Your churn rate went up this quarter. Before you panic, work through the diagnostic. Most churn increases have one of 8 specific causes, and the right fix depends entirely on which one is hitting you.
Panic is usually worse than the actual problem.
The 8 most common causes
1. Acquisition mix shifted
You added a new marketing channel or paid campaign that is bringing in wrong-fit customers. They convert (which looks like growth) but churn fast (which shows up 60-90 days later).
Diagnostic signal: Recent cohorts have worse retention than older ones. Segment by acquisition channel and see if one channel is worse.
Fix: Reduce spend on the underperforming channel. Tighten targeting. Or accept the higher volume with lower retention if unit economics still work.
2. Older cohorts naturally maturing
Every SaaS has a natural retention curve. Customers you acquired 2-3 years ago are reaching the point where their natural churn kicks in. This can raise blended churn without indicating any new problem.
Diagnostic signal: Newer cohorts are retaining fine; older cohorts are the ones churning. Cohort retention chart makes this obvious.
Fix: Nothing tactical, but plan for the revenue impact. Consider expansion motion to offset the base decay. See NRR guide.
3. Recent price change fallout
You raised prices, changed pricing structure, or removed a plan. Some customers who tolerate the current price will churn at the new one.
Diagnostic signal: Churn spike happens 30-90 days after a pricing change, concentrated in the affected plans.
Fix: Grandfather at-risk accounts, offer transition discounts, or accept the churn if it was expected and the unit economics still work.
4. Competitor launched or improved
A competitor released a new product, cut prices, or ran a big campaign. Some of your customers are evaluating alternatives now.
Diagnostic signal: Cancellation reasons mention the competitor by name. Cancellation cluster is spread across plans and cohorts.
Fix: Personalized competitive comparisons in the save flow. Show what customers would lose by switching. See competitive displacement.
5. Recent product change broke a workflow
You shipped a product update that changed how customers use a core feature. Power users who depended on the old workflow are churning.
Diagnostic signal: Support ticket volume spiked around the same time. Churn is concentrated in high-usage accounts.
Fix: Rollback or migrate the affected workflow. Communicate proactively to affected accounts. In-product help for the new pattern.
6. Seasonal effect
Certain SaaS products have predictable seasonal churn (tax software after April, education tools in summer, retail tools in Q1). If you did not adjust for seasonality, you might mistake it for a real problem.
Diagnostic signal: Compare to the same quarter last year. If churn is similar year-over-year, you are seeing seasonality, not a new trend.
Fix: Nothing new. Just report seasonally-adjusted metrics so you do not spook the board.
7. Measurement methodology changed
Someone changed how you count churn: switching from logo to revenue, or including previously-excluded free-to-paid transitions, or fixing a bug that was hiding failed payments.
Diagnostic signal: The "increase" happened suddenly on a specific date correlated with the measurement change, not gradually.
Fix: Confirm the new measurement is correct. Restate historical numbers using the new methodology. Communicate the change to stakeholders.
8. Genuine product-value decline
The rare but real case: your product is no longer delivering the value it used to. Customers are quietly disengaging. Cohort curves are declining across the board.
Diagnostic signal: Newer cohorts retain worse than older ones. Support and NPS trends are also declining. The problem is everywhere, not concentrated.
Fix: Product work. This is the hardest problem on the list and takes 3-6 months of investigation and iteration.
The diagnostic order
- Check if involuntary churn spiked. Pull failed payment data. If involuntary is 25%+ of the churn increase, fix it with AI dunning first. Quick win.
- Compare to same quarter last year. Rules out seasonality.
- Segment by acquisition cohort. Rules out mix shift.
- Segment by plan/tier. Rules out price/product change.
- Check cancellation reasons. Surfaces competitor or workflow issues.
- Compare cohort retention shapes. If newer cohorts are worse, look at product-value decline.
What NOT to do
- Do not lower prices in response to churn without diagnosis. You will attract more churn-prone customers.
- Do not fire your CS lead. One quarter is not enough time to attribute cause.
- Do not launch a big marketing campaign to "get ahead of it." That treats the symptom (revenue) not the cause.
- Do not panic on one data point. Look at trend across 3-6 months before drawing conclusions.
Score your setup
Take the 60-second Churn Health Check. It scores your measurement and diagnostic setup and tells you whether your data is even accurate enough to trust the trend.