The Discount Incrementality Trap: Most Promotional Spend Goes to Buyers Who Didn't Need It

The Promotion Assumption

Most ecommerce brands run discount codes constantly. Not as an occasional event, but as a permanent feature of the marketing mix. Welcome discounts, loyalty discounts, seasonal events, re-engagement offers, influencer codes. For many D2C brands, there is rarely a moment when a discount code isn’t available somewhere.

The assumption driving this behaviour is straightforward: discounts drive sales. A customer who was hesitating buys. A customer who had abandoned a cart returns. A lapsed customer reactivates. The code closes the gap between intent and purchase.

The assumption is wrong - or at least, it’s far less true than most brands believe.

What Incrementality Actually Measures

When a marketing mix model looks at promotional activity, it’s trying to answer a specific question: of all the transactions that happened during a promotional period, which ones would not have happened without the promotion?

This is the incremental effect. A customer who was going to buy regardless and uses your discount code is not a win for the promotion. You’ve just given margin away to someone who didn’t need the incentive. The sale would have happened at full price.

The non-incremental transactions are the ones where the discount is doing nothing - except reducing your revenue.

The Finding: 8% Incremental

For a D2C ecommerce brand running a marketing mix model across two-plus years of data, the results were significant.

Approximately 75% of all transactions were made using a discount code. This is a brand that runs always-on promotions, has a large active subscriber base receiving offer emails, and uses discount codes across its influencer and paid social activity. At any given time, the majority of their catalogue was discount-accessible.

Of those promotional transactions, the model found that only 8% were incremental. The other 92% were customers who would have purchased at full price. They happened to use a discount code, but the code was not the reason they bought.

Run the numbers: if 75% of transactions use a discount code, and 8% of those are genuinely incremental, then roughly 6% of your total transactions are driven by promotions. The other 94% of discount redemptions are margin transferred to customers who didn’t need the incentive.

Why This Happens

It seems counterintuitive. Why would a customer with a valid discount code not be influenced by it?

The answer is in purchase intent. A customer who is in market, has already found the product they want, and is ready to buy will almost always use a discount code if one is available. They’re not buying because of the discount. They’re buying because they’ve already decided to buy. The discount is a bonus they capture at checkout.

When discount codes are pervasive - subscriber emails, influencer codes, site-wide banners, abandoned cart flows - virtually every motivated buyer will have access to one. This pushes the incrementality rate down significantly, because the code is no longer a signal of price-sensitive hesitation. It’s just a checkout habit.

The incremental customers are the ones who were genuinely on the fence and for whom the discount tipped the decision. At high discount coverage, those customers represent a small fraction of total redemptions.

The Cost of Discount Saturation

This isn’t a reason to stop running promotions. Discounts do drive some incremental volume, and they can play a useful role in acquisition, reactivation, and seasonal demand generation.

But the cost of running always-on, high-coverage discount programs is substantial.

If a brand generates $10 million in revenue and 75% of transactions carry a 15% discount, that’s $1.125 million in margin given away per year. If 8% of those promotions are driving genuinely new purchases that wouldn’t otherwise happen, the incremental revenue gain from promotions might be $600,000 - $800,000 in additional top-line sales. The ROI on the discounts varies heavily by margin structure, but in a category with thin margins, the business case can easily be negative.

The bigger problem is structural. Once customers learn that discounts are always available, full-price purchase behaviour erodes. Customers delay purchases, waiting for the next offer. The discount becomes a floor, not a lever.

What Targeting Incrementality Looks Like

The model finding points to a specific opportunity: rather than broad discount coverage, focus promotional activity on the customers most likely to be incremental.

In practice, this means:

Suppressing discounts for high-intent customers. If a customer has visited the product page three times in 48 hours, they’re likely in the final stages of a purchase decision. Serving them a discount at this point gives away margin you didn’t need to spend.

Using discount triggers more precisely. Cart abandonment at the final payment step is a stronger signal than adding to cart and leaving. Discount codes served to the former are more likely to be incremental than codes served to the latter.

Testing discount removal. For subscriber segments with very high purchase frequency, running a holdout test - removing discount access for a cohort - can directly measure how much volume disappears. Most brands find the answer is much less than expected.

Distinguishing acquisition from retention. First-time customer acquisition via discount is more likely to be incremental (removing a barrier to a first purchase) than retention discounts going to already-loyal customers who would buy regardless.

The Search Equivalent

The same incrementality logic applies to other channels. For this brand, the model also found that within search, Dynamic Search Ads had a significantly lower cost per transaction ($48) than general search terms ($92). The gap was driven partly by precision: DSA campaigns match specific product queries, reaching customers further along in the purchase journey.

The higher-spend general terms were capturing broader, less purchase-ready traffic. The cost per transaction was higher because a larger share of clicks were going to customers who were researching rather than buying.

In both cases - discounts and search - the question is the same: of the customers you’re reaching and converting, how many of those sales are actually caused by your marketing?

The answer, consistently, is less than the headline number suggests.

Key Takeaways

• Of promotional transactions for a D2C ecommerce brand, only 8% were incremental - the other 92% went to buyers who would have purchased without the discount
• Pervasive discount coverage trains customers to expect discounts and suppresses full-price behaviour
• Targeting incremental customers specifically (through suppression, triggers, and holdout testing) can recover margin without losing volume
• The same incrementality question applies across channels: Dynamic Search Ads outperformed general search 2:1 in cost per transaction by reaching higher-intent customers
• Always-on promotions are not inherently wrong, but treating every redeemed discount as a successful conversion is a measurement error