The Same Price Does Not Cost the Same Depending on How It Is Displayed

In August 2015, StubHub, the American ticket resale platform, split its visitors into two randomly assigned groups. The first saw the total price, fees included, from the very first page. The second discovered the fees, around 15% of the ticket price, only at checkout. The tickets, the available seats and the final amount paid were identical. According to the analysis of the experiment (opens in a new tab) published by four economists using the company's data, the second group spent almost 21% more. It bought more often, with 14.1% more transactions over ten days. It also chose more expensive seats.

The price paid was the same. Only the moment it appeared in full had changed. Price psychology, a branch of consumer behavior research, has accumulated dozens of effects of this kind, which pricing guides turn into psychological pricing recipes. Their robustness varies widely. Some rest on millions of real transactions. Others, among the most cited, have not survived replication, meaning experiments rerun by other teams with the same protocol.

Economist Avner Strulov-Shlain analyzed checkout data on 3,500 products sold by 25 American retail chains. The paper, an article in the Review of Economic Studies (opens in a new tab) published in 2023, concludes that consumers respond to a one-cent increase from a 99-ending price "as if it were more than a twenty-cent increase". Moving from $4.99 to $5.00 costs as many sales as an increase twenty times larger. Retailers know about this bias without exploiting it fully. The author estimates that they forgo 1 to 4% of their potential gross profits as a result.

The phenomenon is called left-digit bias. It reaches well beyond supermarket shelves. Across more than 22 million wholesale used-car transactions, Nicola Lacetera, Devin Pope and Justin Sydnor observed sharp drops in price (opens in a new tab) at each 10,000-mile threshold on the odometer. Two cars a few hundred miles apart sell at clearly different prices as soon as one crosses a new 10,000-mile threshold, although the buyers are automotive professionals.

Manoj Thomas and Vicki Morwitz specified the condition for the effect in 2005. It appears only when the leftmost digit changes (opens in a new tab), between $2.99 and $3.00 for example. Between two prices that start with the same digit, a one-cent gap is perceived at its actual size. The effect also grows when the compared price is close, which is the case for a grid of two or three plans displayed side by side.

The 9 ending also works as a signal. Eric Anderson and Duncan Simester tested several prices (opens in a new tab) for the same items in the catalog of a women's clothing retailer. They report in the Harvard Business Review (opens in a new tab) that demand for a dress rose by a third when its price went from $34 to $39. Moving from $34 to $44 changed nothing. The increase sold more because the 9 reads as the mark of a bargain.

That signal does not suit every offer. Robert Stiving showed in 2000 that firms use round prices more often for higher-quality products (opens in a new tab). The relationship is even stronger in categories where customers cannot judge quality before buying. Consulting, coaching or a custom-built website belong to that category. A round price carries the quality signal there, whereas a 9 ending carries the discount signal.

The StubHub experiment measures what transparency costs. The group that discovered fees at checkout spent more for two reasons. It bought more often. Above all, it chose better seats, an effect to which the authors attribute at least 28% of the revenue difference (opens in a new tab). Visitors compare seats on a partial price, which makes every upgrade look cheaper than it really is. The authors observe that the effect persists beyond the first purchase and also affects experienced users of the platform.

Vicki Morwitz, Eric Greenleaf and Eric Johnson had described the mechanism as early as 1998. A price split into several mandatory parts, a base price followed by shipping charges for example, decreases the total cost customers recall and increases demand (opens in a new tab).

The practice is now regulated. In the United States, the Federal Trade Commission's rule on junk fees, in force since May 2025, applies to live-event tickets and short-term lodging. It requires businesses to display the total price more prominently than any other pricing information (opens in a new tab). StubHub's 21% therefore measures the revenue from a display that is now banned in those two sectors.

For a service business website, the lesson concerns the order in which amounts appear. A package shown without its options makes each option feel heavier when it is added, since the customer has already retained the first figure. A total shown from the start looks higher on first reading and does not change afterward. According to the StubHub experiment, it sells slightly less in the moment. In return, the customer discovers no additional amount at checkout.

Anchoring is the tendency to estimate an amount from the first number encountered, even when that number has nothing to do with the question. It is one of the classic effects that best withstood Many Labs (opens in a new tab), a project that reran thirteen famous experiments in 2014 across 36 independent samples totaling 6,344 participants. Ten effects replicated, anchoring among the clearest. A 2021 meta-analysis (opens in a new tab) devoted to willingness to pay confirms anchoring on prices. It measures a more modest effect than the early studies did.

The precision of the number changes the strength of the anchor. Chris Janiszewski and Dan Uy showed in 2008, across five studies, that people adjust less away from a precise starting point (opens in a new tab) than from a round one. Faced with a round amount, the other party reasons in broad steps of hundreds or thousands. Faced with a precise amount, they adjust in small units and stay closer to the stated figure. Malia Mason and coauthors found the same phenomenon in negotiation in 2013. A precise first offer (opens in a new tab) draws more conciliatory counteroffers because the recipient credits its author with better knowledge of the subject.

David Loschelder, Malte Friese, Michael Schaerer and Adam Galinsky qualified this principle in 2016. They ran five experiments with 1,320 experts and amateurs in real estate, jewelry, cars and human resources. Among amateurs, the more precise the offer, the stronger the anchor (opens in a new tab). Among experts, the effect rises and then reverses, because excessive precision strikes them as a sign of incompetence. The negative effect disappears when the offer comes with a rationale that demonstrates the competence of the person making it. A quote itemized line by line fulfills exactly that function. Its precise total is explained by the lines that make it up.

Another rule about precision circulates widely in pricing guides. Monica Wadhwa and Kuangjie Zhang published in 2015 that round prices suit purchases driven by feelings and precise prices suit reasoned purchases, because each form "feels right" in its context. In 2018, a German team reran their main experiment with 588 people in a preregistered replication (opens in a new tab), whose protocol is filed before data collection. It found neither the effect nor its mechanism. Its authors conclude that the effect is "either smaller than originally reported" or contingent on factors not yet investigated.

In 2008, Hilke Plassmann and colleagues from the California Institute of Technology and Stanford had 20 people taste wines inside a functional magnetic resonance imaging scanner, which measures activity in different brain regions. The same $5 wine was presented to them once at that price and once at $45. A $90 wine was presented at $90 and at $10. Participants found each wine more pleasant when told it cost more (opens in a new tab). Activity in the medial orbitofrontal cortex, a region associated with experienced pleasure, rose with the stated price. The price therefore changed the experience itself, not only the judgment made about it.

The study covers 20 people and a single product. Its result also depends on what people know about the price. Robin Goldstein and coauthors gathered 6,175 blind ratings (opens in a new tab) from 506 participants across 17 tastings held in the United States. Without knowing the price, non-experts on average enjoy expensive wines slightly less. Only participants with wine training show a positive relationship between price and enjoyment. For most people, the extra pleasure of an expensive wine comes from its displayed price.

This role of price as a quality cue explains another effect, the order of display. Kwanho Suk, Jiheon Lee and Donald Lichtenstein showed in 2012 that options presented from highest to lowest price (opens in a new tab) lead people to choose more expensive options. Presented in the reverse order, the same options lead to cheaper choices. The effect appears only among people who associate price with quality. A pricing grid that starts with the most complete plan therefore sets a high reference before the others. The study has not been independently replicated. This link between price and perceived quality matters all the more when the purchase requires deliberation, as the article on what converts depending on what you sell (opens in a new tab) shows.

The decoy effect is one of the most cited pricing recipes. It consists of adding a third option designed to be less attractive than another, in order to steer customers toward that other option. The canonical example comes from Dan Ariely's book Predictably Irrational, published in 2008. In it, an advertisement for The Economist offers a web subscription at $59, a print subscription at $125 and print plus web for the same $125. According to the book, the print-only option, which nobody chooses, tips most students toward the combined offer. This example is a classroom demonstration given to MIT students. It was never published in a scientific journal.

Shane Frederick, Leonard Lee and Ernest Baskin looked for the effect with realistic choices in 2014, for instance by having people taste drinks or by showing photographs of the view from apartments. Across 19 studies with 68 to 681 participants each, they write that they found no evidence for the effect in any of them (opens in a new tab). The decoy reduced the appeal of the targeted option as often as it increased it. In 2021, an experiment built to meet the criteria set by the effect's own authors measured a "precisely zero" effect (opens in a new tab) on choices between ordinary products.

The effect has not disappeared entirely. A registered report published in 2025 (opens in a new tab) finds it robustly when options are described by numerical attributes. Its strength varies with the display order of the options and with whether attributes are shown as numbers or as a graphic. The decoy therefore works on numerical comparison tables, which is exactly the form of a pricing grid. It weakens sharply once the choice concerns a product people taste, look at or try.

Removing the currency symbol is another common recipe. It relies on a study from Cornell's hotel school published in 2009. At St. Andrew's, a restaurant of the Culinary Institute of America, 201 tables received menus with prices written in three ways, with a dollar sign, in words or as bare numbers. The study report (opens in a new tab) states that the overall effect of the format was not statistically significant. The often-quoted $5.55 difference only appears after controlling for party size, dining duration and table location. The authors write themselves that these factors have "a much larger impact" on spending than price typography does. Money priming, the theory meant to explain the effect, is one of the two experiments Many Labs failed to reproduce.

Price size follows the same pattern. The 2005 study by Keith and Robin Coulter, often summed up as "a price in small type looks cheaper", involved 65 students and an advertisement for fictitious in-line skates. It showed that a sale price looks like better value (opens in a new tab) when it is written smaller than the regular price it replaces. It did not measure the effect of a small price in absolute terms. No replication has been published. The role of typography in conversion is covered more broadly in the article on typography and conversion rate (opens in a new tab).

Card payments followed a similar path. Drazen Prelec and Duncan Simester measured in 2001 a markedly higher willingness to pay (opens in a new tab) by card than in cash. A 2024 meta-analysis (opens in a new tab) of 392 effect sizes from 71 papers finds a small but significant effect that has weakened over time as cards became the ordinary means of payment.

Several famous pricing recipes rest on the work of Dan Ariely, two of whose papers have been retracted by the journals that published them. The first, retracted in 2021, contained an insurance experiment whose data turned out to be fabricated (opens in a new tab). The second, a 2002 study on procrastination, was retracted in September 2026 (opens in a new tab) after researchers found signs of data tampering. Neither concerns pricing. These retractions explain why isolated results from this body of work, such as The Economist example, call for independent confirmation.

The most cited pricing effects, ranked by strength of evidence

Since 28 May 2022, any announcement of a price reduction on a product in the European Union must use as its reference the lowest price applied during the previous 30 days. The rule comes from the EU Omnibus Directive (opens in a new tab), transposed in France as Article L112-1-1 of the Consumer Code. In September 2024, the Court of Justice of the European Union specified, in a case involving ALDI Süd (opens in a new tab), that the displayed percentage reduction must also be calculated from that reference price. A reference price inflated a few days before a promotion now counts as a misleading commercial practice.

The rule covers products rather than services. It does not apply to a quote for consulting or design work. It does apply to any online store that shows a former price next to the new one.

TrustRadius, a review platform for business software, surveyed 1,862 buyers and 444 vendors for its 2026 report. Transparent pricing has been buyers' top wish for vendors (opens in a new tab) every year since the question was first asked in 2023. The survey is self-reported and covers standardized offers, sold by subscription on a grid known in advance.

A custom service works differently. Its price does not exist before the need has been studied. It appears for the first time in a proposal or a quote, which then becomes the business's pricing page. The solid effects in the table above all apply there. The first amount read anchors the discussion that follows. A precise total convinces an experienced buyer when every line justifies it. An option added after the total weighs more heavily than an option included from the start. The order in which plans are presented steers the choice when the buyer associates price with quality.

The price psychology effects that withstand scrutiny share a common trait. They depend on the order and timing of reading, on the leftmost digit, on the first amount seen and on fees revealed after the fact. They were measured on millions of transactions or found again by independent teams. The effects that collapse depend instead on details of form, such as the currency symbol, the size of the type or a round number meant to appeal to emotion. They usually rest on a single study, sometimes on a classroom demonstration.

Price also acts as a quality cue, to the point of changing the pleasure people feel when they know it. That signal matters most for offers whose quality can only be verified after purchase, which is the case for most services. How a price appears, on a pricing page as in a quote, is therefore part of what the customer buys. Replicated studies provide a more reliable basis for it than the recipes that circulate most.