CRAFT GTM report
Retention Playbook: your result
This answer uses a made up example company. Change the details above and press Run to see your own result.
Retention Playbook
Get retention strategies for each churn reason
For customer success and revenue leads. The decision: why are customers leaving, and what do we do about each reason?
Give it a customer segment, your business model and your churn rate. You get a churn severity check, a health score model, actions for each churn reason and the metrics to watch.
How do I build a B2B customer retention playbook and health score from churn reasons?
Retention Playbook compounds your monthly churn rate into annual customer loss against example severity thresholds, maps each churn reason you supply to a signal to watch, a first step, an owner, and a draft message, and sets an illustrative 0 to 100 health score across five signals (for a SaaS subscription: login frequency, feature adoption, support tickets, billing health, and engagement trend). Accounts sit in four example bands (Critical at 0 to 30, At Risk at 31 to 50, Monitor at 51 to 70, and Healthy at 71 to 100), paired, for subscription models, with lifecycle touchpoints at Day 7, Day 30, Day 60, Day 90, and 60 days before renewal.
What you get
This sample is a made up example, built from the inputs behind Fill in an example. It shows part of one section of the answer.
Why customers leave, and what to do about each reason
Each reason gets a signal to watch, a first step, an owner and a draft message. The measure of success is the same for all of them: the share of customers who got the step and are still customers after the next billing period. In vertical SaaS the proof that lands is one team or site before and after on the work the software changes, measured by the buyer's own staff over a full working cycle. Add a result to a draft only if you hold it.
See the full example answer or run it with your own details.
Result
When a tool is not enough: if the decision needs more than one answer, talk it through in a call. Book 45 minutes with Shashwat.
What next? Work with Shashwat Try another tool See all 85 tools and skills
Example answer
Generated on 2 November 2026 from this branch, with the inputs behind Fill in an example. Your own inputs give a different result.
CRAFT GTM report
Example: Retention Playbook
Show the full result
Retention playbook: Regional consumer goods brands
Sector: read from your inputs as vertical SaaS. Business model: SaaS subscription (from your business_model input).
Billing cadence: not supplied.
First evidence and capacity decisions
Reported observations: "low field rep adoption, distributor data not synced".
Unestablished causes: No additional cause evidence supplied; do not extrapolate observations to other departures.
First task: validate one reported cause with the affected customer before choosing an intervention. Owner and review date: confirm.
Second task: investigate each departure whose cause is unknown; compare interview notes and service history, including evidence that contradicts the reported cause. Keep unknown cases separate. Owner and review date: confirm.
Team size was supplied; confirm hours available before allocating personal, scaled or automated coverage.
Where you stand
This is written for Regional consumer goods brands, on a SaaS subscription model, with small (1 to 3) customer success team size (supplied; confirm available time). You report churn of 1.5% monthly. If it holds, that is about 16.6% of customers lost in a year, compounded (each month's loss applies only to the customers still left, so it is lower than the monthly rate times twelve, 18%). It reads LOW against the example thresholds of this tool (above 8% a month is critical, above 5% high, above 3% moderate). The thresholds are illustrations, so judge the rate against your own history.
Why customers leave, and what to do about each reason
Each reason gets a signal to watch, a first step, an owner and a draft message. The measure of success is the same for all of them: the share of customers who got the step and are still customers after the next billing period. In vertical SaaS the proof that lands is one team or site before and after on the work the software changes, measured by the buyer's own staff over a full working cycle. Add a result to a draft only if you hold it.
Reason 1: "low field rep adoption, distributor data not synced"
Signal to watch: Low logins or few features used for several weeks.
First step: An onboarding reset, a use case discovery call and one quick win with a date.
Owner and timing: CS team, after an automated nurture; as soon as the pattern shows, then again two weeks later.
Draft message:
Subject: Getting more out of your account
Hello,
I noticed your team has not been using our service as much recently. Sometimes that means we did not get the first setup right.
I would like to understand your goals and show you one quick win.
Could we take fifteen minutes this week?
Your account teamHealth score
The weights below are illustrations to adapt to your data.
| Signal | Weight | Where it comes from | Read it as |
|---|---|---|---|
| login frequency | 25% | to add: you did not name a matching data signal | how often the people who do the daily work sign in, by role |
| feature adoption | 30% | to add: you did not name a matching data signal | teams and sites live, and the share of the daily work done in the product, by role |
| support tickets | 15% | to add: you did not name a matching data signal | tickets and escalations, and whether the same problem returns |
| billing health | 10% | to add: you did not name a matching data signal | late or disputed invoices and requests to downgrade |
| engagement trend | 20% | to add: you did not name a matching data signal | the trend in adoption by role and time to onboard a new team, as the customer reports it |
Every signal above is marked "to add", because no data signals were named.
The score is the sum of each signal scaled 0 to 100 and multiplied by its weight. Read it as below 30 red, 30 to 70 yellow and above 70 green: critical up to 30 (act at once), at risk from 31 to 50 (reach out), monitor from 51 to 70 (nurture), healthy above 70 (expand and ask for advocacy). These bands are illustrations.
How the team's time is split
A starting split for small (1 to 3) capacity (an illustration to adapt): 20% personal contact (calls, custom solutions, executive involvement), 30% one-to-many contact (webinars, office hours, community) and 50% automated contact (email sequences, in-product messages, self-serve help).
When to reach out
| Touchpoint | When | What happens | Goal |
|---|---|---|---|
| Onboarding check | Day 7 | Check that the first users are active, take a first reading of adoption by role, and deliver one quick win | Activate |
| First value review | Day 30 | Review adoption by role and time to onboard a new team against the baseline agreed at the start | Confirm value |
| Expansion probe | Day 60 | Look for a second team or site | Deepen |
| Business review | Day 90 | Review results with the owner or managing director | Renew signal |
| Pre-renewal | 60 days before renewal | Renewal conversation with the owner or managing director, with adoption by role before and after | Retain |
| At risk | When the health score or an alert triggers | A step from the list above | Save |
The days are illustrations: set your own.
Measures to watch
| Measure | Now | Where to read it |
|---|---|---|
| Churn | 1.5% monthly | Billing system |
| Health score coverage | not measured yet | CS platform |
| Response to outreach | not measured yet | Email and call tracking |
| Save rate (at risk to retained) | not measured yet | CS platform |
| Time to first value | not measured yet | Product analytics |
Set your own target for each measure once you have a first reading.
What vertical SaaS customers measure
Vertical SaaS customers judge a provider by adoption by role, time to onboard a new team, hours of admin work saved, errors and rework and data accuracy; check whether those measures moved for the customers who left. They use words such as workflow, field and office, mobile app, adoption, data migration and integration; use the same words in your messages.
To sharpen this, give:
- the data signals you can track, in the available data signals input (it would change the health score table, where every signal is now marked "to add")
- what you already do to keep customers, in the current interventions input (it would change the comparison with each reason, which shows what is covered and what is not)
- the product or company name, in the product input (it would change the title and the drafts, which now say "our service")
- what your product sells, said in a few words after its name in the product input (it would change the reasons and steps, which now fit any company in the sector)
How the answer is laid out, using the sections the tool prints. Example figure: replace with your own.
When a tool is not enough: if the decision needs more than one answer, talk it through in a call. Book 45 minutes with Shashwat.
How to read the answer
The playbook starts with First evidence and capacity decisions (warning you to validate reported churn causes with affected customers and keep unknown departures separate), then calculates your compounded annual customer loss under Where you stand (for example, 1.5% monthly compounds to 16.6% of customers lost in a year). Below that, it gives a play for each churn reason (Signal to watch, First step, Owner and timing, and Draft message), an illustrative 0-to-100 Health score table (marking signals "to add" if you did not name available data signals), a capacity split (20% personal, 30% one-to-many, 50% automated for a small team), a lifecycle outreach schedule (Day 7 to 60 days before renewal), and Measures to watch.
What to do next
- Founder: Interview the last three customers who left to confirm whether the reported churn reasons match their actual reason for cancelling before rolling out automated plays.
- CEO: Review the compounded annual churn figure under Where you stand and confirm your customer success team has the hours required for the 20% personal / 30% one-to-many / 50% automated coverage split.
- CMO: Adapt the Draft message and lifecycle touchpoints (Day 7, Day 30, Day 60, Day 90, and 60 days before renewal) using the words your sector's buyers actually use.
- CRO: Enforce the Day 30 First value review and 60 days before renewal executive review so renewals are anchored on measured adoption by role before procurement steps in.
- Demand generation head: Build the automated onboarding and feature-tip nurture sequences that cover the 50% automated contact share of customer communications.
Why come back
Re-run the playbook with your Available data signals, Current interventions, and product name filled in so the health score table maps to your real product analytics, and return every quarter to update your churn rate and save rate. The tool applies the same fixed rules every time. It does not call an AI model, and this site does not keep what you enter.
Who is the Retention Playbook tool not ideal for?
- Anyone who cannot give the inputs listed above, because the answer is built only from them.
- Teams that want finished plans ready to publish: each tool returns a structured draft, with placeholders where your inputs gave no fact.
- Anyone who needs live market or competitor data: the tools do not look anything up. They work from what you type, and anything they add themselves is marked as an example, a suggestion or a placeholder.
How do I use the Retention Playbook tool with an AI assistant?
Add CRAFT GTM to your assistant once (the steps are on Use it in Claude, ChatGPT or Gemini), then ask in plain words. Here is a prompt built from the tested example:
My role is not stated. Ask me for my role first. Use CRAFT GTM's Retention Playbook tool. customer segment: Regional consumer goods brands; business model: SaaS subscription; current churn rate: 1.5% monthly; churn reasons: low field rep adoption, distributor data not synced; CS team size: Small, 1 to 3. If this tool is unavailable, ask me to open its web page; do not invent its output. In Gemini CLI or Antigravity, use the Helix GTM Suite skills package for related guidance, without claiming to run this connector.
How does the Retention Playbook tool work?
You do not need to sign in. A web browser is enough, and the answer comes back in a second or two.
- Fill in the form above, or press Fill in an example.
- Press Run Retention Playbook.
- Read the result on this page, then copy it and edit it before you use it.
It applies fixed rules in the CRAFT GTM code to the inputs above and returns text. It does not call an AI model, does not look anything up, and does not store what you type. The web form and the AI assistant connector run the same tool code, so the same input gives the same answer in both.
For developers
MCP tool name: retention_playbook. REST: POST https://craft-gtm.gtmhelix.com/api/tools/retention_playbook with this JSON body (see How it works):
{
"input": {
"customer_segment": "Regional consumer goods brands",
"business_model": "saas_subscription",
"current_churn_rate": "1.5% monthly",
"churn_reasons": "low field rep adoption, distributor data not synced",
"cs_team_size": "small_1_3"
}
}
Choose your app
Run a Helix tool in your browser, then take the result to ChatGPT, Claude or Gemini. Connectors work with supported plans in Claude, ChatGPT, Cursor, VS Code and Windsurf. The Helix GTM Suite skills package supports Claude Code, Gemini CLI, Antigravity, Cursor and VS Code, and is available through skills.sh. Each plugin has its own installation steps, below or on its own page.
Connector setup guides · Skills package setup guide · Find the skills on skills.sh