Measure New Feature Adoption With an Eligible-Customer Cohort

A new feature’s usage count can rise because more customers became eligible, because more eligible customers adopted it, or because existing users used it more often. Separating those effects gives product and customer teams a fairer view of adoption and a clearer next action.

Define the eligible population

Specify which plans, roles, regions, product versions, or permissions make a customer able to use the feature. Exclude accounts that could not access it during the measurement period. If eligibility changes, keep the date and rule so the denominator can be reconstructed.

Choose the unit that matches the adoption question: user, account, workspace, or team. Multiple users at one account can inflate a user-level rate even when the business question is account adoption.

• Document eligibility and exposure dates.
• Keep a reason for exclusions that can be audited.
• Report both account and user views when they answer different questions.

Separate exposure, activation, and repeated use

A feature announcement does not prove a user saw it. Capture exposure where possible, then define the first meaningful action that counts as activation. Avoid counting a page view or accidental click as adoption if the feature’s intended value requires a completed task.

Repeat use indicates a different behavior from first use. Choose a follow-up window that fits the feature’s natural cadence; a monthly workflow should not be judged by daily return behavior.

• Eligible → exposed → activated → repeated use.
• Define event names and properties before release.
• Check instrumentation with representative accounts and roles.

Compare cohorts and learn from gaps

Compare adoption by release cohort, role, customer maturity, and use case, while controlling for differences in eligibility and exposure. A low adoption rate may signal poor discoverability, unclear education, an onboarding gap, or a feature that is not relevant to the segment.

Pair quantitative patterns with customer feedback and support questions. Do not treat low usage alone as evidence that customers need more promotional email; the problem may be fit, workflow, or access.

• Track activation time and repeat use separately.
• Review unactivated eligible accounts for common blockers.
• Use the result to change education, access, or product design.