- Four metrics: cohort size, enrollment rate, reactivation rate, revenue attributed.
- 14-day attribution window - not 30, not 60.
- Hold out 10% of cohort with no messages - that's your baseline.
- Exclude customers with upcoming bookings from the denominator.
- Revenue Recovery Dashboard launching Q2 2026.
1. Cohort Size
Snapshot day 0. Apply exclusions up front. Log contact IDs. Typical US SMB: dormant cohort 20-40% of total contacts.
2. Enrollment Rate
Real values 65-90%. Below 80% means operational issues - template approval lag, invalid numbers, throughput caps. Fix those before optimizing copy.
3. Reactivation Rate
Qualifying event = completed purchase, showed-up booking, deal advanced. Not opens. Not clicks. Attribution window: 14 days.
Holdout Group (Mandatory)
Randomly assign 10% to receive no messages. Measure their reactivation rate too. True lift = sent - holdout. Without this you cannot separate saves from customers who were returning anyway.
4. Revenue Attributed
Subtract the discount. Net out holdout baseline. Only events within the 14-day window count. Gross revenue minus discount = actual campaign contribution.
Common Attribution Mistakes
- Attribution window too long.
- No holdout.
- Counting opens as recoveries.
- Including upcoming-booking customers in denominator.
- Ignoring discount cost.
- Comparing cohorts with different dormancy thresholds.
The Dashboard You Need
All 4 side-by-side, filterable, with holdout logic built in. App-ening's Revenue Recovery Dashboard (Q2 2026) computes these automatically with 14-day attribution and 10% holdout pre-wired.
Related: reactivation journey template, win-back offer design.