Average churn rate: choose a comparable baseline

An average churn rate is useful only when its observation period, customer group and definition match yours. Start with a baseline from your own fixed starting cohorts, then use outside figures as context. A lower percentage can reflect larger accounts or longer contracts rather than better retention.

Decide what the average describes

The word average can hide several different questions. Do you want the typical reporting month, the pooled loss rate across business units, or the share of a particular customer cohort that disappeared? Those answers use different weights and sometimes different denominators.

Use the churn definition for the underlying measure. For benchmarking, the important step is to write a short comparison label: customer accounts active at the start, lost during the calendar month, excluding accounts added during that month. Keep the customer identity and loss rule unchanged across every period in the baseline.

That discipline also applies to published metrics. The SEC's guidance on key performance indicators calls for a clear definition, calculation method and explanation of how management uses a metric. Use those disclosures to assess comparability; a familiar metric name alone is insufficient evidence.

Make a comparison card before collecting benchmarks

Record the following beside every candidate average:

An unavailable field is a limitation, not a detail to fill in by assumption. If a report does not disclose its customer mix or denominator, it may still suggest a question to investigate. It cannot establish whether your own performance is above or below a comparable norm.

Industry differences deserve a separate comparison. The guide to average churn rate by industry explains why industry labels need matching contract and measurement rules too.

Build an internal baseline from the export

Use a customer status export and a recurring revenue export. Keep a stable account identifier, segment at the start, active status at the start, effective loss date, starting recurring revenue and the recurring revenue attached to lost accounts. Resolve multiple subscriptions to the customer unit before counting customers.

Here is a fictional US subscription business for September 2026. Both segments use the same monthly observation window and effective cancellation rule. All amounts are monthly recurring revenue; there are no partial cancellations or revenue adjustments in this example.

These are illustrative internal figures, not published industry averages. They answer two separate questions:

Percentages are rounded to one decimal place. Customer churn is higher because the lost accounts carry less recurring revenue than the customer base overall. Neither percentage replaces the other: customer success needs the account losses, while finance needs the revenue exposure.

Prove the baseline before interpreting it

The count reconciliation is 1,000 starting customers minus 46 losses equals 954 retained customers. Those retained customers must come from the starting list. Newly acquired accounts are a separate movement and cannot erase a loss from this cohort.

The company rate must also equal its weighted segment rates. For customer churn, the check is (900 × 5.0% + 100 × 1.0%) / 1,000 = 4.6%. A simple average of the two segment percentages gives 3.0%, which describes equally weighted segments rather than the company customer base.

Revenue uses different weights: (180,000 × 5.0% + 120,000 × 1.0%) / 300,000 = 3.4%. The lost recurring revenue should tie to the canceled accounts' starting recurring amounts, rather than to the value of final invoices or cash receipts.

Keep an average of repeated monthly exposures separate from a full-period cohort result. A customer present at several monthly starts contributes to several monthly exposure counts. A fixed annual starting cohort counts that customer once. The renewal, retention and churn denominator guide covers the underlying distinctions.

Use the baseline to choose the next question

Finance should own the approved measure and reconcile revenue. Customer success or account management should own the loss reasons. Sales operations should maintain customer identifiers, segments and contract terms so the comparison survives account changes.

In the example, small accounts deserve a closer look because their loss rate exceeds the large-account rate under identical rules. That finding does not identify the cause. Compare onboarding age, acquisition channel and cancellation reasons before choosing an intervention; churn rate analysis takes that investigation further.

For recurring software contracts, use average SaaS churn to examine renewal opportunities by contract term. Reassess the baseline after changes to customer mix, but preserve the prior segment rates so a mix shift does not look like an operational improvement.

Where it goes wrong

• Mixing time windows. A monthly loss percentage and an annual loss percentage describe different exposure periods. Label both and compare equal windows.
• Changing the starting population. Adding new customers to the denominator dilutes losses from the original base. Rebuild the starting list independently.
• Changing revenue settings. Stripe's billing analytics documentation describes configurable discount treatment and active-subscriber definitions. Record the settings behind an export before comparing it with another source.
• Treating missing reasons as voluntary losses. Keep unknown classifications visible until an owner resolves them. A complete total does not guarantee a complete explanation.

Covirage is not affiliated with Stripe.

Compare your own baseline in Covirage

Bring a sample of your own customer status and recurring revenue exports to Covirage. Covirage maps the columns, and its tools compute and check comparable cohort figures; the external AI model explains the results with every figure traced to its rows. Create a free Covirage account, which takes about a minute, then set up your business and bring your file.

See renewal analysis with Covirage for the supporting business context.