Anchoring Bias in Pricing: The Number Your Customer Sees First Changes Everything

In 1987, two researchers at the University of Arizona brought professional real estate agents to a house in Tucson. Not a hypothetical house. A real one, on a real street, with real comparable sales data. Each agent walked through the property, inspected the rooms, reviewed a 10-page information packet with neighborhood details and recent comparable listings. Then they were asked to estimate four things: the appraised value, the appropriate listing price, a reasonable purchase price, and the lowest acceptable offer.

The agents didn't know they were all looking at the same house with one difference. The listing price on the information packet had been changed. Some agents saw a high number. Others saw a low one. Every other detail was identical.

The listing price moved their professional judgment on all four measures. Agents who saw the higher number appraised the house higher, recommended a higher listing price, and set a higher floor for the lowest acceptable offer. These weren't students in a lab. They were professionals whose entire career depended on pricing homes accurately. Anchoring bias is the cognitive phenomenon where an initial number, even an arbitrary or irrelevant one, shifts subsequent judgments toward that number. It operates below conscious awareness, affects experts and novices alike, and shapes every pricing decision your customers make, whether you design for it or not.

When Gregory Northcraft and Margaret Neale asked the agents what factors had influenced their appraisal, only 19 percent mentioned the listing price. The amateurs in the study were more honest about it: 37 percent admitted the listing price mattered. The agents were just as anchored. They just couldn't see it.

What Is the Anchoring Effect?

The founding experiment is almost comically simple. In 1974, Amos Tversky and Daniel Kahneman spun a wheel of fortune in front of research participants. The wheel was rigged to land on either 10 or 65. After it stopped, participants were asked: "Is the percentage of African countries in the United Nations higher or lower than this number?" Then they were asked to give their actual estimate.

The wheel was random. Everyone in the room knew it was random. It had zero informational content about African UN membership. Participants who saw 10 estimated a median of 25 percent. Participants who saw 65 estimated a median of 45 percent. A transparently irrelevant number moved their estimates by twenty percentage points.

In the same paper, published in Science, Tversky and Kahneman ran a second experiment. They asked one group to estimate the product of 8 x 7 x 6 x 5 x 4 x 3 x 2 x 1 within five seconds. They asked another group to estimate 1 x 2 x 3 x 4 x 5 x 6 x 7 x 8. Same numbers. Same product. The answer is 40,320. The descending group estimated 2,250. The ascending group estimated 512. Both were wildly wrong, but the group that started with 8 estimated four times higher than the group that started with 1, because each group anchored to the first few numbers in the sequence and adjusted upward insufficiently.

The paper launched an entire subfield. In the five decades since, anchoring has been replicated across legal sentencing, salary negotiations, consumer purchasing, medical diagnoses, and financial forecasting. A 2011 literature review by Adrian Furnham and Hua Chung Boo called it "one of the most robust cognitive heuristics" in the psychological canon. It persists even when the anchor is implausible, operates automatically, and resists correction even when people are warned about it.

How Does an Arbitrary Number Set What You'll Pay?

In 2003, Dan Ariely and his colleagues at MIT ran an experiment that should make anyone who sets prices uncomfortable. They showed MBA students six products: a cordless trackball, a cordless keyboard, two bottles of wine (one average, one rare), a design book, and Belgian chocolates. Each product was described enthusiastically.

Then Ariely asked each student to write down the last two digits of their Social Security number.

After writing the number, students were asked: would you pay this amount (your Social Security digits, treated as a dollar figure) for the product? Then they submitted their maximum willingness to pay in a real auction with real money.

Students whose Social Security numbers ended in the top quintile (80-99) bid between 216 and 346 percent more than students in the bottom quintile (00-19). For a bottle of 1998 Cotes du Rhone, high-number students bid an average of $27.90. Low-number students bid $8.64. The anchor was a government-issued identification number. It contains exactly zero information about the price of wine.

Ariely called the finding "coherent arbitrariness." Once an arbitrary anchor sets an initial price level, subsequent valuations become internally consistent relative to that anchor. Students bid more for the rare wine than the average wine, more for the keyboard than the chocolates. The relative ordering made sense. But the absolute level, the baseline from which all those reasonable-sounding judgments were made, was arbitrary. Determined by the last two digits of a number the Social Security Administration assigned before the student was old enough to buy wine.

"We could have just as well asked for the current temperature or the manufacturer's suggested retail price," Ariely wrote in Predictably Irrational. "Any question, in fact, would have created the anchor."

This is the part that matters for pricing. Your customers don't evaluate your price in a vacuum. They evaluate it against whatever number they encountered first, which is why the science of pricing starts with understanding reference points, not cost structures. If that number is your competitor's price, your price is judged against theirs. If it's your own premium tier, your mid-tier looks reasonable by comparison. If it's nothing at all, the customer's brain generates its own anchor from whatever is available: a similar product they bought last year, a number mentioned in a review, the first price they see when they land on your page.

You're not choosing whether to anchor. You're choosing whether to control the anchor or leave it to chance.

Does the Anchoring Effect Hold Up to Scrutiny?

Anchoring is one of the better survivors of psychology's replication crisis. But "it replicates" and "the numbers in the papers are accurate" are two different claims.

In 2025, Ye Li and Carly Weigel published a study in Economic Inquiry that did something most anchoring researchers hadn't bothered to do: they calculated the statistical power of existing anchoring studies. The results were grim. Study designs in the anchoring literature "routinely fail to achieve statistical power of more than 30%." Li and Weigel then replicated an anchoring study that had originally reported a 31 percent increase in participants' bids. With power increased from 46 to 96 percent, they found an effect of 3.4 percent, not statistically significant. The original effect had been exaggerated by roughly a factor of seven.

This doesn't mean anchoring doesn't work. It means the dramatic numbers in many published studies are probably inflated by publication bias and underpowered designs, the same pattern that has plagued psychology since the replication crisis began. The directional finding, that an initial number shifts subsequent judgments toward itself, replicates robustly. The magnitude, how much it shifts, is almost certainly smaller than the literature's headline figures.

The honest position: anchoring is real, it's directional, and it operates in real-world pricing contexts. JCPenney lost $4.3 billion learning that the hard way. But claiming that "a random number changes willingness-to-pay by 300%" is overselling the lab data.

Why Did Removing Anchors Destroy JCPenney?

In February 2012, Ron Johnson, the executive who'd built Apple's retail stores, launched "Fair and Square" pricing at JCPenney. The idea was elegant. Customers hate fake markups. They hate coupons. They hate the game of pretending a $25 shirt is "originally" $50. So Johnson eliminated all of it. No more coupons. No more clearance events. No more marking items up to mark them down. Honest prices, all the time.

Annual sales fell from $17.26 billion to $12.98 billion. A $4.3 billion drop in one year. Net loss: $985 million. Stock price fell more than 50 percent. Tens of thousands of jobs eventually disappeared. Johnson was fired after seventeen months.

What went wrong was anchoring. Under the old model, a shirt "originally" $50, marked down to $25, gave the customer two things: a shirt and a victory. The $50 was the anchor. It made $25 feel like a win. The customer's brain registered a positive prediction error: I expected to pay $50, I'm paying $25, this is a deal. Remove the anchor and the same shirt at $25 is just... $25. No reference point. No win. No prediction error.

"Coupons were a drug," Johnson said afterward. He was closer to the truth than he realized. The prediction error generated by a perceived deal activates the same dopamine pathways the brain uses to process actual rewards. The anchor doesn't just frame the price. It generates the neurochemistry that makes buying feel good. It's the same principle behind the elevator mirror effect: what you change about the context changes everything about the perceived value.

What Happens in Your Brain When You See a Price?

The two dominant models of anchoring both point to something more interesting than "you start high and adjust down."

The original model, Tversky and Kahneman's anchoring-and-adjustment, works like this: your brain uses the anchor as a starting point and adjusts toward the correct answer. But the adjustment is effortful, requires cognitive resources, and typically stops too short. Under time pressure or cognitive load, anchoring effects get stronger because the adjustment gets weaker.

The more recent model, selective accessibility, proposed by Fritz Strack and Thomas Mussweiler in 1997, goes deeper. When your brain encounters an anchor, it doesn't just "start there." The anchor changes which information your brain retrieves from memory. A high anchor activates reasons why the true value might be high. A low anchor activates reasons why it might be low. Anchoring isn't a failure of arithmetic. It's a form of confirmation bias triggered by a number. And because customers already overvalue what they have by roughly nine times what they value what you're offering, the anchor you set is the only lever that closes that gap.

A 2017 study by Li and colleagues used transcranial direct current stimulation to test this. They boosted activity in the right dorsolateral prefrontal cortex (a region involved in filtering irrelevant information from memory) and found that it reduced anchoring effects. When they suppressed that same region, anchoring increased. The prefrontal cortex helps gate which information the anchor pulls from memory. Weaken the gate and the anchor's influence goes unchecked.

Through the lens of predictive processing, the neuroscience framework that runs through Wired, an anchor functions as a prior expectation. When you see $499 first, your brain generates a price prediction centered around that number. Every subsequent price is evaluated as a prediction error against that anchor. A price lower than the anchor generates a positive surprise, the neural signature of getting a deal. A price higher generates negative surprise. You're not comparing the price to some objective standard of value. You're comparing it to whatever prediction your brain generated first.

Steve Jobs understood this before the neuroscience existed to explain it. At the iPhone launch in January 2007, Jobs spent the first part of his keynote establishing that people were already spending around $499 when they bought a smartphone and an iPod separately. Then he asked the audience: "So how much more than $499 should we price iPhone?" The audience braced for a higher number. The reveal was $499 for all three in one device. Standing ovation. The phone was expensive by any historical standard. But Jobs had set the anchor at "the cost of what you already own," and $499 landed as a deal.