The truth about churn
By the time a customer cancels, the decision was made weeks ago. Sometimes months.
The signals were there. A support ticket went unresolved. A key user stopped logging in. A billing issue created friction. A champion left the company.
Your tools captured every one of these moments. Your CRM logged the contact change. Your product analytics tracked the usage drop. Your support platform recorded the escalation. Your billing system flagged the failed payment.
But no one connected the dots.
This isn't a people problem. It's a systems problem. The data exists, but it lives in silos. Each team sees their slice. No one sees the full picture until the cancellation email arrives.
Why lagging metrics fail you
Most companies track churn with lagging metrics: churn rate, net revenue retention, renewal rate. These tell you what already happened. They're the scoreboard, not the game.
Leading indicators are different. They measure behavior before the decision is made. They give you time to act. They turn churn from an outcome you report into a risk you manage.
The difference between a company that loses 15% of revenue annually and one that loses 8% isn't better customer success reps. It's better signals.
The 7 leading indicators
1. Product usage decline
What it looks like: Login frequency drops. Session duration shrinks. Key features go unused. The account that used your product daily is now logging in twice a week.
Why it matters: Usage is the most direct measure of value delivery. When usage drops, the customer is finding less value — or finding it elsewhere.
Where the data lives: Amplitude, Mixpanel, Pendo, Heap, or your own product analytics.
The nuance: Don't just measure logins. Measure meaningful usage — the actions that correlate with the value your product delivers. A customer who logs in daily but only checks their dashboard is different from one who actively uses core workflows.
2. Champion departure
What it looks like: Your main contact leaves the company. The person who championed the purchase, who drove adoption, who defended your budget line — gone.
Why it matters: In B2B SaaS, relationships drive retention. When your champion leaves, you have approximately 90 days to build a relationship with their replacement. If you don't, that replacement will evaluate alternatives with fresh eyes and no loyalty.
Where the data lives: LinkedIn notifications, CRM contact updates, email bounces, calendar patterns.
3. Support ticket patterns
What it looks like: A spike in support tickets. Repeated issues that don't get resolved. Escalation requests. Frustrated tone in communications.
Why it matters: Support tickets are a direct signal of friction. But it's not just volume — it's the pattern. Unresolved tickets, repeated issues on the same topic, and sentiment shifts matter more than raw ticket count.
Where the data lives: Zendesk, Intercom, Freshdesk, or your support platform.
4. Engagement with communications
What it looks like: They stopped opening your emails. They don't respond to CSM check-ins. They skip QBRs. They ghost your renewal conversation.
Why it matters: Silence is a signal. When a customer disengages from communication, they've mentally downgraded your priority. They're not angry — they're indifferent. And indifference is harder to recover from than frustration.
Where the data lives: Outreach, Gainsight, your email platform, calendar tools.
The nuance: It's about change, not absolute levels. A customer who never opened emails isn't necessarily at risk. A customer who used to open every email and suddenly stopped? That's a signal.
5. Billing friction
What it looks like: Failed payments. Disputes. Requests to pause. Downgrade inquiries. Questions about cancellation terms.
Why it matters: Money follows priority. When a customer lets payments fail or actively questions their spend, they're signaling that your product isn't worth the friction of keeping it running. The pattern matters — one failed payment is noise; repeated billing issues are a signal.
Where the data lives: Stripe, Chargebee, Recurly, your billing system.
6. Feature adoption gaps
What it looks like: The customer bought your product for a specific use case and never implemented it. They're using 20% of what they're paying for. Key features that drive stickiness remain untouched.
Why it matters: Customers who use more of your product are harder to replace. Customers who use one feature are one competitor away from churning. The gap between what they're paying for and what they're using is a measure of risk.
Where the data lives: Product analytics, feature flags, onboarding completion data.
The nuance: Identify the 3-5 features in your product that most correlate with retention. Track adoption of those specific features, not everything.
7. Contract and timing signals
What it looks like: Short-term renewals instead of annual. Annual contracts switching to monthly. Asking about cancellation terms mid-contract. Procurement asking for updated pricing.
Why it matters: Contract decisions are forward-looking. When a customer shortens their commitment, they're hedging. When procurement gets involved outside of renewal, something is shifting internally.
Where the data lives: CRM, contract management tools, billing system.
Why these signals stay hidden
The challenge isn't that these signals don't exist. It's that they live in different systems, owned by different teams, with different update cadences.
- Product usage lives in your analytics platform
- Champion changes live in your CRM (maybe)
- Support patterns live in Zendesk
- Billing signals live in Stripe
- Engagement data lives in Outreach or Gainsight
Each team sees their slice. Product knows usage is down. Support knows tickets are up. Sales knows the champion left. But no one sees all three happening to the same account at the same time.
Churn isn't a single-signal event. It's a multi-signal pattern. And you can only see patterns when you connect the data.
This is why churn feels unpredictable. It's not. It's a coordination failure.
From signals to system
The companies that beat churn don't just track these signals — they build systems around them.
- Connect signals across tools into a unified view per account
- Score accounts by composite risk, weighted by signal importance
- Route alerts to the right team at the right time
- Act before renewal with playbooks matched to risk type
This isn't about adding another dashboard. It's about building an early warning system that turns scattered data into coordinated action.
Eru connects churn signals across your CRM, product analytics, support platform, and billing system — so you can see risk before it becomes revenue loss.
Book a churn audit →