Is this a good deal?

A $79 jacket marked down from $160 is either a 51% discount or a $79 jacket. Which one it is depends on a question the tag doesn't answer: did anyone ever pay $160? In 2025, Consumers' Checkbook tracked sale prices at 25 national chains for 24 weeks and found that 21 of them advertised sale prices more than half the time. Across most of those stores, the tracked items were "on sale" about 76% of the time. So the markdown on the tag tells you where the seller started counting, not whether the price is good. Here's how to find out.

How do you tell if something is a good deal?

Compare the price in front of you with what the same item sells for right now from other sellers, not with the "was" price printed next to it. That's the whole method. Everything below is a way of doing that comparison faster or more carefully.

A good deal has three parts. The price is at or below what you'd pay elsewhere for the same thing in the same condition. You'd buy the item at that price even if it weren't marked down. And you can see the comparison, not just feel it. A tag that says "Save 40%" gives you none of the three. A search that turns up the same item for $70 at two other stores gives you all of them.

What if you can't find the same item? Then you're pricing against comparable items: same category, same brand tier, same condition. That's a looser comparison, so leave yourself more room. A used dresser priced 10% under the nearest comparable listing isn't a clear win. One priced at half is.

Is a sale price really a discount?

Usually not, at least at national chains. The Checkbook study above found that only Apple, Costco and Dell ran discounts that were consistently real, meaning the items sat at their regular price most of the time. At 12 of the 25 companies, more than half the tracked items sat at a supposed discount every week or almost every week.

The rules on this aren't new. The FTC's Guides Against Deceptive Pricing say a former price can support a bargain claim only if it was "the actual, bona fide price at which the article was offered to the public on a regular basis," and that a price set up just to allow a big markdown is fictitious. The guides aren't enforced much, as the Checkbook authors point out, which is why the tags look the way they do.

Does the timing of a sale help? Less than you'd think. Which?, the UK consumer group, tracked 175 Black Friday 2024 deals for six months on either side of the event and found 83% were the same price or cheaper at some point outside the sale window. None were at their cheapest only on Black Friday. That's UK data, but the pricing habits are the same ones Checkbook found in the US.

Is 40% off a good deal?

Only if 40% off lands below what the item sells for elsewhere. The percentage measures distance from a number the seller chose, and you've just seen how those numbers get chosen. A $200 "compare at" price cut by 40% is $120. If the same item is $95 at the store across the road, the 40% is a decoration.

Why does the percentage feel so persuasive? Partly because it's doing arithmetic for you, and partly because of how it's framed. Jonah Berger describes a rule marketers use: under $100, percent off reads as bigger than the same discount in dollars, and over $100 the reverse holds. A 25% markdown on a $20 shirt sounds better than $5 off, and $500 off a $2,000 laptop sounds better than 25% off. Sellers pick whichever version looks larger. You can undo it by ignoring both and looking only at the final price.

One more thing to know: a 2023 study in Marketing Science found that some online sellers raise an item's price and add a list price at the same moment, so a price increase shows up on the page as a discount. If a "deal" appeared the same week the item got more expensive, that's what you're looking at.

How do you check a price on the spot?

Search the exact item and read what other sellers are asking for it today. For a new product, the model number or the full product name on the tag is enough. Three or four current prices from different sellers tell you more than any sale sticker.

For anything sold online, look at the price history, not just today's price. Price-tracking sites for Amazon and other large retailers chart what an item has cost over the past year, so you can see whether today's "deal" is the usual price with a new label. If the line is flat, the discount isn't one.

For used items, asking prices aren't enough. What you want is what comparable items actually sold for, because sellers can ask whatever they like. eBay's search can be filtered to sold listings, which is the closest thing to a public record of what people paid. A gap between the asking prices and the sold prices is normal. Price against the sold ones.

For consumables, compare the unit price. The shelf label's price per ounce or per count is the only number that lets you compare a 12-pack with an 18-pack, or a store brand with a national one.

What about used and secondhand items?

There's no list price to argue with, so the comparison is to other used examples in the same condition. Start with sold prices, as above. Then adjust for what's in front of you: condition, completeness, whether it's the common version or the one people hunt for.

Thrift stores, estate sales and marketplace listings all price by hand, which cuts both ways. Some tags are set from a quick look at online asking prices, which runs high. Some are set from nothing at all. Either way, the tag tells you what the seller hopes for. A minute spent looking at what the item has sold for tells you what it's worth to the next buyer, which is the number that matters if you're buying.

How Zozy checks a price

Zozy is an iPhone app that answers this question from a photo. You photograph the item in the store, Zozy reads the price on the tag (you can correct it), and within seconds it gives you an estimated fair price and live comparable listings, plus one of four verdicts: Great Price, Good Price, Fair Price or High Price. When there isn't enough data to judge, it says so rather than guessing.

A few things you should know before you trust it. Results are automated estimates meant to help you decide whether to buy. They can be wrong, and they are not a professional appraisal. The listings it shows are asking prices, not sale prices, so treat them as context for your decision, not a guaranteed number. Zozy doesn't promise an exact valuation or that you'll find the item cheaper somewhere else.

What it does promise is independence. Zozy makes money from memberships and nothing else: no advertising, no affiliate commissions, no payments from brands or sellers, and it doesn't sell your data. It earns the same whether you buy or walk away. It's free to start, with 10 price checks a month. Membership is pay what you can, at $1.99, $3.99 or $7.99 a month, and every option gets the same access, including unlimited price checks. The details are on the FAQ.

Before you pay: a checklist

• Ignore the "was" price and the percent off. Write down the final price.

• Search the exact item and find two or three current prices from other sellers.

• Buying online? Check the price history. A flat line means no discount.

• Buying used? Compare with sold prices, not asking prices, in the same condition.

• Buying consumables? Compare the unit price, not the package price.

• Ask whether you'd buy it at this price with no sale sign on it.

• If the answer is yes and the price is at or below what you found, it's a good deal.