Customer retention
Keep customers from churning — before they decide to leave
Churn is cheaper to prevent than to win back. Five free Agent Skills — churn analysis, QBRs, onboarding, returns prevention, and support resolution — form a retention stack that catches at-risk customers early. Learn the frameworks and run them in Claude.
Why is retention worth more than acquisition?
Every business obsesses over acquiring customers; far fewer systematically keep them. That’s backwards. Acquiring a new customer costs several times more than retaining an existing one, existing customers spend more over time, and — crucially — churn compounds against you: every point of monthly churn is a leak that grows as you scale. A retention program is often the cheapest growth you can buy.
The reason teams under-invest is that retention feels reactive — you find out a customer is leaving when they tell you. The shift these skills enable is proactive: watch health signals, intervene early, and design the onboarding, reviews, and support experiences that keep customers by default. Retention stops being a fire drill and becomes a system.
The frameworks behind it
The skill isn’t magic — it encodes proven frameworks anyone can learn. Here’s the method itself, so you understand what it’s doing for you.
The churn risk-signal taxonomy
churn-prevention-analystYou can’t save a customer you don’t know is leaving. The core of proactive retention is a taxonomy of early-warning signals across three types. Usage signals: a >30% usage drop, key-feature abandonment, or the admin not logging in for two weeks. Engagement signals: no response to your CSM for 21 days, a declined QBR, a falling CSAT trend. Business signals — the most severe: the champion left, layoffs announced, an acquisition in the news.
Each signal carries a risk level (medium → high → critical), and combined they produce a risk score you can act on. The point is to move retention from reactive firefighting — noticing when the cancellation email arrives — to proactive health management, where you intervene weeks earlier, when a save is still possible.
The return-reason taxonomy
returns-prevention-strategistFor e-commerce, returns are a retention problem in disguise — a bad return experience is a churned customer. The fix starts by classifying why returns happen: fit/size (30–40% of returns — the biggest bucket, solved with size guides), not as expected (20–30% — better images and descriptions), quality (10–15%), wrong item (fulfillment), and changed mind (impulse). Each reason has a prevention strategy — and the insight is that you cut returns with better information, not restrictive policies (which just cut sales too).
Benchmarks make it concrete: apparel runs 25–30% returns (excellent is under 15%), footwear 30–35% (under 20%), electronics 10–15% (under 5%). Knowing your category’s benchmark tells you how much headroom there is.
The skills that get you there
Use any one on its own, or chain them — each hands its output to the next. Every one is free in the public library.
- 1
Churn Prevention Analyst
churn-prevention-analystSpots at-risk customers from early-warning signals, then deploys the right save motion or win-back — the anchor of proactive retention.
- 2
Customer Onboarding Playbook
customer-onboarding-playbookGets new customers to value fast with milestones and health scoring — because most churn is decided in the first 90 days.
- 3
QBR Architect
qbr-architectBuilds quarterly business reviews that prove value and surface expansion — the relationship ritual that pre-empts churn at renewal.
- 4
Returns Prevention Strategist
returns-prevention-strategistThe e-commerce edge of retention — cuts returns (and the churn they cause) through better pre-purchase information, not restrictive policies.
- 5
Customer Inquiry Resolver
customer-inquiry-resolverResolves support inquiries with the right tone and authority, so a bad service moment doesn’t become a lost customer.
When you’d reach for it
Any time keeping customers matters more than it’s getting attention — which, for most subscription and e-commerce businesses, is always. Especially when churn is creeping up, onboarding is inconsistent, or renewals feel like a scramble.
- SaaS and subscription teams fighting monthly or annual churn
- Customer Success teams that react to cancellations instead of preventing them
- E-commerce brands losing margin (and customers) to high return rates
- Support teams whose service moments quietly cost renewals
What the retention stack does
It turns retention from something you react to into something you run. The stack catches at-risk customers early (churn-prevention-analyst), gets new ones to value before they can churn (customer-onboarding-playbook), proves value at the renewal ritual (qbr-architect), and — on the e-commerce side — cuts the returns and support failures that quietly lose customers (returns-prevention-strategist, customer-inquiry-resolver).
Retention is the highest-leverage growth lever most teams under-invest in: keeping a customer is far cheaper than acquiring one, and a small churn improvement compounds. These five free skills give you the frameworks to do it systematically instead of by gut feel.
Try asking it
Real prompts, and what the skill hands back.
You ask
“This account’s usage dropped 40% and they declined our last QBR. What’s the churn risk, and what save motion should we run?”
What comes back
A risk assessment flagging the high-risk usage and engagement signals, plus a recommended save motion and offer framework from the intervention playbook.
You ask
“Our apparel return rate is 32%. Break down the likely reasons and give me the highest-impact prevention moves.”
What comes back
A return-reason breakdown against the taxonomy and category benchmark, with prevention strategies ranked by impact (starting with sizing).
What makes the stack work
Early-warning signals
A risk-signal taxonomy catches churn weeks before the cancellation, while a save is still possible.
Value in the first 90 days
Milestone onboarding with health scoring — most churn is decided before month three.
Renewals as a ritual
QBRs prove value and surface expansion, so renewal isn’t a surprise.
Information beats policy
Returns fall when you inform buyers better, not when you make returns harder.
Every service moment counts
Resolving inquiries well keeps a bad day from becoming a lost customer.
How the five work together
They cover the customer lifecycle: onboard to value → monitor health and catch risk → review value at renewal → and, for commerce, prevent the returns and support failures that leak customers. Run the whole stack, or start with the one leak that’s costing you most.
Retention data also feeds the top of the funnel: what makes customers stay tells you who to acquire — pairs naturally with synthetic-persona research.
Who it’s for
- Customer Success leaders who want to prevent churn, not react to it
- E-commerce operators fighting return rates and repeat-buyer drop-off
- Founders who know retention is cheaper growth than acquisition
Do it yourself — build the retention stack
Run the skills in Claude and customize each with your product, signals, and thresholds:
Point Claude at the right skills
agentman_skillschurn-prevention-analystcustomer-onboarding-playbookqbr-architectreturns-prevention-strategistcustomer-inquiry-resolverAlready connected your library to Claude? Just tell it: “Use the churn-prevention-analyst skill (and the others below) from the agentman_skills MCP server.” Not connected yet? The Try in Claude / ChatGPT button above loads the first one from the public library — no setup. (connect your library.)
- 1
Load the skills
Click “Try in Claude / ChatGPT” above, or — if you’ve connected your library — tell Claude: “Use the
churn-prevention-analystskill from theagentman_skillsMCP server.” Add the others by slug as you go. - 2
Score your at-risk accounts
Run
churn-prevention-analystagainst your customer data — it maps the usage, engagement, and business signals into a risk score and recommends a save motion. - 3
Fix the first 90 days
Use
customer-onboarding-playbookto design milestone-based onboarding with health scoring, so new customers reach value before they can churn. - 4
Make renewals a ritual
Run
qbr-architectto build value-proving quarterly reviews that surface expansion and defuse renewal risk early. - 5
Cut returns and service churn
On the e-commerce side, run
returns-prevention-strategistto attack your biggest return reasons, andcustomer-inquiry-resolverso support resolves issues without losing the customer.
That’s a customer-success and CX function most teams staff up for — assembled from five free skills you keep and tune.
Try Churn Prevention Analyst right now
One click loads the skill into Claude or ChatGPT — no account, no setup.
FAQ