Why Do Customers Only Buy When There’s a Discount?

Monday is quiet. Tuesday is quieter.

On Wednesday, the business announces 20% off. Orders arrive. Existing customers return. New names appear in the sales report. The promotion ends on Sunday night, and by Tuesday the business is quiet again.

The conclusion seems obvious.

Customers have learned to wait for discounts.

Perhaps they have. But the sales report does not prove it.

The discount may have brought the price within reach. It may have made the offer feel worth the money. It may have attracted people who would never buy at full price. The deadline may have forced an overdue decision. The promotion may also have landed during a week when demand was going to rise anyway.

Some of those discounted customers might have bought next month. Others might have bought tomorrow at full price.

The important question is not simply why sales increased.

It is this:

What changed when you added the discount?

Until you can answer that, removing promotions, increasing them or trying to “retrain” customers is guesswork.

First, find out whether the discount actually created more sales

A promotion can increase sales during the promotional period without creating much additional business.

Imagine a company that normally sells 80 units during a comparable week. It runs a 20% promotion and sells 100.

The promotion appears to have produced 100 sales. It did not.

Some of those customers may have been planning to buy at full price. Some may have moved a purchase forward from the following week. Others may have waited because they expected the promotion. Only part of the increase may represent demand that would not otherwise have appeared.

The useful question is:

How much extra business did the discount actually buy?

Suppose the normal price is 100. Selling 80 units produces revenue of 8,000 before costs.

During the promotion, the price falls to 80 and the business sells 100 units. Revenue is still 8,000 before costs.

That does not automatically make the promotion unsuccessful. The additional customers may return. The business may have moved stock that had a deliberate reason to move. The offer may serve a useful acquisition or capacity objective.

But the headline result, 100 sales instead of 80, tells a more flattering story than the economics do.

The discount was applied to every promotional purchase, including purchases that might have happened anyway. The business needs to distinguish the sales the promotion created from the sales it merely made cheaper.

Look beyond the promotional week where possible. Did the following week weaken? Did customers who usually buy before the promotion postpone their orders? Did the promotion bring genuinely new demand, or rearrange demand that already existed?

You will rarely know what every customer would have done in an alternative reality. You do not need perfect certainty. You need a better answer than “sales went up”.

Who buys when there is no discount?

Aggregate sales can hide several different customer behaviours.

Start with customers who continue buying at full price. Do they behave normally before, during and after promotions? If they do, the business may not have a general discount-dependency problem.

Now look at customers who appear mainly during promotional periods. Are they existing customers changing when they buy, or a different group attracted by the lower price?

That distinction changes the diagnosis.

If established customers increasingly postpone purchases until the next offer, promotions may be changing behaviour inside the existing customer base.

If full-price customers continue buying normally while promotions bring a separate, highly price-sensitive group, the discount may be changing the customers you attract.

New customers require another distinction. A promotion can introduce worthwhile buyers who later return at a profitable price. It can also attract people whose interest ends the moment the discount does.

Check what those customers do next rather than assigning them a future value because they entered the database.

Useful comparisons may include:

• the share of existing and new customers during promotional and full-price periods;
• how frequently each group buys without a promotion;
• whether existing customers move purchases towards predictable offers;
• whether promoted customers return, what they buy and what they pay;
• whether the products or services bought on promotion differ from normal purchases.

The point is not to produce an elaborate segmentation exercise. It is to answer a commercial question:

Are discounts changing customer behaviour, or changing the customers you attract?

Those are different problems. They should not receive the same response.

Are customers waiting because you taught them to wait?

Repeated and predictable promotions can give customers a sensible reason to postpone a purchase.

If a discount appears at the end of every month, a customer buying on the twenty-fifth may learn that waiting five days is worthwhile. If a business sends a recovery offer whenever somebody leaves a basket or ignores a proposal, hesitation can become part of the buying process.

Look for the pattern before accepting the explanation.

Sales may weaken immediately before familiar promotions. Existing customers may concentrate more of their purchases around offers. Promotional periods may become stronger while comparable full-price periods become progressively weaker.

None of those signals proves that every customer has been trained to wait. Together, they would justify a closer investigation.

Listen to customers as well. Do they ask when the next promotion will happen? Do they refer to the previous discounted price as though it were the normal one? Do sales conversations pause when buyers believe another offer is close?

The comparison inside the buyer’s head matters.

A regular price of 100 can feel reasonable when it stands alone. Repeated exposure to the same product at 80 can change what the customer regards as an acceptable price. The 100 has not changed, but the customer now has another number against which to judge it.

A recent study offers a useful, limited warning about that comparison. Andong Cheng and Ashley Stadler Blank examined conditional promotions, where customers must meet a requirement to receive a discount. Across five studies, promotions combining a costly precondition with a large discount were the most likely to reduce total sales compared with no promotion. The requirement discouraged some customers from taking the offer, while the visible discount made the regular-price purchase less attractive. The research concerns conditional offers, not discounting in general.

That distinction is important. The study does not show that discounts reduce sales. It shows that a promotion can alter how the alternative is perceived, even among people who cannot or will not claim the offer.

Your own promotion may work differently. The practical lesson is to inspect more than the discounted option. Ask what the promotion does to the full-price choice sitting beside it.

Or is your full price telling you something?

Suppose promotional demand rises across several customer groups. New customers buy. Existing customers buy. People who did not know a promotion was coming also respond.

The business may not have trained everybody to wait.

The full-price offer may be failing a different test.

A lower price can move the purchase across an affordability boundary. A buyer who cannot spend 1,000 may be able to spend 800, regardless of how strongly they value the offer.

It can also change the perceived exchange. At 1,000, the scope may feel too narrow, the risk too high or the alternatives more attractive. At 800, the same uncertainties become acceptable.

Ask:

What becomes acceptable at the discounted price that was not acceptable before?

Do not answer “value communication” automatically.

Sometimes buyers do not understand what they receive. Sometimes they understand perfectly and still consider the normal price too high. The offer may contain features they do not value, require a commitment they do not want or compare badly with a credible alternative.

The problem could sit in the price, the offer, the packaging, the comparison or the explanation. Treat each as a hypothesis.

Sales conversations can help if the evidence is recorded honestly. “Too expensive” is not enough. What did the buyer expect to pay? What were they comparing? Which part of the offer failed to justify the difference? Did the discount solve the objection, or merely make it small enough to tolerate?

If customers repeatedly become willing to buy at 800 but not at 1,000, the 200 difference is telling you something. Find out what before deciding whether the answer is a lower price, a different offer or a clearer reason to choose it.

Maybe the discount is supplying urgency, not affordability

A limited promotion changes at least two things.

The customer pays less, and the customer has a reason to decide now.

Businesses often credit the lower price for the whole result. The deadline may be doing more work than it appears.

Consider somebody who has been thinking about a purchase for six weeks. They can afford the normal price. They believe the offer is useful. Nothing requires a decision today, so the purchase remains behind more immediate work.

A promotion introduces an expiry date. Waiting now has a consequence.

The customer may buy because the saving matters. They may also buy because the promotion converts an open decision into a dated one.

Look at what happens when the deadline approaches. Do most orders arrive immediately after the announcement, or in the final hours? Do customers ask questions about the price, or about when the offer ends? Does a promotion move people who were already engaged, or bring people who previously showed no interest?

Then ask:

Would these customers respond differently if there were a legitimate reason to decide now without lowering the price?

That is a diagnostic question, not an invitation to invent scarcity.

A real deadline might come from capacity, delivery timing, an application window, a seasonal requirement or the customer’s own operating calendar. If no genuine reason exists, the business should not manufacture one.

The aim is to separate the effect of urgency from the effect of affordability. If you combine them every time, the sales report cannot tell you which one moved the customer.

Check whether seasonality is getting credit for the discount

Promotions rarely happen on random dates.

They appear during Black Friday, holidays, seasonal peaks, payday periods, weather changes, replacement cycles and annual purchasing windows.

Demand may already be unusual.

A retailer discounting winter products during the first cold week cannot assume that the offer created the whole increase. A business promoting annual services when customers normally plan next year’s budgets has the same problem.

Compare reasonably similar periods where you can. Look at previous years, nearby non-promotional periods and products that faced similar demand without the same offer.

The comparison will not be perfect. Its job is to stop the discount receiving credit for every favourable condition around it.

Seasonality may explain little, some or most of the uplift. Treat it as another possibility to test, not a convenient way to dismiss a successful promotion.

Now look at what the discounted sales are worth

By this point, you may know more about who bought and why.

Now ask whether those purchases made economic sense.

Begin with the additional sales that the promotion probably created. Estimate what they contributed after the discount, direct costs and any extra fulfilment or service required.

Then account for the discount given to purchases that may have happened at full price.

A promotion can create twenty useful new orders while reducing the margin on eighty orders the business was likely to receive anyway. The new volume is visible. The margin surrendered across the existing demand is easier to miss.

Do not stop at the first transaction if repeat behaviour can be observed. Promotional customers may later buy profitable work at full price. They may also return only when another offer appears.

Use what the records show. Do not assign a theoretical lifetime value to customers whose later behaviour is unknown.

A promotion can also serve another deliberate purpose. It may fill otherwise unused capacity, support a product launch, reduce ageing stock or introduce buyers to an offer with sound follow-on economics. Lower margin per transaction is not automatically a failure when the promotion has a clear commercial job.

But revenue and activity can make weak economics look healthier than they are. That broader problem is examined in Your Business Is Busy. So Why Isn’t It Making Money?

The useful calculation is not whether promotional revenue exceeded zero.

It is whether the promotion created enough worthwhile additional business to justify what the company gave up and what it had to deliver.

Run one promotion as an experiment, not a reflex

The next promotion should have a question attached to it.

Choose the question before the offer launches. Otherwise, any increase will be interpreted as success and any disappointing result will produce another theory.

You might want to learn whether existing customers are postponing purchases. Compare their timing around the promotion with comparable periods and look for weakness before or after the offer.

You might want to know whether the promotion attracts genuinely new customers. Identify who is new, then observe whether they return and how they buy.

You might suspect that urgency matters more than the price reduction. Look at when customers act and what they say about the deadline. Where a genuine non-price deadline exists, compare the response rather than assuming a discount is the only way to create movement.

You might want to understand incremental margin. Estimate which purchases were genuinely additional, what margin they produced and what margin was surrendered on purchases likely to happen anyway.

Keep the comparison proportionate to the decision. A small business does not need a research department before sending an offer. It does need to stop running promotions that answer no question beyond “will the sales line move this week?”

Change as little as practical when the purpose is learning. If the price, product, audience, deadline and sales message all change together, the result will be difficult to interpret.

Record what you expected to happen. Record what actually happened. Include the period before and after the promotion, not just the days when the chart looks exciting.

Then decide what the result supports.

The business was quiet. A discount appeared. Sales jumped. The discount disappeared. Sales fell.

At the beginning, that pattern seemed to prove that customers only buy on sale.

It does not.

Which customers bought? Which sales were additional? Which purchases moved in time? Did the lower price make the offer affordable, more worthwhile or merely more urgent? Was demand already rising? Did the extra revenue leave enough value behind?

A discount is a change to the offer. Treat the response as evidence, not a verdict.

The dangerous discount is not necessarily the biggest one. It is the one whose job you cannot explain.