A 20% discount on a 40% margin product does not cost you 20%. It costs you half your profit.
Revenue is holding up but profit is not, and the discount calendar has quietly become the marketing plan.
4 min read
Discounting is the easiest lever in ecommerce to pull and the hardest to stop pulling. It works immediately, it is visible in the same-day numbers, and its costs arrive later and in places the promotion report does not look.
This is not an argument against ever discounting. It is an argument for knowing the actual price, which is considerably higher than the percentage on the code.
The arithmetic that surprises people
A discount comes entirely out of gross profit. Your cost of goods, shipping, packaging and payment fees do not fall because you ran a sale.
| Full price | 20% off | |
|---|---|---|
| Selling price | $100 | $80 |
| Cost of goods | $40 | $40 |
| Shipping and fulfillment | $8 | $8 |
| Payment fees | $2.50 | $2 |
| Gross profit | $49.50 | $30 |
| Change in profit per unit | — | -39% |
To break even on that promotion in absolute profit, you now need to sell about 65% more units. Most promotions do not lift volume by 65%. Many lift it by 15% and are reported as a success because revenue rose.
The second cost: who a discount attracts
The margin hit is the visible cost. The retention penalty is the larger one and it arrives months later.
35–45%
Reported reduction in 12-month lifetime value for discount-acquired customers against full-price customers
Niblin discount strategy analysis
Half the profit
What a 20% discount removes from a 40% margin product, because costs do not fall with the price
Saras Analytics on discount profitability
Customers acquired on a discount repeat at materially lower rates. That means their true acquisition cost, measured across the relationship rather than the first order, is far higher than the figure on your dashboard — and the dashboard will keep telling you the discount campaign was your most efficient one.
Promotional dependency, and how to spot it early
The third cost is structural. Discount often enough and customers stop buying between promotions — not because they left, but because they learned to wait.
By the time this is obvious it is difficult to reverse, because unwinding it means a quarter of visibly worse revenue while the audience re-learns your pricing.
- Track full-price revenue as a percentage of total revenue, monthly. A declining line is the earliest reliable signal.
- Track the gap between promotions. If revenue collapses in the weeks between sales, the sales are not incremental — they are your entire demand, rescheduled.
- Track blended gross margin, not just revenue. Stable revenue with falling margin has exactly one common cause.
- Track what share of new customers arrive on a code.
What to do instead, when you need the volume
The goal is not zero discounting. It is discounting that does not train the behavior you are trying to avoid.
- Raise perceived value rather than cutting price: bundles, a free add-on with real margin, faster shipping, an extended guarantee.
- Make the offer conditional on something useful to you — a threshold that lifts average order value, a subscription, a first-party data exchange.
- Give the discount to segments where it changes a decision, not to your whole list where most recipients were going to buy anyway.
- Time-box it genuinely and honor the end date. A permanent sale is a price change with worse optics.
- Prefer a first-order incentive that leads into a strong post-purchase experience over a recurring discount that becomes the relationship.
And measure it against a holdout
Any promotion sent to a list will claim revenue from people who were going to buy anyway. Hold out a random 10% of the segment, compare, and you will learn what the promotion actually caused rather than what it collected credit for. This single habit changes most brands' view of their own promotional calendar.
Common questions
How much does a 20% discount actually cost?
It comes entirely out of gross profit. On a 60% margin product a 20% discount removes about 39% of profit per unit; on a 40% margin product it removes roughly half. To break even you would need a volume lift most promotions do not achieve.
Do discount-acquired customers repeat less?
Reported figures put their 12-month lifetime value 35% to 45% below full-price customers. Split your own cohorts by whether the first order used a code — the gap is usually large enough to change how you budget promotions.
How do I know if my brand is dependent on promotions?
Track full-price revenue as a share of total, monthly. A declining line, plus revenue collapsing between sales, means the promotions are not incremental — they are your demand, rescheduled.
What should I do instead of discounting?
Raise perceived value rather than cutting price — bundles, a margin-positive add-on, faster shipping, a longer guarantee — or make the offer conditional on a threshold or a subscription that pays you back.
How do I measure whether a promotion worked?
Against a holdout. Reserve a random 10% of the target segment, send them nothing, and compare. Without that, you are measuring what the promotion collected credit for rather than what it caused.
How Glimmio handles this
Because Glimmio holds Shopify order data alongside ad spend, campaign performance can be read against what the store actually recorded — including what a promotion did to order value and margin, not only to order count.
Marketing Mix modeling asks where the next unit of budget contributes most, which is the same question a promotional calendar should be answering and usually is not.
- Manual approval by default — nothing runs unattended
- New campaigns and ads are always created paused
- 7-day recovery window on eligible changes
- 48 permissions across 13 roles, scoped per client account
Go deeper on this
The product pages and setup guides that cover what this article describes.
Searches this answers
- discounting impact on ecommerce margin
- do discounts hurt customer lifetime value
- promotional dependency d2c
- discount strategy ecommerce brand
- full price sales percentage benchmark
Read next
Stop Optimizing for ROAS. Optimize for Contribution Margin.
Your best-performing campaign by ROAS might be your least profitable, and your reporting has no way to tell you.
ReadRepeat Purchase Rate: The Number That Decides Survival
You are pouring money into acquisition to replace customers who bought once and never came back.
ReadAverage Order Value Is the Fastest Lever You Have
Acquisition costs keep rising and you have run out of things to fix inside the ad account.
Read