Your last promo campaign hit +50% revenue lift. Or… did it?
The number your promo dashboard shows and the number your revenue team cares about are rarely the same. A promo campaign that looks like a revenue win can quietly destroy contribution margin, and most teams don't find out until the finance review later. It all happens because the promo performance gets measured by one metric and business health gets measured by another.
So I did the math for you – here are 4 silent margin killers behind every discount and a small gift at the end, so stick around.
Take a standard scenario. You run a 20% off coupon. You show the offer to one group of customers (the treatment group) and withhold it from another (the control group) to measure the difference. Treatment converts at 3%, control at 2%. That's a clean +50% lift.
Now let’s analyze 4 things most promo dashboards ignore.
Your control group (the customers who weren't supposed to see the offer) saw the promo anyway. Through the homepage banner, an email forward, an influencer post.
If even 25% of your control was contaminated, the true baseline conversion wasn't 2%; it was closer to 1.67%. The lift looks bigger on paper, but only because you were comparing against a group that was already influenced, not because more people actually bought.
A chunk of those "incremental" orders were heavy buyers who just moved their purchase up a week to grab the discount. They were going to buy anyway, but you paid them to do it sooner. Research on promotional economics consistently finds this pattern: when more than half your redeemers are repeat buyers, a big portion of the lift is just shifted timing, not new demand.
Your cost-per-order looks like €16. But next month, baseline orders dip because those sales already happened. Factor that in and the real cost is closer to €21.
Of everyone who redeemed, how many actually needed the discount to convert?
If 60% of your redeemers are high-frequency buyers (3+ orders per year), the subsidization rate (the share of discounts that went to people who didn't need them) can hit 54%. That's more than half your promo budget going to people who were already reaching for their wallet.
At scale, ecommerce stores lose tens of thousands per year on dedicated buyers who would have paid full price.
Customers who bought on deep discount return products at much higher rates than full-price buyers. The reason is simple: when something costs less, people think less before buying. That boosts cart size, which boosts returns. Fashion is the worst offender, with return rates in the 25-35% range during promo periods, and Gen Z shoppers showing nearly 45% higher return rates than average.
Stack those four and the +50% headline lift collapses to roughly +24% in real revenue terms.
Revenue is only half the picture.
Every discounted order also carries less contribution margin (what's left per order after product cost, shipping, and fulfillment), because the discount eats into the same pool the four harms already thinned out.
On a product with 35% contribution margin, a 20% discount means each order keeps only 43 cents of every margin dollar it would have earned at full price. That compression applies to every order in the promo, including the ones that would have happened without the discount.
Run the full math and that +50% headline revenue lift becomes a -47% contribution margin story. On a €1 million campaign, that's €110,000 in imagined margin: money the dashboard said you earned but the P&L never saw.
The answer is simple – it's because the tools they use were built to measure one thing (did people convert) and the business cares about something else entirely (did we make money).
Most promo dashboards show conversion lift. Maybe revenue lift if you're lucky. Almost none of them subtract holdout contamination, cannibalization, subsidization, or returns drag. And none of them translate what's left into contribution margin terms.
So teams optimize for the metric they can see (conversion) and quietly erode the metric they can't (margin).
The structural fix isn't necessarily to discount less. A better approach is to measure honestly, then discount smarter. Know which customers actually need an incentive to convert and which ones are just cashing a coupon they didn't need. Know whether your holdout is clean. Know your real return rate on promo cohorts, not the blended number.
We built a calculator that does this math live.
Plug in your discount depth, your heavy-buyer share, your baseline return rate, holdout contamination, and contribution margin. It shows you the headline lift, the true lift after all four harms, and the gap between them, in both revenue and margin terms. Maybe even grab a person from the finances team to go through this exercise together.
And don’t get me wrong – incentives work. They're one of the strongest levers in commerce when they're designed well and measured honestly. The point is that most teams are measuring wrong, and the cost of that gap compounds every campaign.
If the calculator tells you a story you didn't expect, that's the conversation worth having before the next campaign goes live, not after.
.svg){kind=icon}
