Is That Sale Real? How Price History Exposes Fake Discounts

When the UK consumer group Which? tracked 175 home, tech and health appliances across eight major retailers for a full year, it found something worth remembering the next time a banner promises 50% off: on Black Friday itself, not a single one of those products was at its cheapest price of the year. A discount is a claim about the past, and price history is the only thing that can check it. This guide shows you how fake discounts are built, what a "was" price is legally required to mean where you shop, and four checks that tell you whether a sale is real before you spend anything.

What You'll Learn

• How a discount that isn't one gets manufactured
• What a full year of tracked prices actually showed
• What the "was" price legally has to mean in the EU, UK and US
• Four checks that expose an inflated discount, three of them in about two minutes
• How to have your own price history ready before the next sale starts

How a Fake Discount Is Built

A genuine sale and a staged one look identical on the page. Both show a struck-through number, a lower number, and a percentage. The difference is entirely in what happened before you arrived. These are the patterns that show up again and again:

• The inflated reference price. The price goes up shortly before the sale, then comes down to roughly where it always was. The US Federal Trade Commission describes exactly this: an "artificial, inflated price was established for the purpose of enabling the subsequent offer of a large reduction."
• The brief high price. A product sits at its normal price for months, jumps for a short window, and that window becomes the "former" price. Nothing about the number is invented — it just never really applied.
• The stale reference. The "was" price was genuinely charged, and charged for a while — months ago. It stopped being the price long before the banner went up, and nothing on the page tells you when it last applied.
• The permanent sale. Some listings are discounted essentially all the time. When the sale price is the real price, the "was" figure is decoration.
• The list-price comparison. The saving is measured against a manufacturer's recommended price that no retailer actually charges.
• The resetting deadline. A countdown timer creates urgency, then restarts the following week with the same "final hours" copy.

Drawn out over ninety days, that stale reference looks like this. The line is what the product actually cost, recorded every time the price moved. The dashed line is the "was" price the sale advertises against.

E

Example 27-inch 4K Monitor

https://example.com/p/27-inch-4k-monitor

$289.00advertised was $449

was $449

Low $279.00High $449.00

Price ranged from $279.00 to $449.00 across 9 recorded changes. The advertised reference price was $449.00. An alert was sent on 2026-08-04, when the price moved to $279.00.

A sale advertising "was $449" here is not inventing a number — $449 is in the record. It is quoting a price that stopped applying three months ago, which is precisely the case the FTC calls out: a former price "not used in the recent past but at some remote period in the past, without making disclosure of that fact." Meanwhile the drop worth having — $349 down to $279, on an ordinary Tuesday in August — carried no banner at all. That is the one the red marker is on.

What a Year of Real Prices Showed

Which? published its analysis(opens in a new tab) on 25 November 2025, covering 175 products from eight retailers tracked from May 2024 to May 2025 — six months either side of the sales period it reviewed, 15 November to 12 December 2024.

Two findings are worth keeping apart. Measured against that sales period, 83% of products were cheaper or equal in price on at least one occasion during the year. Measured against the Black Friday day price alone, the figure is 100% — every product in the analysis was cheaper or the same at some point in the six months either side. And in Which?'s own words, "there were no deals in our analysis that were at their cheapest price of the year on Black Friday 2024."

Which? published a per-retailer breakdown for most of the eight. The share of products that had been cheaper or the same price at other times:

• John Lewis — 94%
• Very — 93%
• Amazon — 88% (outright cheaper in 63% of cases)
• AO — 85%
• Argos — 79%
• Richer Sounds — 55%

Currys is the exception, and it cuts the other way. After earlier Which? investigations it committed that no Black Friday deal would have been cheaper in the previous six months, and Which? found it kept that promise: every Currys product in the analysis was at its cheapest on the day compared with the six months before it — though all were cheaper or equal again in the six months after. Boots was in the study but not individually quantified.

The concrete examples are more instructive than the percentages, and all come from the same analysis. A Samsung Jet Bot robot vacuum was £350 on Black Friday at John Lewis, having sold for £299 across May and June 2024. A Samsung TV was £1,039 on Black Friday at AO and later dropped to £759. A Dyson V11 was £349 at Very during the sale — and the same £349 on 29 days before it and 41 days after, which Which? puts at 32% of its analysis period.

None of this necessarily means anyone broke a law. Prices move for ordinary reasons: stock levels, competitor moves, model refreshes. That is precisely the point. Because prices move constantly, the sale-day price is rarely the floor — and a discount badge is not evidence that you are near it.

What the "Was" Price Legally Has to Mean

Regulators have been closing in on this, and the rules differ enough by region that it is worth knowing which one applies to you.

European Union. The Omnibus Directive (2019/2161)(opens in a new tab) added Article 6a to EU price-indication law, applicable since 28 May 2022. Whenever a trader announces a price reduction, it must state the prior price — defined as "the lowest price applied by the trader during a period of time not shorter than 30 days prior to the application of the price reduction" — and the discount must be calculated against that figure. Member states may set different rules for goods liable to deteriorate or expire rapidly, and a shorter period for products on the market for less than 30 days. They may also require that a progressively deepened discount measure itself against the price in force before the first markdown — closing off the ladder that would otherwise let each step quote the one before it.

United Kingdom. The Digital Markets, Competition and Consumers Act 2024 lets the Competition and Markets Authority rule that consumer law has been broken and impose penalties directly, instead of litigating through the courts first. On 18 November 2025 the CMA published its final price transparency guidance(opens in a new tab) and opened its first investigations(opens in a new tab) under those powers into eight businesses, with advisory letters to 100 more across 14 sectors. In April 2026 two of those eight became the regime's first financial penalty: the AA and BSM driving schools were ordered to refund learner drivers(opens in a new tab) more than £760,000 over a £3 booking fee that was not shown upfront, and fined £4.2 million.

United States. There is no 30-day rule. The FTC's Guides Against Deceptive Pricing (16 CFR 233.1(opens in a new tab)) instead require a former price to be bona fide — "openly and actively offered for sale, for a reasonably substantial period of time, in the recent, regular course of his business". A former price "is not necessarily fictitious merely because no sales at the advertised price were made"; it becomes fictitious when it existed in order to set up the markdown.

Four Checks Before You Buy

1. Look for the floor, not the percentage

Ask one question: what is the lowest this has been recently? If you can answer it, the discount badge stops mattering. If you cannot, the badge is the only evidence you have — and it was written by the seller.

2. Widen the window past 30 days

Where a 30-day reference price is shown, treat it as a starting point rather than an answer. The interesting question is what the price was two and three months ago — the stretch a briefly inflated "was" price is designed to sit outside of.

3. Compare across retailers at the same moment

A staged discount is usually one retailer's decision. Open the same model number at two or three other stores, including the manufacturer's own shop. If everyone is within a few percent of the "sale" price, the sale is mostly typography.

4. Watch what happens after the sale ends

This is the check almost nobody runs, and it is the most revealing — the one that costs patience rather than minutes. If the purchase can wait, let the sale end and keep the product on your radar for two or three weeks. If the price returns to the sale figure — or drops below it — you have learned what that product actually costs, and the next banner has nothing left to tell you.

What You Can Check Right Now, and What You Can't

Being straight about this matters more than a neat pitch: if you never tracked a product, nobody can hand you its complete price history. Amazon listings are well served by third-party price trackers with their own historical databases. For most of the internet — regional retailers, specialist shops, direct-to-consumer brands — that data was never collected by anyone, and no tool can reconstruct it after the fact.

So for a purchase you are weighing today, work with what exists: the retailer's legally required reference price, a cross-retailer comparison, the manufacturer's own store, and a search on the exact model number rather than the product name.

Then fix the real problem — that you were relying on the seller's account of the past. That fix is not something you do during a sale. It is something you do before one.

Build Your Own Price History Before You Need It

The reliable way to know a product's price history is to have been watching it. That sounds like effort; in practice it is a couple of minutes per product, once.

AlarmBot is a price tracker that lives in your browser: it watches a product page and records the price every time it changes. Install it, open the product page, and add a track the first time you start considering something — weeks before you intend to buy. By the time the banners appear, you are not reading the retailer's claim about the past. You are reading your own record of it.

A few things worth knowing in practice:

• History starts when you do. There is no retroactive lookup. The timeline builds forward from the moment you add the track, which is why adding it early is the whole trick.
• The free plan keeps 90 days of change history for each tracked item — three times the EU's 30-day reference window, which is the point: 30 days is exactly the span a staged "was" price is built to hide behind.
• You see the shape, not just the number. Each track records its lowest and highest values and how far the price has moved since you started, so the floor is something you look up rather than guess.
• You choose what is worth an alert. Set a rule like "drops more than 15%" and the product stays quiet until something meaningful happens.
• It works on any store. AlarmBot reads the page directly rather than relying on retailer partnerships, so niche shops are tracked the same way major marketplaces are — see how tracking prices on any website works.

A Buying Routine That Beats the Banner

Put together, the habit is small:

1. Track on interest, not on intent. The moment a product enters your consideration set, add it. You are not committing to buy; you are starting the clock.
2. Group what you are watching. Organise tracks into lists — a kitchen renovation, a camera setup, back-to-school — so a category is one glance rather than ten.
3. Set the rule that matches how you shop. A percentage drop for things you will buy eventually, a target price for things you will only buy at a number. Price drop alerts 101 covers how to pick.
4. Route alerts where you will see them. Browser, email, or a Telegram or Discord channel.
5. Check the record before you check out. Ten seconds with your own history beats any badge on the page.

Common Questions

Can I look up a product's price history if I never tracked it?

Not in the general case. Amazon products are covered by third-party history services, and some regional comparison sites publish historical charts. For most stores that data does not exist anywhere. This is why starting early matters more than any single feature.

Is a fake discount illegal?

It depends on where you are and how it was constructed. In the EU, a reduction must as a general rule be measured against the lowest price of the previous 30 days; quoting a higher figure breaches the Omnibus Directive as transposed, with narrow exceptions for perishables and products newer than 30 days. In the US, a "former price" that was set up to enable the markdown is deceptive under the FTC guides. But a retailer can also stay entirely within the rules and still show you a discount that is not a good price.

How long should I track before buying?

Long enough to see the price move at least twice — for most electronics and appliances, a few weeks. Ahead of a major sale event, six to eight weeks puts your baseline outside the period when reference prices tend to be adjusted.

Do I need to catch the absolute lowest price?

No, and chasing it is how people end up not buying anything. The goal is to avoid paying meaningfully above the recent floor. Knowing the floor exists — and roughly where it is — does almost all of the work.

The Discount Is a Story. The History Is the Record.

Retailers are not obliged to tell you where a price has been, only what it was under a narrow legal definition, for a narrow window, in some jurisdictions. That is the whole reason "is this sale real" is a question worth asking at all — the information you need to answer it is the one piece the page does not show you.

You can keep guessing, one banner at a time. Or you can spend two minutes adding a track today, and let the next sale prove itself against a record you own.

---

Brand, retailer and product names in this article are the trademarks of their respective owners and are used for identification and commentary only. AlarmBot is not affiliated with, endorsed by or sponsored by any company named. Pricing figures are reported from the sources cited and describe the periods those sources analysed; the chart above uses illustrative data and does not represent a real product.