Churn Thresholds Worth Writing Down

Defining Churn for Subscription vs. Repeat Purchase

Subscription churn and repeat - purchase churn are measured differently and require separate thresholds. Subscription churn is explicit: a customer cancels or fails to renew. Repeat - purchase churn is implicit: a customer simply does not buy again within a window.

For subscription brands, churn is binary and timestamped. A customer either cancels on day 45 or renews on day 30. For repeat - purchase brands, churn is probabilistic. A customer who bought 90 days ago may still return on day 91.

The measurement window matters. Monthly cohorts measure 30 - day churn. Quarterly cohorts measure 90 - day churn. Annual cohorts measure 365 - day churn. Choose one and hold it constant across reporting periods.

Subscription Churn Thresholds

Monthly subscription churn (MRR loss / starting MRR) should be tracked separately from customer churn. A customer may churn but be replaced by a new subscriber, masking underlying retention problems.

Healthy subscription churn benchmarks vary by category but follow a pattern. Beauty and wellness subscriptions typically see 5 - 8% monthly churn. Meal kits and perishable goods see 8 - 12%. Premium or luxury subscriptions see 3 - 5%.

Month 1 churn (immediate post - purchase): 15 - 25% is normal. Anything above 30% signals onboarding or product fit issues.
Month 2 - 3 churn: Should drop to 8 - 15%. If it stays above 20%, the product is not meeting expectations.
Month 6+ churn: Should stabilize at 3 - 8%. Churn above 10% at this stage indicates a cohort - wide retention problem, not early friction.
Calculate net revenue retention (NRR). If NRR is above 100%, churn is offset by expansion. If below 80%, churn is outpacing growth.

Repeat - Purchase Churn Thresholds

Repeat - purchase churn is harder to define because there is no explicit cancellation event. Instead, set a repurchase window based on average order frequency (AOF).

If average order frequency is 60 days, define churn as no purchase within 120 days (2x AOF). If AOF is 30 days, define churn as no purchase within 90 days (3x AOF). The multiplier accounts for natural variance in purchase timing.

Healthy repeat - purchase churn for apparel: 40 - 50% at 180 days post - purchase.
Healthy repeat - purchase churn for consumables: 30 - 40% at 90 days post - purchase.
Healthy repeat - purchase churn for home goods: 50 - 65% at 365 days post - purchase.
If churn exceeds these thresholds by more than 10 percentage points, investigate product quality, email cadence, or pricing changes in that cohort.

Failure Modes and Red Flags

Churn accelerates when specific operational failures occur. Recognizing these patterns allows operators to intervene before cohort - wide damage.

Sudden churn spikes (5+ percentage points month - over - month) indicate a triggering event: price increase, product reformulation, shipping delay, or email frequency change. Correlate churn timing with operational changes.

Cohort - specific churn (e.g., one acquisition channel has 2x churn of another) points to audience mismatch or expectation misalignment. Pause that channel and audit messaging.

Churn that rises with customer lifetime value (high - value customers churn faster) suggests product quality issues or unmet premium expectations. This is not a pricing problem; it is a product problem.

Measurement Rules and Hygiene

Churn measurement breaks down without clear rules. Establish these before building dashboards.

Rule 1: Define the cohort start date. For subscriptions, use the first charge date. For repeat - purchase, use the order date of the first purchase.

Rule 2: Exclude refunds and chargebacks from the denominator. A customer who refunded is not a customer who churned; they are a failed transaction.

Rule 3: Exclude gift purchases and test orders. These inflate the denominator and lower churn artificially.

Rule 4: Do not mix subscription and one - time purchase customers in the same cohort. They have different churn mechanics.

Rule 5: Measure churn at the customer level, not the order level. One customer with two orders is one customer; two orders do not equal two customers.

Churn Segmentation Checklist

Aggregate churn hides problems. Segment churn by the factors that drive it.

Acquisition channel: Organic vs. paid. Paid channels often have 15 - 20% higher churn than organic.
Cohort age: Month 1, Month 2 - 3, Month 6+. Churn curves should show improvement over time.
Product line: If selling multiple SKUs, track churn by product. One product may be driving overall churn.
Price tier: Higher - priced tiers often have lower churn but higher absolute dollar loss. Track both.
Geography: International customers may have different churn curves due to shipping, returns, or currency fluctuation.
Email engagement: Segment by email opens and clicks. Non - openers churn 2 - 3x faster than engaged customers.

Churn Reduction Priorities

Not all churn is worth fighting. Prioritize based on impact and tractability.

Month 1 churn is high - impact and often tractable. Improve onboarding, set expectations in post - purchase email, and reduce time to first value.

Cohort - specific churn (one channel, one product) is tractable. Pause the channel or reformulate the product.

Churn driven by price is low - impact if it affects only 5 - 10% of customers. Churn driven by product quality affects 30 - 50% and is high - impact.

Churn in high - LTV segments (repeat customers, high - AOV buyers) is high - impact. Churn in low - LTV segments (one - time, low - AOV) is low - impact.

Questions

FAQ

Should we measure churn monthly or quarterly?

Monthly is standard for subscription brands and allows faster detection of operational changes. Quarterly is acceptable for repeat - purchase brands with longer purchase cycles (90+ days). Choose one and hold it constant. Do not mix monthly and quarterly cohorts in the same report.

What is a good churn rate?

There is no universal good rate. Subscription beauty brands with 6% monthly churn are healthy. Meal kit brands with 10% monthly churn are healthy. Repeat - purchase apparel brands with 45% churn at 180 days are healthy. Benchmark against your category and your own historical trend, not against other brands.

How do we distinguish between churn and dormancy?

Dormancy is temporary inactivity; churn is permanent loss. In practice, set a reactivation window. If a customer does not purchase within 2x their average order frequency, they are churned. If they return after that window, they are a reactivation, not a retention. Track reactivations separately; they are a different lever.

Should we include failed payment attempts in churn?

For subscriptions, yes. A failed payment that is not retried is a churn event, even if the customer did not explicitly cancel. Track failed payments separately so you can distinguish between product churn and payment infrastructure churn. Payment churn is often recoverable; product churn is not.

Want this on your account?

Thirty minutes. Bring the number that keeps you up.

Book a demo