Discount Customers Have Lower Lifetime Value. Ours Didn't.

Someone in your last planning meeting said the brand needs to stop discounting. Somebody else agreed. Nobody asked to see the cohort data, because everyone in the room already knows what it would say.

Discount buyers are worse customers. They churn faster, buy less often, and never pay full price again. It's one of the few things in ecommerce that gets treated as settled.

In the discount-heavy apparel brands I've worked with, it doesn't hold. Full-price first buyers repeat at a little over half the rate of every discounted band. Not the deepest band. Every one of them.

The consensus is real, and thinner than it looks

This isn't a strawman. The position is everywhere, stated with confidence, usually with a churn multiple attached.

Go looking for what those numbers rest on and the ground gets soft. The more careful articles admit their figures are aggregated ranges from published benchmarks and shared cohort analyses rather than first-party data. Some cite forum threads. Almost none show a cohort table from an account they operate.

That doesn't make the consensus wrong. It makes it untested in the only place that decides your budget, which is your own customer file.

Where the reversal shows up

The scope matters, so let me be precise about it. This pattern appears in brands where discounting is already the default. Promotional calendars running most weeks, welcome offers as standard, a customer base that has learned when to wait.

In those brands, split first-time buyers by discount band on the first order and track repeat rate at 180 days. Full price lands at the bottom, around seventeen percent. Every discounted band sits well above it, in the high twenties to around thirty.

The detail that matters is the absence of a gradient. If price sensitivity drove repeat behavior, deeper discounts should produce worse customers. They don't. Shallow and deep bands behave similarly, and both beat full price comfortably.

A flat gap across bands rules out price as the variable. Something else is doing the work.

At a discount-heavy brand, full price is the anomaly

Here's the reading I'd defend, and it rests on a mechanism rather than a correlation.

At a brand that discounts constantly, the discount is the normal way in. It arrives in an email or an SMS, sent to someone who subscribed. To receive it at all, the customer had already handed over a direct channel. The discount is the visible thing. The subscription is the thing doing the work.

So who buys at full price in that environment? Someone who arrived outside the usual path. Paid search, paid social, a link from somewhere, converting before they ever entered an owned channel. For a large share of them, the brand has no way to reach them again except by buying the impression a second time.

The same accounts confirm it from the other direction. Customers acquired through owned channels repeat at roughly 1.7 times the rate of customers from paid search and shopping at the same point in their life.

What the cohort table ranks is whether the brand can reach the customer again. Price sensitivity has very little to do with it, and that puts the real problem somewhere the discount debate never looks.

What this doesn't mean

It isn't an argument for discounting more, and I want to be blunt about that, because the finding is easy to weaponize.

Discounting still destroys margin and the arithmetic hasn't moved. Cut thirty percent off a product carrying forty percent cost of goods and you compress gross profit by far more than thirty percent. You need substantially more volume to stand still. I've written before about brands drifting into calling themselves discount-heavy as though it were a strategy rather than the accumulated result of never deciding otherwise. All of that still holds.

There's a sharper version of the point. The reversal shows up because the brand is discount-heavy. Having trained the base to buy on promotion, the brand converted its owned channels into the primary acquisition path, and full price became the route taken by people it has no relationship with. The pattern is a symptom of the discount problem, not a defense of it.

You can hold both at once. Discounting is expensive. Discounted customers aren't the ones to blame for your repeat rate.

Why this ranks your channels backwards

Most brands judge acquisition on first-order economics. Contribution margin on the first purchase against what the customer cost. It's the only figure available on day one, so it becomes the figure that moves budget.

Full-price cohorts win that comparison automatically. No discount, higher contribution, better first-order return. They score highest on the metric and repeat worst in reality.

Run that for a few quarters and you steadily shift spend toward the acquisition sources least likely to produce a second order, while the reporting says efficiency is improving throughout. It's the failure mode behind LTV Drift, arriving through a different door.

The compounding version is worse. Weakening repeat behavior in newer cohorts is the pattern behind cohort compression. A brand optimizing toward full-price acquisition has been manufacturing it deliberately while believing it was protecting margin.

The decision this forces

Stop segmenting acquisition by discount and start segmenting it by whether the customer entered an owned channel.

That changes who owns the number. If owned-channel acquisition produces the customers who come back, list growth stops being a retention activity reported after the fact and becomes an acquisition line with a budget attached. Right now it usually sits with lifecycle as a soft metric while acquisition budget gets set on channel ROAS in a different meeting entirely.

The second decision is about the room. If your acquisition mix is being set on first-order economics, whoever owns repeat rate should be there before the allocation is made, with the standing to say the ranking is inverted.

Monday morning diagnostic

Two pulls. The first one is the whole argument.

One. Take first-time buyers from a single month at least six months old. Split them by whether a discount code was applied to that first order. Calculate 180-day repeat rate for each group. If your discounted cohorts match or beat full price, the belief your budget rests on doesn't survive contact with your own data.

Two. Split the same cohort by acquisition channel, separating owned from paid. Compare the two splits. If the channel gap is wider than the discount gap, you've found the real variable, and it isn't price.

Before either, check whether the scope applies. If your promotional calendar runs most weeks and your welcome offer is standard, you're the brand type this shows up in. If you hold full price and discount rarely, expect the conventional result, and the second pull still matters more than the first.

If you can't run pull one because discount usage isn't stored against the order, that's the finding. You've been making a categorical judgment about a customer segment you have never been able to measure.

If you run it and the answer surprises you, that's usually where the interesting work starts.

---

FAQ

Do discount customers have lower lifetime value than full-price customers?

The widely repeated position in ecommerce is yes, but much of the published evidence is aggregated benchmark data rather than first-party cohort analysis. In discount-heavy apparel brands the pattern reverses: full-price first buyers repeat at a little over half the rate of every discounted band, with no gradient by discount depth. The likeliest explanation is channel rather than price sensitivity.

Where discounting is the default, discounted buyers arrive through owned email or SMS channels they had already subscribed to, leaving the brand able to reach them again, while full-price buyers more often arrive through paid media and never enter an owned channel at all.

Does this apply to every ecommerce brand?

No, and the scope is the point. The reversal appears in brands where discounting is already the norm, with promotional calendars running most weeks and a standard welcome offer. In that environment the discount is the ordinary route in and full price is the exception, which inverts what each cohort represents. Brands that hold full price as a rule should expect the conventional result. The way to know which you are is to run the split rather than assume.

How do you measure whether discounting is hurting retention?

Take first-time buyers from one acquisition month at least six months old, split them by whether a discount applied to the first order, and calculate repeat rate at 180 days for each group. Then run the same cohort split by acquisition channel instead. Comparing the two shows whether discount or channel is the variable actually driving repeat behavior. Blended retention figures across all customers hide both.