Price promotions: why 75% can't say which discounts pay

October 9, 2026 · 9 min · Free, with full transcript

Brands spend about twice as much on price promotions as on paid media, yet 75% of 250 UK consumer goods leaders surveyed by Censuswide for Les Binet and Kantar's Dom Boyd can't estimate which promotions make money. The episode weighs practitioners' faith in discounts against margin data, including a Nielsen analysis where nearly two thirds of 300 million promotions lost money, and argues for adjusting promotions rather than abolishing them. You'll learn how the promotion dependency spiral erodes pricing power, why judging promotions over less than 13 weeks hides the hangover, how econometrics puts promotions and advertising on the same basis, and how to bring your CFO a list of loss-making promotions to cut.

Listen to the podcast

8 min

Chapters

Key takeaways

  • Ask finance or revenue management for last year's ten largest promotions, with each one's cost and sales during and for eight weeks after it ended.
  • Judge promotions over 13 weeks or more, including the dip afterwards, instead of the one-week spike.
  • Propose cutting only the bottom slice of promotions you can show lost money, and keep the ones that worked.
  • Use econometrics to hold price, promotions and advertising to the same standard of evidence.
  • Keep the Black Friday deals you have committed to, but decide now how you will measure them.
Read the full transcript

Host:Welcome back to MBA Training. Price promotions cost twice paid media, and 75% can't say which pay, and why it matters this week. On Wednesday, at the IPA Effectiveness Conference in London, Les Binet compared price promotions to class A drugs. He said it to a room full of people who run promotions every month. What did the research find?

Expert:The headline is that brands spend about twice as much on price promotions as on paid media, which means the advertising you buy. That comes from the study Binet did with Dom Boyd, Kantar's chief strategy officer, as Marketing Week reported it. Promotions are now the biggest single line in marketing spend. And 75% of the marketers surveyed say they can't estimate what share of those promotions make money.

Host:Who did they ask?

Expert:The research firm Censuswide surveyed 250 senior marketing and insight leaders at UK consumer goods companies. LBB says the sample included board members and top executives in food and drink, durables and other categories, and every respondent was directly involved in promotional decisions. These are the people who sign off the deals.

Host:Let me push on that. Three quarters can't say which promotions are profitable. Isn't that marketers admitting they don't do their jobs?

Expert:Partly, yes, and it gets worse. The Media Leader reports that 76% repeat promotions even when nobody has checked whether they were profitable. Marketing Week reports that 81% judge promotions over less than 13 weeks. So they measure the spike and almost never the hangover. I wouldn't put all of it on marketing, though. In many consumer goods companies the promotional budget sits with sales or revenue management, and it gets negotiated with retailers. Marketing often doesn't own it and doesn't see the P&L, the profit and loss account, line by line. That's my experience. The study doesn't say it.

Host:What's the hangover, in plain terms?

Expert:Profit leaks out three ways. Some promoted volume comes from people who would have paid full price anyway. Some is pulled forward from next month. Some just moves from one retailer to another. LBB sums up the study as saying much of that volume isn't incremental, meaning you'd have made those sales without the discount. The respondents know this. 80% said many promotional buyers would have bought anyway.

Host:They know it and they keep doing it.

Expert:And 68% say they'd prefer to run fewer promotions. To me that's the most interesting number. People aren't unaware. They can see the trap and they can't get out of it.

Host:Why can't they?

Expert:The respondents gave the reasons themselves. They named consumer expectations, short-term volume and revenue targets, defending market share, retailer pressure, habit inside the organisation, and a lack of reliable evidence. Binet and Boyd call the result a promotion dependency spiral. You discount, your price looks less believable, and your pricing power falls. Pricing power is your ability to raise prices without losing much volume. Then you need more discounts to hit the number.

Host:Here's the awkward part. Respondents rated brand strength as less effective than cutting prices. Maybe the people closest to the tills are right and Binet is wrong.

Expert:That's the real disagreement. The Media Leader says respondents rated brand strength 1.3 times less effective than lowering prices at driving revenue and profit growth, and pricing power 1.5 times less effective than promotions. On the other side, Boyd's separate analysis found that brands with the strongest pricing power earn margins 1.5 times higher than the weakest. The Media Leader also cites a Nielsen analysis of 300 million price promotions in which nearly two thirds lost money. So practitioners' instinct points one way and the margin data points the other. I'd go with the margin data, but it's a different kind of evidence. One is what people believe. The other is what happened.

Host:Give me the strongest case for promotions.

Expert:There's a good one. First, the short-term lift is real. Marketing Week says plainly that promotions can lift short-term sales. If you have a quarterly volume target, stock close to expiry, or a competitor running a deal next to you on the shelf, a promotion is the fastest lever you have. Second, retailer pressure is real. If a retailer expects you in its promotional calendar, pulling out can cost you shelf space. Third, this is a self-reported survey of UK consumer goods leaders. It tells you what 250 people believe and do. It isn't an audit of their accounts. The coverage I read also doesn't explain how the "twice paid media" figure was calculated, so I'd read it as a direction, not a precise ratio.

Host:And the people presenting it have something to sell.

Expert:That's fair to raise. Kantar sells measurement, and one of the recommended steps is econometrics. I don't think that invalidates the findings, because the Nielsen figure is independent of them. Still, you should know who is telling you to buy better measurement.

Host:Then what is econometrics, and why does it matter here?

Expert:Econometrics, often called marketing mix modelling, is statistics that separates out what drove your sales: price, promotions, advertising, weather, distribution. Binet and Boyd want price, promotions and advertising judged on the same basis. Most companies hold advertising to a tough standard and approve promotions because sales spiked.

Host:What does their plan say?

Expert:According to LBB there are six steps: improve marketing training, learn promotional best practice, use econometrics consistently, measure short and long-term effects, assess pricing power, and move money from ineffective promotions into brand building. Binet's version for Marketing Week is shorter. Find the promotions that lose money, cut them, and put the money into brand advertising.

Host:That sounds easy on a slide. A CMO goes to the CFO and says let's cut promotions. The sales director tells the CEO we'll lose share at our biggest retailer. Who wins?

Expert:Usually the sales director, if you show up with an opinion. You win if you show up with a list. Don't propose cutting promotions in general. Propose cutting the bottom slice you can show lost money, and keep the ones that worked. That's a conversation about margin, and CFOs like those. Laurence Green, director of effectiveness at the IPA, the UK trade body for agencies, described the research as being about judging marketing on profitability and long-term effectiveness. Use that language in the room.

Host:The study covers UK consumer goods. Why should a B2B or direct-to-consumer marketer care?

Expert:Because discounts come in many forms: free months on a subscription, a 20% code for signing up to the newsletter, end-of-quarter deals in software sales. They work the same way. You subsidise sales you'd have made anyway, you pull demand forward, and customers learn to wait. The study doesn't cover those sectors, so this is my own extrapolation. But you can ask what share of your discounts make money in any business.

Host:Black Friday is seven weeks away. Is this bad timing?

Expert:It's good timing. Most of you have already committed to Black Friday deals, and I wouldn't cancel them now. But you can decide today how you'll measure them. Judge them over one week and you'll see a spike and call it a win. Judge them over 13 weeks or more, including the dip afterwards, and you may find a different answer.

Host:Last one. What's the risk if someone takes this too literally?

Expert:They cut promotions across the board, lose distribution, miss a quarter, and the effectiveness argument loses credibility inside their company for years. Binet himself told Marketing Week back in 2022 that adjusting promotions could be "immensely profitable". He meant adjusting them. He didn't mean abolishing them.

Host:So what should a listener do this week?

Expert:Ask your finance or revenue management team for last year's ten largest promotions, with what each one cost and sales for the eight weeks after it ended as well as during it. If nobody can produce that table, you're in the 75%. If they can, mark the two that lost the most money and take them to your next planning meeting as candidates to cut.

Host:Drawn from Marketing Week, LBB and The Media Leader, links in the show notes. Done. Want to know where you actually stand? Take the CMO self-assessment at mba-training.com.

Read and go deeper