Your retention program almost certainly looks more profitable than it is. Not because the numbers are wrong, but because “attributed revenue” is not the same as “incremental revenue” — and most platforms only report the first one.
The attribution problem
When a customer opens your reorder email and buys within seven days, that purchase gets credited to the email. But here’s the question most brands never ask: would that customer have bought anyway?
If the answer is yes — and for your most loyal customers, it often is — then the email didn’t cause the purchase. It just happened to be nearby. You’ve paid for the send, you’ve credited the revenue, but the causal relationship isn’t there. Multiply that across thousands of sends and you can be spending real money on emails that are doing nothing.
What incremental revenue actually measures
Incremental revenue answers a different question: how much did we sell because of this program that we wouldn’t have sold without it? The measurement method is a holdout: send your campaign to 90% of eligible customers, withhold it from 10%, compare purchase rates. The difference is your true lift.
This is more complex to set up than standard flow reporting. It’s also the only way to know if your program is actually working.
Three metrics that matter more than attributed revenue
- Holdout lift. Incremental purchase rate in treated group vs. control. This is the number. Everything else is noise.
- Incremental revenue per send. Divide total incremental revenue by total sends. This is your true cost-effectiveness metric and tells you whether reducing send volume would increase or decrease profitability.
- Margin impact, not revenue. If you’re offering discounts to drive retention, revenue is the wrong denominator. Gross margin tells you whether the program is actually profitable after the incentive cost.
A program that generates £400k in attributed revenue and £60k in incremental revenue isn’t a success story. It’s a measurement problem waiting to be discovered.
The practical path forward
Start with a holdout test on your most active retention flow. Keep 10% of eligible customers out of it for 90 days. Compare their purchase behavior to the 90% who received the flow. That gap — adjusted for baseline purchase probability — is your real number.
Most brands run this test once and revise their entire view of what their retention program is actually worth. Some discover the program is genuinely additive. Others discover it’s mostly crediting revenue that was going to happen anyway. Both outcomes are valuable. Only one of them is optimistic.
Key takeaways
- Attributed revenue overstates program value — incremental revenue is the real metric.
- Holdout testing is the only reliable way to measure true lift.
- Gross margin impact matters more than top-line revenue when discounts are involved.