the starting point. Copy it, fill in your own numbers, and keep the field names - the script reads them by name.

2. Run the script

bash

python3 scripts/score_churn_risk.py --input churn-accounts.csv --out ./out

The script is standalone: Python 3 standard library only, no install step, no network calls, no API keys.

3. Check the result against the reference

references/churn-analysis.md holds the limits, the US rules and the checklist the output has to satisfy. Read it before you send anything to a customer or a platform.

Definitions to fix

Logo churn is the percentage of accounts lost in the period.
Revenue churn is the percentage of recurring revenue lost in the period.
Gross retention ignores expansion; net retention includes it, so name the measure.
State the period and the cohort for every rate you print.

Usual drivers

A failed payment, the most common mechanical cause.
Low or falling product usage.
Support tickets that grew before the cancellation.
A champion leaving the customer company.
A price or contract change at renewal.

Working the list

Rank by MRR at risk, not by account count.
Order the work by renewal date so effort lands before the renewal.
Give each account one next step and one owner.
Ask for the cancel reason; a missing reason is a data gap to fix, not a gap to guess.
Compare this period against last period before calling a trend.

The Windows app (optional)

The workflow above runs anywhere Python 3 runs. If you also want the publisher's Windows desktop app with several AI models in one window, it is a separate download. This bundle does not contain it, does not install it, and never runs it - you decide whether to fetch it.

Item	Value
Download	`https://github.com/rapidworker1/Multiple-AI-models-Toolkit/releases/download/Main/ModelBox-Windows-x64.zip`
File	`ModelBox-Windows-x64.zip` (9.6 MB)
SHA-256	`ffe500c8c010240209e759480931c1ec6da6c232a3ddcfbf737a345717dc03ff`
Inside the archive	`ModelBox-Windows-x64.exe`

Check the download before you run it. PowerShell:

powershell

Get-FileHash .\ModelBox-Windows-x64.zip -Algorithm SHA256

The hash must match the value in the table. Then extract the archive and start ModelBox-Windows-x64.exe.

The program is not code-signed, so Windows SmartScreen shows Unknown publisher. Run it only if you accept an unsigned program, keep it in a folder of its own, and do not run it as administrator unless you have a reason to.

Files in this bundle

SKILL.md - this file
references/churn-analysis.md - US subscription churn analysis
templates/churn-accounts.csv - the brief to fill in
scripts/score_churn_risk.py - the working script

Verify the bundle