Discounting is one of the most familiar tools in ecommerce.
When sales slow, marketers launch an offer. When they want to acquire new customers, they introduce a welcome discount. When someone abandons a basket, a percentage-off code often follows. When a seasonal event approaches, another promotion appears on the calendar.
Discounts are easy to create, easy to communicate, and easy to measure. We can see whether the code was redeemed, how much the customer spent, and how much revenue the campaign generated.
That visibility can make discounting feel reassuringly scientific.
But are we measuring whether the discount worked, or simply whether someone used it?
That question came to mind when I read Seguno’s 2026 Shopify Unique Discount Code Benchmark (https://www.seguno.com/unique-discount-code-benchmarks). The study analysed 6,099 discount sets containing 182 million unique, single-use discount codes from 1,348 Shopify brands.
The findings are useful. Percentage-off discounts were associated with higher average order values. Amount-off offers achieved higher redemption rates. A 20% discount appeared to provide the strongest balance between the two. Minimum-spend requirements substantially increased average order value for amount-off promotions, while relationship-based discounts dramatically outperformed seasonal offers on redemptions.
There is plenty here for ecommerce teams to test.
However, the research also highlights a much bigger question about the way we approach discount strategy.
Too often, we begin with the discount and then ask what it achieved.
Whereas, we should begin with the behaviour we want to create.
A Discount Is Not a Strategy
Brands often talk about their discount strategy as though discounting were a single activity.
It isn’t.
A percentage reduction, fixed monetary amount, minimum-spend threshold, loyalty reward, welcome offer, and seasonal promotion may all reduce the price paid, but they perform very different behavioural jobs.
They frame value differently. They require different levels of mental effort. They may attract different types of customer and encourage different purchasing behaviour.
This is why asking which discount works best is almost impossible to answer.
Best for what?
Seguno’s data illustrates this nicely. Percentage-off promotions were associated with an average order value of $95.75, compared with $67.35 for amount-off offers. Yet amount-off codes achieved the higher redemption rate: 0.87% compared with 0.69%.
Neither is therefore the universal winner.
One appears better at encouraging larger baskets. The other appears better at prompting more people to act.
The right choice depends on the objective, the customer, the margin, and the behaviour the retailer wants to encourage.
Yet many promotional calendars still treat discounts as interchangeable. Ten pounds off becomes 15% off, which becomes 20% off when the first offer fails to generate enough response.
That isn’t strategy. It is escalation.
Redemption Does Not Necessarily Equal Success
Redemption rate is an obvious measure of performance. If more people use the code, the offer appears to have worked.
But redemption only tells us that the customer accepted the discount.
It does not tell us whether the discount changed their behaviour.
Would they have purchased anyway? Did it generate an incremental purchase, or simply reduce the margin on an order that was already going to happen? Did it attract a customer likely to return, or someone who will simply wait for the next promotion?
Seguno itself makes an important distinction here. The research does not estimate incremental revenue lift because that would require a control group.
That caveat matters. The findings show a relationship between certain discount structures and outcomes such as redemption and average order value. They do not tell us whether the discount caused the purchase.
This is why discount measurement needs to move beyond how many people used the code. We should also be asking whether the offer changed what customers did.
Beware the 20% Shortcut
One of the headline findings is that 20% off provided the best balance between average order value and redemption rate among the most common percentage discounts.
Useful? Absolutely.
A recommendation that every retailer should start offering 20% off? Definitely not.
Benchmarks can quickly become defaults, and defaults can become habits. When brands repeat the same promotional pattern often enough, customers learn it too.
They begin to recognise that another offer is probably coming. They delay purchasing. They ignore the first promotion because experience tells them a stronger one might follow. Full price stops feeling like the real price and starts feeling like the amount paid by people who didn’t wait.
I recently wrote about how retailers can deliberately build purchasing habits by reinforcing repetition and making the next purchase easier.
But habit formation works both ways. Brands can accidentally train customers to associate buying with discounts.
The cue becomes the promotion. The behaviour becomes purchasing during the offer. The reward becomes the saving.
At that point, the brand may not be creating a habit of buying. It may be creating a habit of waiting.
Relevance May Matter More Than Discount Depth
One of the most interesting findings in the research is that relationship-based codes were redeemed at around five times the rate of seasonal promotions. These included offers given to groups such as employees, friends and family, or students.
It would be easy to conclude that exclusivity makes discounts more attractive, and that may well be part of it. But I think relevance is doing a lot of the work.
A seasonal discount is based on the retailer’s calendar. A relationship-based discount is connected to something about the recipient.
It reflects who they are, which group they belong to, or their relationship with the brand. It therefore feels less like a generic attempt to generate sales and more like a benefit intended for them.
The same pattern appeared in segmentation. Codes restricted to eligible customer groups were associated with considerably higher average order values than unrestricted promotions.
Which raises an interesting question. Are brands relying on bigger discounts because they are not investing enough in relevance?
A well-timed, targeted 10% offer could potentially be more persuasive than a blanket 20% reduction. A reward tied to a customer’s progress may carry more meaning than another generic seasonal sale.
When the offer feels relevant, the customer is not only evaluating the saving. They are also evaluating why they received it.
Minimum Spend Changes the Decision
Seguno also found that adding a minimum-spend requirement to amount-off codes was associated with a 2.6-times increase in average order value.
Behaviourally, this makes sense. A minimum spend turns a discount into a goal.
Get £10 off offers a saving.
Spend £50 and get £10 off creates a target.
Once the customer starts building their basket, they begin evaluating their progress towards that threshold. Someone with £42 worth of products may add another item because they do not want to lose the £10 saving.
That can be an effective way to increase basket size. But again, higher average order value does not automatically mean better performance.
The additional product has a cost. The discount has a cost. Delivery may have a cost. The order may also cross a free-delivery threshold.
The smarter question is not simply whether the customer spent more. It is whether the threshold encouraged profitable incremental spending.
Discounts Can Solve the Wrong Problem
Another danger is using discounts as the automatic response whenever performance drops.
Low conversion? Offer a discount.
Basket abandonment? Offer a discount.
Poor re-engagement? Offer a discount.
But price may not be the actual barrier.
Customers may be worried about delivery costs, unsure about quality, confused by the returns policy, overwhelmed by choice, or simply not ready to buy.
Offering a discount without understanding the barrier is like reducing the price of a door the customer cannot work out how to open. This is also why discount strategy cannot be separated from the rest of the ecommerce experience.
The offer, product price, delivery threshold, loyalty programme, returns policy, and post-purchase journey all influence the same decision.
Begin With the Behaviour
The Seguno research gives ecommerce leaders some genuinely useful benchmarks, but the bigger lesson is that discounts should not be treated as a single promotional lever.
Before deciding whether to offer 10%, 20%, £10 off, or a minimum-spend reward, define the behaviour you are trying to influence.
Are you trying to encourage a first purchase? Increase basket value? Bring a lapsed customer back? Reward an established customer? Encourage a second purchase? Move excess stock?
Those are different situations, and they should not automatically receive the same incentive. Then decide how you will recognise success.
Redemption rate and average order value matter, but depending on the objective, so might incremental conversion, contribution margin, repeat purchase rate, time to second purchase, and longer-term customer value.
And wherever possible, use a control group. Without one, you know what happened after the discount was offered. However, you do not know what would have happened without it.
The Discount Is Teaching the Customer Something
Every discount communicates more than a saving. It teaches customers how your brand behaves.
It teaches them whether offers are rare or routine. Whether patience is rewarded. Whether the first offer is worth accepting. Whether full price is credible.
That makes discounting a form of behavioural design.
Used carefully, discounts can reduce uncertainty, encourage trial, reward progress, increase basket value, or strengthen a customer relationship.
Used habitually, they can train customers to delay purchasing and wait for the next incentive.
So the question is not whether discounts work. Of course they do. Customers like paying less.
The better question is: what behaviour are they teaching? Because a campaign can generate more redemptions, a higher average order value, and an impressive revenue figure, while still encouraging exactly the wrong customer behaviour.
