Start with the misconception: a big percentage off tells you how good a deal is. It does not. A percentage is arithmetic applied to a reference price, and the seller chooses the reference price. If that number is a price the item rarely or never sold at, the percentage is decoration. The Competition Bureau has a name for the practice, fake ordinary selling price, and in a news release dated November 25, 2025 it warned that inventing a higher regular price to make a sale look like a bargain violates the Competition Act. Its example is blunt: a business cannot advertise an item as $100, now $50, if it never actually sold at $100.
The number to trust is the sale price itself, compared against what the item has actually been selling for. Every step that follows is a way to get that comparison right.
Three reference prices, three different checks
Not every crossed-out number is the same kind of claim. The Competition Bureau notes that businesses use two types of regular prices as a reference for claiming savings, and large marketplaces add a third. Each type fails in its own way, so each one gets checked differently.
| Reference price | What it claims | How it can mislead | The shopper check |
|---|---|---|---|
| The seller's own regular price ("our regular price $100, now $50") | This store genuinely sold the item at that price | The item sat at that price briefly, or never, while spending most of the year on sale | Price history at that retailer: how long, and how often, the item actually sat at the regular price |
| A market or list price ("list price $100, our price $50") | The wider market, or the manufacturer's suggested price, is that high | The list price is real on paper but few sellers charge it, so the saving is measured from a price nobody pays | Compare the exact same model number across several stores and compare selling prices, not list prices |
| A typical price (shown next to some marketplace deals) | Customers have recently paid about that much on that marketplace | Recent can be a short or favourable window, so the typical price can sit above the longer run average | Open the longer price history, 90 days where available, and compare the sale price with the average, not only with the typical figure |
One live example of the first type, checked on November 4, 2026: the Roborock Canada official store listed a Qrevo Curv 2 Flow robot vacuum at $999.99 against a crossed-out $1,599.99, an advertised saving of $600, about 37.5% off the crossed-out price. The arithmetic is correct. Whether it is a strong deal depends on evidence the tag does not show, namely what that model has been selling for over the months before the sale. That is the gap the rest of this check fills.
The two legal tests, in shopper language
Canadian rules give a regular price two ways to prove itself, and a claim is in trouble when it passes neither. The volume test: more than 50% of sales of the product were at that price or higher within a reasonable period, usually within a year, before or after the promotion. The time test: the product was offered for sale, in good faith, at that price or higher for a substantial period of time, usually within a year, immediately before or after the promotion. Good faith means the retailer honestly believed the price was fair and expected customers to actually pay it.
A shopper cannot see a store's sales volume, so the volume test is out of reach from outside. The time test is not. Price history shows how long a price was on offer, which makes history the shopper's working version of the test. A regular price that appears for a few days in October and vanishes is weak evidence. A regular price that held for months is strong evidence. The guides that cover fake markdowns describe the same underlying tactics from the other side: an inflated list price, a price raised in the weeks before the sale, and a discontinued model cleared out under a fresh label.
The five step crossed-out price check
Run this on any Black Friday tag that leans on a crossed-out price. It takes a few minutes per item, so save it for the purchases that matter.
- Copy the exact identity of the item. Record the model number, size, specifications and included accessories before comparing anything. Prices for a near match prove nothing about this item.
- Name the reference type. Decide which of the three reference prices the tag is using, using the table above. The type decides which check comes next.
- Pull the price history. For Amazon Canada items, CamelCamelCamel and Keepa track Amazon.ca prices and show history charts, and Amazon's own price history link shows what an item cost over the last 30, 90 or 365 days. For other retailers, use the prices recorded during prep, or the same store's earlier flyers and pages.
- Redo the percentage against the typical selling price. Use the worked example in the next section. The advertised percentage is the seller's version; the recalculated one is yours.
- Check the final total and the exit. Compare the final checkout total in Canadian dollars, including shipping, handling and taxes, and confirm the return policy before paying. A price in US dollars can change character once exchange, shipping, duties and brokerage land on it.
Worked example: redo the percentage against the typical price
This is a worked example with labelled assumptions, not a live listing. Assume a tag reads: regular price $200, sale price $120, advertised as 40% off, a saving of $80. The arithmetic on the tag is correct. Now assume the price history from step 3 shows the item actually sold at $150 for most of the preceding months, with the $200 price appearing only briefly. Measured against the price shoppers have really been paying, the saving is $150 minus $120, which is $30, and the real discount is $30 divided by $150, which is 20%. Half the advertised saving was the reference price, not the deal.
The decision framework that follows from the math:
- Buy on the recalculated number. If the sale price beats the typical selling price by a margin worth acting on, and it is at or near the lowest price in the history window, the deal is real regardless of what the tag claimed.
- Wait when the two numbers converge. If the sale price is close to the typical price, the sale is a label, not an event. The item will very likely be available at that price again without the crowd.
- Pass when only the tag looks good. If the advertised percentage is large and the recalculated saving is small, the crossed-out price was doing the selling. That is the pattern the Bureau's fake ordinary selling price warning describes.
When a real discount is still not worth buying
A discount can pass both tests and still be the wrong purchase. Three common cases. First, the final total: a genuinely reduced item shipped from outside Canada can end up costing more than a higher priced domestic listing once the full checkout total in Canadian dollars is compared. Second, the terms: a real sale price attached to final sale terms converts a pricing question into a risk question, and the return policy check in step 5 is what catches it. Third, the item itself: a discontinued model cleared at a genuine discount is still a discontinued model, with the parts and support horizon that status implies. The discount being honest does not make the purchase wise.
The record to keep, and where a complaint goes
The Bureau recommends that businesses keep detailed records of promotions, including dates, discounts and regular prices. A shopper who suspects a fake regular price can keep the mirror image of that record, and it takes under a minute: a dated screenshot or saved page showing the sale price, the crossed-out price and the model number, plus the price history chart or the earlier prices recorded during prep. The Bureau investigates deceptive marketing practices, and it states that a business or consumer who comes across fake sales should report them to the Competition Bureau. A complaint with dates and prices attached is evidence. A complaint that says the discount felt fake is an opinion, and only one of those can be investigated.
What success looks like: every major Black Friday purchase has a typical selling price recorded beside the sale price, the percentage in the decision was the recalculated one, and any crossed-out price that could not support itself was treated as advertising, not information.