Short answer: sometimes, and knowing which kind of "sometimes" matters more than most pricing advice admits. The effect is real — demonstrated in a controlled study in 1982, and famous from a genuine, well-documented experiment with actual subscription prices. It is also fragile: a large attempt to reproduce it on real products, with real descriptions rather than bare numbers, found a reliable effect in a small minority of tries. Both facts are true at once, and a page that only tells you one of them is not being straight with you.
The tool Decoy Pricing Calculator Price a third option that nudges customers toward the one you actually want sold, with the reasoning shown as arithmetic.Where it comes from
The underlying idea is called asymmetric dominance, first demonstrated by Joel Huber, John Payne and Christopher Puto in a 1982 paper in the Journal of Consumer Research. Add a third option to a choice between two things — an option that is clearly worse than one of the originals but not clearly worse than the other — and people become more likely to choose the option it is worse than. The new option does not have to be attractive. It only has to make something else look better by standing next to it.
A decade later, Itamar Simonson and Amos Tversky proposed a mechanism for why, in a 1992 paper in the Journal of Marketing Research they called extremeness aversion: an option's attractiveness rises when it sits in the middle of a choice set and falls when it sits at an extreme. A decoy does not need to be liked to work — it only needs to push the option you actually want to sell into that comfortable middle.
The demonstration everyone actually remembers is Dan Ariely's, run on real Economist magazine subscription prices with real MIT students. Offered three options — web access for $59, print only for $125, and print plus web also for $125 — 16% chose the web-only option, nobody chose print-only, and 84% chose the $125 bundle. Remove the print-only option, leaving just the $59 and $125 choices, and the split flipped: 68% chose the cheap option and only 32% chose the bundle.
Nobody bought the decoy. Its entire job was to make the $125 bundle look like it was worth having print for free. Ariely put a number on what that was worth: predicted revenue across the group rose from $8,012 to $11,444 — a 43% increase, from doing nothing but adding an option nobody chose.
Why it mostly does not replicate
That result is thirty-odd years old and became one of the most quoted numbers in pricing psychology, which is exactly the kind of fact worth pressure-testing. In 2014, Sybil Yang and Michael Lynn tried to reproduce the effect systematically — 91 separate attempts, across 23 different real product categories, using realistic descriptions and images rather than bare numbers on a page. Only 11 of the 91 attempts produced a reliable effect.
The original researchers themselves responded the same year, in a paper candidly titled "Let's Be Honest About the Attraction Effect." They did not disown the phenomenon, but they conceded it is less common in real markets than a related, gentler effect — the compromise effect, where the middle of three options simply gains share for being in the middle, described by Itamar Simonson in 1989. The 1982 result stands as real. Applying it to a specific product on a specific page and expecting an 84/16 split is where the evidence runs out.
A frequently retold account illustrates the softer version. Williams-Sonoma, according to a story traced to Barbara Buell writing in Stanford's own Stanford Business magazine — which this guide has not been able to read directly, so treat the specifics as reported rather than independently confirmed — once struggled to sell a $275 bread maker. It added a second, similar model at $429. Few people bought the expensive one, and sales of the original $275 machine nearly doubled. The cheap machine had not changed. Its position in the comparison had: from "the expensive option" to "the reasonable one in the middle," which is the compromise effect at work rather than the sharper asymmetric dominance above.
Put plainly: this is a real lever, and it is not a reliable one. The gap between "documented once, dramatically" and "works whenever you use it" is most of what separates good pricing advice from bad.
Why bare numbers behave differently from real products
One detail from the 2014 replication is worth keeping specifically: the effect held up better in abstract choices — plain numbers, no pictures, no brand — and weakened or vanished once real descriptions and images were added. A subscription price is close to the abstract case; a physical product with a photo, a brand and a description is not.
Which is a reasonable account of why it is worth testing on your own audience rather than assuming: the more your product looks like "a photo and a story" and the less it looks like "a number on a comparison table," the less this specific mechanism has been shown to survive.
What still makes it worth trying
Even at eleven working attempts out of ninety-one, that is real upside for something that costs nothing to test — no development, no inventory, just a third listed price. A few things improve the odds, based on how the effect is defined rather than on a fixed number:
- Make the decoy genuinely worse. Not merely different — worse in a way a buyer registers in a second, on the specific thing your target option does better.
- Keep it close to the target option in price. The calculator here shows this as the gap up to your expensive option — the smaller it looks next to the gap already covered from the cheap one, the harder it pulls.
- Measure it. Add the option, watch what actually happens to your numbers for a real stretch of time, and be willing to find out it did nothing. That is the whole lesson of the 91 attempts.
- Put a real number on the decoy itself. If it is going on an invoice or a quote once somebody does buy it, the invoice generator handles a line item priced however oddly you decided on.
The thing this is not
A decoy is a real option somebody can genuinely buy. It is not a price with a line through it that was never a real selling price — that is a different tactic with a different name, fictitious former pricing, and it is not merely less persuasive, it is against the law in a lot of places.
In the United States, the FTC's Guides Against Deceptive Pricing say a former price is a legitimate basis for comparison only if it was "the actual, bona fide price at which the article was offered to the public on a regular basis for a reasonably substantial period of time." An inflated price invented specifically to make a discount look bigger is, in their words, a false bargain. Advertisers are told to avoid language like "formerly sold at" unless real sales were actually made at that price.
The European Union went further in 2022: under the Omnibus Directive's Article 6a, any advertised price reduction has to show the lowest price actually charged in the previous thirty days. A shop cannot raise a price for a week and then announce fifty per cent off it.
Neither rule touches decoy pricing as such, because a decoy is not a claim about the past — it is a real third product sitting on the page today, at a real price, that a customer could actually choose. That is precisely the distinction worth keeping straight: one is a pricing strategy with mixed evidence behind it, the other is a specific, well-defined kind of false advertising. If what you actually want is to show a genuine markdown rather than add a third option, that is a plain percentage calculation instead — our discount calculator handles it, including why two discounts in a row do not simply add up.