Them: "The sale crushed it. We did double our normal week."

Me: "What did the month after look like?"

Them: "...why?"

That conversation, or a version of it, has happened in more rooms than I can count.

Here's what a great sales week hides. Say a 25%-off week does $120K instead of your normal $60K. Feels like found money. But some of those buyers were paying full price until you offered not to charge it - margin handed back for nothing. Some were buying next month anyway and simply moved up, so next month opens soft. The genuinely, really, honestly new customers (the ones the sale was for) are usually the smallest group of the three.

Run that math and the "extra" $60K often nets out to a fraction of its headline. That's before counting the slowest cost of all.

A Stanford study found customers who get a deep discount become 22% more responsive to every future discount. Read that carefully - not more loyal, more responsive to discounts. The promo doesn't just move product. It trains people. And each sale has to be a little deeper than the last one to produce the same spike.

Before you conclude the answer is to never discount - the most famous attempt at that went worse. In 2012, JCPenney's new CEO killed coupons overnight in favor of 'honest everyday pricing'. Revenue fell roughly 25% in a year. He was out in 17 months. Once customers are trained, quitting cold turkey is catastrophic too.

So the lesson isn't "promos are bad." It's that a promo is a loan from your future self, and most brands never check the interest rate.

Before your next sale, write down what next month looks like if it works. If you can't say which sales will be new, which are borrowed, and which are margin you gave away - you're not running a promotion. You're running an experiment without a control group.