011 min
Where price anchoring shows up
An illustrative case. A PM is redesigning the pricing page of a subscription app. It has three plans: Basic at $9, Pro at $19 and Team at $49 per month. In the first version the page lists the plans from cheapest to most expensive, and most visitors pick Basic. In the second version the page lists Team first, and Pro is marked as the recommended plan. The prices are the same, but visitors now read $19 as the moderate option, and more of them pick Pro.
The PM added no information. Only the order changed, so the first number a visitor sees became the number every other price is compared with. That first number is the anchor.
The same structure appears in many places:
- A crossed-out "was" price next to the current price.
- A premium plan or a bundle shown beside the plan the seller wants to sell.
- A menu or price table that starts with its most expensive item.
- An opening quote that starts high in a sales negotiation.
- A price for something the buyer did not come for, shown before the price of what they did come for.
The tactic changes how the buyer compares prices. It does not change the product.
021 min
Where price anchoring comes from
Price anchoring applies a finding from psychology, not a pricing theory of its own. In 1974, Amos Tversky and Daniel Kahneman described anchoring as a heuristic used to make estimates under uncertainty. A heuristic is a quick rule of thumb for making a judgment. Their experiments showed that people who start from one number and adjust from it end up closer to that number than people who start from another.
Marketing took the idea over because a price is a number that buyers must judge, and most buyers have no firm sense of what a product is worth. No single person introduced "price anchoring" as a named technique. Sellers have long listed a high price first, and behavioral economists later measured why it works.
032 min
How price anchoring works
Price anchoring is an application of anchoring: a number a person has just seen moves their estimate toward it. The effect is largest when the buyer is unsure what the product is worth. With no internal price to compare against, the buyer uses the nearest number available, and that number is the anchor.
The best-known test of this comes from Dan Ariely, George Loewenstein and Drazen Prelec, who published it in the Quarterly Journal of Economics in 2003. In one experiment, fifty-five MBA students were shown six ordinary products, such as a cordless computer keyboard and bottles of wine. The students first wrote down the last two digits of their social security number. A social security number is a government identity number that has nothing to do with the value of a product. The students then said whether they would buy each product for that many dollars, and finally stated the most they would pay.
The valuations of the top quintile subjects were typically greater by a factor of three.
A quintile is one fifth of a group. Students in the top fifth of the number range were willing to pay $56 on average for the keyboard, compared with $16 for students in the bottom fifth. Students with above-median numbers stated values from 57 percent to 107 percent greater than students with below-median numbers.
Two details matter for pricing work. First, the students did not know what the products were worth in dollars, but they agreed about which was worth more. The authors call this combination coherent arbitrariness: the absolute price a person will pay is arbitrary, but the order in which they rank products is stable. Second, the effect did not go away when real money was at stake. The students' answers decided whether they bought the product, and the number still moved them.
The practical conclusion is that a buyer's willingness to pay is partly formed at the moment they are asked. If you ask right after showing a price, part of what you measure is the price you showed.
041 min
Does competition remove price anchoring?
A common assumption is that a real market corrects the effect, because buyers who overpay lose money and learn. Ariely, Loewenstein and Prelec tested this. They gave students an arbitrary anchor as before, and then had them trade in a multiperson auction instead of stating a private price.
The authors report that the market did not remove the bias. They found that markets can even strengthen it: when everyone in the auction has seen the same anchor, their bids agree with each other and the shared number looks like a fair price.
For a seller this means a buyer's reaction to an anchor is not weakened just because buyers can compare alternatives. For a buyer, such as a founder negotiating a vendor contract, it means the first quote they hear is the number to question.
051 min
Do experts resist price anchoring?
Knowledge reduces the effect less than most people expect. Gregory Northcraft and Margaret Neale studied how people price a house in 1987. Students and professional real-estate agents were each shown the same house and were given different listing prices. A listing price is the price the seller asks.
In the follow-up interviews, the real-estate agents denied being influenced by the initial price, but the results showed that both groups were equally influenced by that anchor.
The agents had professional experience and said they had ignored the listing price, yet their estimates followed it as closely as the students' did. A buyer who says the list price did not affect them is weak evidence that it did not.
There are two consequences for people who set prices. An informed customer is not immune to an anchor. And a customer who says price was not a factor may not be describing their own decision accurately.
061 min
Common ways to set a price anchor
Wikipedia's summary of the pricing literature groups the common tactics. The table below adds what each one does and what can go wrong.
| Tactic | What the buyer sees first | Why it helps the target price | What can go wrong |
|---|---|---|---|
| Sort from high to low | The most expensive item on a menu or price table | Items lower in the list look cheaper than expected | A buyer who is very sensitive to price may leave before scrolling |
| Decoy | An inferior option priced close to the target | The target looks like a better deal than the decoy | A buyer who notices the decoy may distrust the whole page |
| Reference price | A higher list price or "was" price | The current price looks like a saving | If the reference was never charged, the claim is false |
Wikipedia notes that the high prices at the top of a menu act as anchor values. In its words, buyers "will be pleased to see the cheaper products at the middle and bottom of the list."
The decoy row is a separate effect with its own entry, the decoy effect. A decoy does not give a high number to compare against. It gives a worse option to compare against.
071 min
How price anchoring shows up in tech
Each role that touches a price page makes a different decision about the anchor.
- A PM chooses the plan order and which plan is marked as recommended. The order decides which price is read first.
- A designer chooses the visual weight of each price. A large, high price at the left edge of a table becomes the anchor even if the copy never mentions it.
- An engineer builds the A/B test and the event logging. If the test records sign-ups but not cancellations, it cannot show whether an anchor moved revenue or only moved cost.
- A founder quotes a price on a sales call. The first number said aloud sets the reference for the rest of the talk.
Annual billing is a common case. A page that shows the monthly price of the annual plan next to the full monthly price gives the buyer a reference that the plan alone would not.
081 min
How to use price anchoring without misleading customers
Use these steps in this order. Each depends on the one before it.
Anchor with a real price.
The reference must be a price you charge, or charged, for something comparable. A crossed-out price that was never charged is a false claim, not an anchor.
Make the anchor plausible.
A reference that the buyer cannot connect to the product may be ignored or may reduce trust.
Show the anchor before the target, in the same view.
The effect depends on order, so a reference seen after the buyer has judged the price does little. This step only works once step 1 is done, because a reference shown first but not credible harms the page.
Test it.
Run an A/B test of the order and of the presence of the reference. Measure paid conversion, refunds and cancellations, not clicks.
Check the price against value.
An anchor changes how a price looks. It does not change what the product is worth to the buyer, so it will not rescue a price that the buyer's real alternative undercuts. That limit is a matter of pricing power.
The goal is a comparison that is fair and easy to make. It is not to confuse the buyer about what the product costs.
091 min
Common mistakes with price anchoring
Teams assume a higher anchor is always better. A reference that is far above any believable price is ignored, and it can make the seller look dishonest. The anchor must be plausible for the product.
Teams assume the effect is permanent. The experiments above measured one decision at a time. A buyer who sees the page again, or compares with a competitor, forms a new reference. Effects measured in a first-visit test may not hold for returning customers.
Teams assume a fake reference price is only a marketing choice. A "was" price the seller never charged can be a legal problem in many countries and it damages trust when buyers notice. Check the rules where you sell.
Teams assume anchoring replaces research on willingness to pay. An anchor moves what buyers say they will pay. The tactic does not tell you what they will pay once they have used the product for a month.
101 min
Price anchoring versus the general anchoring effect
Anchoring is the general bias: any number seen before an estimate moves the estimate toward it. Price anchoring is one use of it, where the number is a price and the estimate is how much a product is worth.
The deciding difference is who chooses the number. A researcher uses a random number to show that the bias exists. A seller chooses a relevant number, such as a real list price, to guide the buyer's comparison.
Two other ideas sit nearby. The framing effect changes how the same facts are described, for example "90% fat free" instead of "10% fat". Perceived value is the worth the buyer assigns to the product, and an anchor is one of the things that shapes it.
?7 questions
Questions people ask
What is price anchoring?
What is an example of price anchoring?
Does price anchoring work on experienced buyers?
Does the anchor have to be a real price?
How do you test a price anchor?
What is the difference between price anchoring and the decoy effect?
When should you not use price anchoring?
§5 sources
Sources
Ariely, D., Loewenstein, G. and Prelec, D. (2003). "Coherent Arbitrariness: Stable Demand Curves Without Stable Preferences." Quarterly Journal of Economics, 118(1), 73-106.
Ariely, D., Loewenstein, G. and Prelec, D. (2003). "Coherent Arbitrariness: Stable Demand Curves Without Stable Preferences." Author-hosted copy of the same paper.
Northcraft, G. B. and Neale, M. A. (1987). "Experts, amateurs, and real estate: An anchoring-and-adjustment perspective on property pricing decisions." Organizational Behavior and Human Decision Processes, 39(1), 84-97. (87)90046-X
Tversky, A. and Kahneman, D. (1974). "Judgment under Uncertainty: Heuristics and Biases." Science, 185(4157), 1124-1131.
Show all 5 sourcesShow fewer sources
Wikipedia. "Anchoring effect."
