A record sales day tells you almost nothing about whether you made money

2026-07-01

Big sales events bring in big revenue, and those numbers often get celebrated. But we should look at them more carefully. A record day built on deep discounts and smaller orders can actually lose money on many sales.

  • Ecommerce
  • Pricing
  • Margin
  • Strategy

Big sales events bring in big revenue, and those numbers often get celebrated. But we should look at them more carefully. A record day built on deep discounts and smaller orders can actually lose money on many sales. The important number is not total sales, but what remains after discounts, returns, fees, and fulfilment.

What Prime Day just showed

Amazon's Prime Day 2026 took place from June 23 to 26 and set a new record. Adobe Analytics reported 26.4 billion dollars in US online spending across all retailers during those four days, up about 9 percent from 2025. That makes for a simple headline.

Looking deeper, the picture changes. On the first day, Numerator reported the average order value was 48.36 dollars, down about 17 percent from last year, and average household spend was 89.04 dollars, down 16 percent. There were more orders, but each was smaller. The record came from deeper discounts and more orders, not from customers spending more. That difference is key.

Why the topline flatters you

Revenue is the most flattering number a business can report, because it comes before all the extra costs that discounts add. Let’s follow one discounted order all the way through.

You start with a lower gross margin, since the discount reduces it right away. Then subtract the payment processing fee. Next, take off fulfilment and shipping costs, which stay the same even if the item is cheaper. Add the cost of getting that order, which is often higher during a big sale. Finally, consider the orders that get returned, which bring extra fulfilment costs, reverse logistics, and likely another markdown.

It is very possible to have your best revenue day ever and still lose contribution margin on many of those sales. The orders were real, but the profit was not.

The behavioural cost of teaching the discount

There is another cost that is not obvious right away. Regular, heavy discounting teaches your customers to expect lower prices. This is called anchoring: people compare prices to the last one they saw. If your sale price becomes the anchor, your regular price starts to seem too high, and fewer people buy at full price.

Once customers learn to wait for sales, your growth depends on running more events. You have not built lasting demand; you have only borrowed it, and the cost keeps rising. Brands that discount on a schedule often cannot sell at full price between those sales.

What disciplined operators steer by

Focus on contribution margin per order, not GMV. After a sale event, the main question is not how much you sold, but how much you kept per order after discounts, returns, fees, and fulfilment. If that number is low or negative for some orders, the event lost money there, no matter what the headline says.

Use a thoughtful promotion plan, not just blanket discounts. Adjust depth, eligibility, and exclusions as needed. Offering a targeted deal on slow-moving stock to a specific group is a smart move. Giving a sitewide discount on everything usually just gives away margin without a real strategy.

Protect your full-price sales. Strong businesses track how much they sell at full price and work to keep that number high. Sales events should clear out old stock and attract valuable customers, not take away sales from people who would have paid full price.

Look at what customers do after the event, not just the revenue during it. The real test of a discount-driven campaign is whether those customers return. If they only come back for the next sale, you bought revenue, not loyal customers.

The leadership point

This does not mean discounting is always bad. Sales events help clear inventory, bring in new customers, and create helpful demand spikes. But you need to measure the right things. When someone shares a record revenue number, the important questions are: what was the contribution margin per order, what happened with returns, and how did it affect full-price sales in the following month?

A record is only worth celebrating if you kept some of the profit.