Promotions Get Repeated Without Proof of Profit. Marketing Should Treat the Repeat as a Decision.

A new survey of UK consumer-goods leaders finds that three in four cannot estimate what share of last year's promotions made money, and about as many repeat promotions whose profitability has not been assessed.

19 October 2026 · 4 min read

What happened

On 7 October 2026 the Institute of Practitioners in Advertising (IPA) published research by Les Binet (visiting professor at Ravensbourne University) and Dom Boyd (Global Chief Strategy Officer at Kantar), presented at the IPA Effectiveness Conference, on how UK consumer-goods businesses judge their price promotions.

The survey was conducted by Censuswide on behalf of Binet and Boyd among 250 senior marketing (60%) and insight (40%) leaders at UK consumer-goods businesses, all directly involved in promotional decision-making, planning or evaluation. It was fielded from 18 to 26 August 2026.

Three in four (75%) could not estimate the proportion of their promotions that had been profitable across the previous year. Roughly the same share (76%) said they repeat promotions even when profitability has not been assessed.

The IPA release adds that spending on price promotions now exceeds investment in paid media advertising, at around twice its size.

What the evidence actually says

Observed fact. A disclosed sample: 250 senior leaders, UK consumer goods, fieldwork in late August 2026, published by the IPA on 7 October and reported by Research Live on 8 October.

Reported claim. 82% believed their latest promotion was profitable, but more than 60% of those assessments rested on personal judgement or partial analysis rather than formal event-level profit-and-loss measurement. 81% evaluate promotions over less than 13 weeks. 68% say they would prefer to run fewer promotions.

Inference (ours, in line with the authors). Near-term sales movement may be standing in for incremental profit. The authors argue that evaluation windows under 13 weeks are too short to see it; the survey does not test it.

Unknown / limitations. These are respondents' own accounts, not audited financial results. The sample is UK consumer goods, not marketing teams everywhere. The questionnaire and subgroup cuts are not public. The wider claim that nearly two-thirds of price promotions lose money comes from the earlier evidence the authors reviewed, not from this survey.

Why this matters for marketing

A promotion crosses brand, revenue, media, channel and finance. A discount can lift volume while paying customers who would have bought anyway, pulling purchases forward, cannibalising full-price sales or teaching shoppers to wait for the next offer. A volume uplift or a redemption rate can be accurate and still not answer whether the promotion made money.

The risk compounds when the promotional calendar becomes a template. A promotion that looked successful is carried into next year's plan, and after a few cycles the plan remembers the decision to repeat but not whether anyone ever checked it paid back.

Respondents themselves name the reasons promotions keep coming back: consumer expectations, short-term volume and revenue targets, market-share defence, retailer pressure, organisational habit and a lack of reliable evidence. In UK consumer goods a promotion is rarely a marketing decision alone: it is negotiated with retailers and owned partly by sales or revenue management. That makes the missing evidence harder to see, not less important.

The operating implication

Make the repeat an explicit decision. Before a promotion goes back into the calendar, the marketing lead, the commercial or revenue-management owner and a finance partner should agree what incremental outcome they expect after discount, trade terms, media and fulfilment costs, against what baseline — and how it will be measured: the authors point to econometrics that evaluates price, promotions and advertising on the same basis, over both short and long horizons.
Give unproven repeats a longer review. If last year's promotion was never evaluated beyond a few weeks of sales, run the repeat with a review window that covers the post-promotion dip and full-price sales after the offer ends, rather than the windows of less than 13 weeks most respondents use.

The non-move: do not cut promotions across the board on the strength of one survey. The question it raises is which repeats were ever proven, not whether promotions belong in the mix.

The Olymuse lens

This sits under Learning. Olymuse is the marketing operating layer above the stack a team already runs; it does not calculate the incremental profit of a promotion, and this survey says nothing about what software can do.

In What Is Process Intelligence in Marketing? we argued that before the next step a team should know what should already exist for it and what is missing. For a repeated promotion, the thing that should already exist is the evidence that it paid back.

Our point of view: repeating a promotion is a new decision, not a calendar default. A repeat should inherit the evidence of the last one, and if that evidence was never produced, the repeat should say so.

What we would watch next

First, whether the authors publish the questionnaire and subgroup cuts, so the finding can be tested across types of promotion and retailer.

Second, independent studies that compare promotion-level incremental profit with how teams reported the same promotions at the time.

Third, whether consumer-goods businesses start writing down a baseline and a longer review horizon before a promotion is repeated.

Read next: What Is Process Intelligence in Marketing? · Olymuse is in founder-assisted private alpha. Apply for the Private Alpha →