Reactivation Rate Calculator

Written out

Reactivation rate is lapsed customers who purchase again divided by lapsed customers targeted: reactivated ÷ targeted × 100.

The meaningful version is incremental, reactivations above the baseline rate at which lapsed customers return unprompted.

Rate = (contacted group return rate − holdout group return rate), which requires a holdout to measure.

In practice

3,000 lapsed customers contacted, 240 return, an 8% reactivation rate. A holdout of 500 comparable customers not contacted sees 22 return, a 4.4% baseline.

Incremental reactivation is 3.6 percentage points, or 108 customers rather than 240. The campaign caused fewer than half of the returns it is credited with.

At $110 of remaining value each, the incremental gain is $11,880 against a campaign cost of $2,300, still a 417% return and less than half the unadjusted figure.

Both numbers justify the campaign and only one supports a decision about how much more to spend on it.

The limitations

Reactivation rates decline sharply with time lapsed. Customers away six months return at several times the rate of those away three years, and a blended figure across the whole lapsed file describes neither.

Whether reactivated customers stay also matters more than whether they return once, and a single-order reactivation is worth far less than a resumed relationship.

Putting it to use

Segment the lapsed file by time since last purchase and by prior value, and target the recent high-value group first where the return is concentrated.

Then track the reactivated cohort for six months. A campaign producing one order and another lapse has bought a transaction rather than a customer.

Why the lapsed file is usually the best list a business owns

Every name on it has bought before, is contactable without media spend, and has demonstrated the product works for them at least once.

Compared against cold acquisition at $27 a customer, reactivation at under $10 is the cheapest source of customers most businesses have, and it is typically worked once a year if at all.

The reason it goes unused is that lapsed customers feel like a failure rather than an asset, and the list sits in the platform unexamined. Treating it as a defined audience with a known value, segmented and worked quarterly, is one of the more reliable improvements available to an established business.

Suppressing the genuinely unreachable: hard bounces, unsubscribes, customers who asked not to be contacted, before measuring the rate produces a figure that describes the addressable file rather than the whole database.

It also avoids the deliverability damage that comes from repeatedly mailing addresses that have been dead for years.

Comparing reactivation performance across the reasons customers originally lapsed, where that is known, focuses the campaign on the causes that are genuinely recoverable.

Measuring the second-order rate among reactivated customers distinguishes a recovered relationship from a single opportunistic purchase, which is the outcome worth optimising for.

Testing subject lines and offers separately on the lapsed file identifies what actually brings people back rather than what is assumed to.

The definition of lapsed has to be set from the actual purchase cycle rather than from a round number of days. A category where customers buy every eight weeks needs a different threshold from one where they buy annually, and using ninety days for both mislabels half of them. Deriving the threshold from the distribution of gaps between orders takes one query and makes every subsequent figure meaningful.