Score enough DTC customer bases and certain shapes stop being surprising. The most consistent one: roughly four in five customers buy exactly once. We have seen it in beverage, in footwear, in consumables. Categories differ, price points differ, brand strength differs, and the one-time buyer rate keeps landing within a few points of 80%.
Most operators know their repeat rate is low. Fewer have internalized what sits directly behind it, so this note works through the whole shape: the multiplier, the concentration hiding inside it, and why the order of operations decides whether the biggest lever in your book prints money or leaks it.
The multiplier
When we score a one-time buyer's 12-month forward value, the number is usually small. Single digits of dollars in many categories. When we score the same customer after a second purchase, the number jumps by a factor of five to nine, depending on the category.
| Customer state | Typical 12-month forward value | Multiple |
|---|---|---|
| One purchase | Single digits of dollars | Baseline |
| Two purchases | Five to nine times the one-purchase figure | 5x to 9x |
| Three or more | Higher still, but the big jump already happened | The tail of the curve |
That is not a rounding-error improvement. It is the single largest per-customer value change that exists in a DTC book. Acquisition doubles your customer count slowly and expensively; a second purchase multiplies an existing customer's expected value overnight. On leverage alone, second-purchase conversion is the highest-yield motion most brands have.
The concentration hiding inside the 80%
Here is the trap. If the multiplier is that large, why doesn't blasting every one-time buyer with a winback offer print money?
Because the multiplier is an average across customers who are not average. Within that 80% of one-time buyers, forward value is wildly concentrated:
A thin slice are customers whose timing, basket, and behavior suggest real future value if they return. The long tail are customers who bought a gift, chased a discount, or churned for reasons no email will reverse. Their expected value after a second purchase is still small, and the cost of chasing them, in discounts and list fatigue, is real.
Send the same offer to all of them and the economics of the tail swallow the economics of the slice. The program "converts" and still leaks money. This is why untargeted winback campaigns feel busy and produce flat contribution.
The strongest objection
"Email is free. Reach costs nothing, so why not send to everyone and let the winners self-select?"
Email is not free; it only invoices in ways the campaign report doesn't itemize. Every tail redemption is margin handed to a customer who was never coming back profitably. Every irrelevant send burns a little deliverability and a little attention, which taxes the emails that do matter. And a standing offer broadcast to the whole file trains your best customers, the slice, to wait for codes they would have purchased without. The tail doesn't just fail to pay; it degrades the economics of the customers who would have.
Selection before intervention
In a completed controlled trial with a DTC apparel brand, selection alone, choosing which one-time buyers to invest in before sending anything, beat untargeted marketing several times over. The offer amplified that advantage; it did not create it. The order of operations matters:
- Score every customer by forward value, with an interval.
- Select the one-time buyers whose scores justify a dollar of investment.
- Intervene on that pool only, and measure the change against a control.
Run in that order, the second-purchase motion is the most valuable play in the book. Run in reverse order, offer first, selection never, it is a discount program with good branding.
The verdict
The one-time buyer wall is the largest store of unrealized value in a DTC business, and the multiplier for breaching it is 5x to 9x per customer. But the multiplier belongs to a thin slice, and only selection can find the slice before the money goes out. Leverage this large has a sign, and targeting decides it.
- Look up your one-time buyer rate today. If it is near 80%, most of your file is a decision being made by default.
- Estimate what moving one point of one-timers to a second purchase is worth at a 5x multiplier. That number is your ceiling for caring about this.
- Before the next winback send, decide who not to send it to. That single subtraction is where the economics turn.
Telling the slice from the tail before you spend is exactly what forward CLV is for, and it is what the free diagnostic shows you on your own customers.