Why Frequent Promotions Are Becoming a Legal Problem as Well as a Pricing One

What has changed in the law on price claims?

Since May 2022, a retailer in the EU that announces a price reduction must show the lowest price it charged in at least the 30 days before. In September 2024 the Court of Justice of the European Union ruled, in a case against the German discounter Aldi Sud, that the announced reduction itself must be measured from that lowest price.¹ Aldi Sud had advertised bananas at 1.29 euros a kilo with 1.69 struck through and "-23%", while its lowest price in the previous 30 days had been 1.29. Measured from that price, the reduction was zero; the German court decides the case itself.³

Britain has no 30-day rule. On 30 July 2026 the High Court in London refused a request from the Competition and Markets Authority (CMA) to require the mattress seller Emma to sell at least one product at its advertised full price for every two sold at a discount. Emma had admitted some breaches, and the terms of any order await a further hearing. The judge did accept that "the longer the mattress is advertised" at the discount, "the less realistic" the advertised full price becomes.⁴ On 9 August 2026 the government announced a consultation for autumn 2026 on whether to add "fake 'was' prices, invented discounts and misleading recommended retail prices (RRPs)" to the practices banned outright.²

The retailer decides the shelf price and makes the saving claim, so under the rule the claim is the retailer's. The manufacturer decides which deal weeks it funds and what recommended retail price it publishes, a direction the retailer is free to ignore.

Regulators now ask whether the regular price was really charged, and the calendar of deals a manufacturer funds goes a long way to deciding the answer.

Why do frequent deals lower the price shoppers expect?

The Academy's lesson on price thresholds treats the reference price, the price a shopper expects to pay, as a running average of the prices the shopper sees, in what it calls the adaptive reference model.

To find the average price a shopper sees, multiply the share of weeks on deal by the average depth, and reduce the regular shelf price by that fraction. So that one number, the share of weeks on deal times the depth, sets how far below the regular price a shopper's expectation sits. The Academy's reference price gap measures how much more expensive the regular price then feels: the actual price less the reference price, divided by the reference price.

The price shoppers expect settles at the average price they see, so the share of weeks on deal times the depth decides how expensive the regular price feels.

What does a busy calendar cost the brand?

Take an invented biscuit pack that sells 100,000 a week at the regular price. Its manufacturer sells it to the retailer at a list price of 3.00 euros. The retailer's regular shelf price is 4.00 euros and its deal price 3.00 euros, 25 percent off, both the retailer's decision. In deal weeks the manufacturer funds the 1.00-euro cut on every pack sold, so it receives 2.00 euros a pack; prices exclude value-added tax, and the funding split is this example's assumption.

The manufacturer compares two calendars of deals it could fund over a year. Calendar A has 13 two-week deals with two full-price weeks between them, so half the year is on deal. Calendar B has seven two-week deals with at least five full-price weeks before each, so 14 of 52 weeks, or 26.9 percent, are on deal.

The share of weeks on deal times the depth is 0.125 for calendar A (half the weeks times 25 percent) and 0.067 for calendar B (26.9 percent times 25 percent). The shopper's reference price therefore rests at 3.50 euros under A (4.00 times 1 minus 0.125, or 4.00 times 0.875). Under B it is 3.73 euros (4.00 times 0.933). At the 4.00-euro shelf price, the reference price gap is 14.3 percent under A (0.50 euros above 3.50, divided by 3.50) and 7.2 percent under B (0.27 above 3.73, divided by 3.73).

The Academy's lesson on baseline and incremental volume uses the same number to model what it calls reference price decay: how the full-price baseline, the volume sold in weeks with no deal, wears down month by month.

At 0.06, the baseline falls by 0.75 percent a month under calendar A (0.06 times 0.125) and by about 0.40 percent under calendar B (0.06 times 0.067). Each month's fall compounds. After 24 months the baseline index, where 100 is today's level, stands at 83.5 under A (0.9925 to the power of 24) and 90.7 under B (0.99596 to the power of 24, using the unrounded 0.404 percent).

The 7.3-point gap in the baseline is worth about 1.13 million euros a year of the manufacturer's sales at the month-24 rate (7.27 percent of 100,000 packs a week for 52 weeks, at the 3.00-euro list price). The Academy's public Reference Price Erosion Projector values it the same way. It is a rough size rather than a forecast: it is revenue, not profit, and deal weeks, which sell several times a normal week, can lose more packs than it counts.

Calendar A's six extra deals also sell more packs, so whether calendar B earns more depends on what each deal week earns on the manufacturer's own margin. The baseline lesson puts the ceiling on deal weeks where the full-price volume lost over two years is worth more than the profit the extra deal weeks bring in.

A busy calendar has two costs that event reports do not show: a regular price that feels more expensive, and full-price volume that wears down faster. Both have to be weighed against what the extra deals earn.

When can a deal still be called a saving?

In the EU, a deal can be advertised as a reduction from the regular price only if the regular price applied for the 30 days before it. A deal lasting D days, with a saving from the regular price every time, can therefore run for at most D divided by (D plus 30) of all days.

If every deal starts on the same weekday, as flyer-led promotions often do, the 30-day gap rounds up to five full-price weeks, or 35 days. On that cycle, two-week deals that are each advertised as a cut from the regular price can fill at most 28.6 percent of weeks (14 days in every 49, or 2 divided by 7). That is slightly under the 30 percent the baseline lesson keeps as its planning ceiling, with no measurement behind it. The law itself sets no limit on deal weeks: it governs what the retailer may announce, and longer deals can fill a larger share. Member states may also set different rules for goods that perish quickly. The rule covers announced price cuts; the Commission places multi‑buys such as "buy one, get two" outside it, though still under the general law, so retailers may lean on those mechanics too, and a brand can plan the mechanics it funds with that in mind.⁵

The 30-day rule favors fewer, longer deals, because each deal needs only one 30-day gap before it. Long deals meet a second test in both the EU and Britain. The European Commission says the fairness of very long reductions, set against the time at the full price, is still judged under the general law.⁵ In the Emma case, the seller accepted that an advertised full price misleads unless the product was offered at it for at least as long as at the discount.⁴ A deal no longer than the 30 full-price days before it passes both tests; a longer one can pass the 30-day rule and still fail the second.

In the EU the gap before each deal now decides whether there is a saving to advertise, and on a weekly cycle a two‑week deal needs five full‑price weeks before it.

How much actually sells at the full price?

Time at the full price is not the same as sales at the full price. The Emma judgment printed both for four mattress lines in the six months to February 2024. Emma Hybrid sat at its full price 93.31 percent of the time but sold only 3.73 percent of its units there.⁴ These are mattresses sold online in Britain, and the court refused to turn a ratio of units into a fixed rule.

The manufacturer's own anchor is its recommended retail price. In the EU a retailer's comparison with that price falls outside the 30-day rule but remains fully subject to the general law on unfair practices, and a comparison that shoppers read as a price cut can breach the 30-day rule as well.⁵ The Academy's price-thresholds lesson makes the same point as a pricing rule: an anchor price is "only fair, and in many markets only lawful" when it is one the pack really sells at. In the EU and Britain, competition law forbids a manufacturer from making retailers charge its recommended price, so what remains in its hands is the recommendation itself.

A recommended retail price that almost nothing sells at is a weak anchor for shoppers and a weak basis for any retailer's comparison, and Britain may ban misleading ones. The manufacturer should check its own recommendation against real sales.

Is it worth funding fewer deals?

Cutting deals carries a risk. In January 2025 Emma began working to the CMA's one-for-two ratio. Its own analysis, cited in the judgment, says it had sold seven discounted mattresses for every one at full price in 2024. In the first three months of 2025 it sold one discounted mattress for every two at full price. The same analysis says its competitors had not changed their approach and that traffic to Emma's website fell. Emma estimated that its net revenue for the quarter fell by 40 to 60 percent. The judge could not find that the change caused the fall but said "there must be a risk that this was the cause".⁴

Emma changed its whole approach within a quarter, in a category where, the judge found, most shoppers can put off buying until a deal comes. The Academy's approach is gradual: cut the funded deal weeks first on the lines promoted most, expect sales to dip when the deals stop, get that dip approved before starting, and judge the recovery over 18 to 24 months. Calendar B in the example still funds seven deals a year. Part of the money saved can go into support that is not a price cut, such as a display or a place in the retailer's flyer, where a retailer sells that space on its own. That keeps the brand visible without teaching shoppers a lower price, and it needs no saving claim. The retailer can still run and fund any promotion it chooses.

The calendar a manufacturer funds now largely decides how much of the brand's volume sells at the regular shelf price, and in the EU whether the retailer has a saving to advertise, provided the retailer runs no deal of its own in the gaps.

A manufacturer that funds fewer deals should cut gradually, move part of the money into non-price support, and judge the result over two years rather than one quarter.

References