Chapter 4: The First Number You Hear

A Vase at the Weekend Market

Beatriz is fifty and loves wandering through the second-hand market that fills a car park every Sunday morning. Today a blue glass vase catches her eye. She picks it up and turns it in the light. The seller, an older man with a friendly face, watches her for a moment and then says, "That one is a beauty. I was asking a hundred for it, but for you, seventy."

Beatriz hesitates. Seventy seems like a lot for a vase. But a hundred seemed like even more, and now seventy feels like a reduction, almost a gift. She offers fifty. The seller sighs, shakes his head and says sixty. They agree on fifty-five. Beatriz walks away pleased with her bargain, the vase wrapped in newspaper under her arm.

At home she places the vase on a shelf and admires it. Later that week, a friend who knows about glass visits and picks it up. "Nice," she says. "These are quite common. You see them for about twenty." Beatriz is surprised. She had felt clever about getting the price down from a hundred to fifty-five. She had never asked herself whether a hundred had any connection to the vase at all.

Beatriz did not lose very much. The vase still pleases her, and an afternoon at the market is worth something too. But the same pattern that cost her a little on a Sunday morning can cost people a great deal in larger decisions: the price of a home, the pay for a new job, the cost of a repair or the time they think a project will take. In each case, a number arrives early, and everything afterwards is measured against it.

The first number in a conversation has a strange power. Once it has been spoken, it becomes a kind of starting point for everything that follows. Our later judgments move away from it, but usually not far enough. This pattern is called anchoring, and it is one of the most reliable findings in the study of decisions.

What an Anchor Does

An anchor holds a ship in place. The ship can drift a little with the wind and the tide, but it cannot move far. In decisions, an anchor is a number that holds our judgment in a similar way. When we have to estimate something uncertain, such as a price, a time or an amount, a number that we heard just before tends to pull our estimate toward it. We may know that the number is not exactly right. We may adjust up or down from it. But we usually do not adjust enough.

The surprising part is that the anchor does not have to be a sensible number. In one classic kind of experiment, researchers asked people to spin a wheel marked with numbers from 0 to 100. Unknown to the participants, the wheel was set to stop at either a low number or a high number. Participants were then asked whether a certain quantity, a share of a group of countries in an international organization, was higher or lower than the number on the wheel, and then to give their own estimate. People who had seen the higher number on the wheel gave higher estimates, on average, than people who had seen the lower number. Everyone could see that the wheel was random. It still pulled their answers.

In other studies, people were asked to write down the last two digits of an identification number they carried, and then to say how much they would pay for ordinary items such as books or bottles. In some versions of these studies, people with higher digits offered more money than people with lower digits. The digits had nothing to do with the items. They were simply the first numbers in the person's mind.

I want to be careful about these examples. Anchoring by numbers that are relevant to the question, such as an asking price, a first offer or a previous estimate, has been found again and again in many studies, and it is one of the sturdier results in this field. Anchoring by completely random numbers, such as the digits of an identification number, has been found in some studies but has been weaker or absent in others when it was repeated. So the safest summary is this: anchors that look relevant have a strong and reliable pull, and even irrelevant numbers can sometimes have an effect, though a less dependable one.

Why the Mind Holds On

Researchers have offered two main explanations for anchoring, and both seem to play a part.

The first explanation is that we adjust from the anchor but stop too soon. Imagine you are asked how long a famous river is, and someone first asks whether it is longer or shorter than a certain distance. You start from that distance and think, "No, it must be longer than that," and you move your estimate upward. But adjusting takes effort, and you stop as soon as you reach a number that seems possible. That number is often still close to where you started. Slow thinking does the adjusting, and slow thinking, as we saw in the second chapter, likes to stop early.

The second explanation is that the anchor changes what comes to mind. When you are asked whether a vase is worth a hundred, your mind starts looking for reasons why it might be: the colour is beautiful, the glass is heavy, it looks old. These thoughts then remain in your mind when you make your own judgment. The anchor does not just give you a starting number. It quietly selects the evidence you think about. This is a form of fast thinking, and it happens without our noticing.

These two explanations also help us understand why anchors work even when we are warned about them. In some studies, people were told clearly that the starting number was random or that it might be misleading. Their estimates were still pulled toward it, though sometimes a little less. Knowing that a number is meaningless does not erase the thoughts it has already started in our minds. This is why practical protection depends on habits that we use before and after we hear a number, not just on good intentions.

Both explanations point to the same practical lesson. Once an anchor has been dropped, it is hard to escape completely. Simply telling yourself "I will ignore that number" does not work very well, because the anchor has already shaped your thoughts. A better strategy is to bring your own anchor before the other person brings theirs, or to deliberately think of reasons why the opposite might be true.

Where Anchors Hide

Once you start looking, you will find anchors almost everywhere numbers appear.

Prices are the most obvious place. A label that shows a high "original price" crossed out next to a lower "sale price" uses the first number as an anchor, so that the second one feels like a good deal. A menu that lists one very expensive dish can make the other dishes look reasonable by comparison, even if they are not cheap. A product sold in three sizes often has its middle size chosen most, partly because the largest size makes it look modest.

Negotiations are another. Whoever names the first number in a negotiation, about the price of a car, the rent of a flat or the pay for a job, often sets the range in which the rest of the discussion takes place. The other side may push back, but they usually push back from that number, not from a number of their own.

Suggested amounts are a third. When a form or a donation box suggests amounts to give, people often choose one of those amounts or something close to them. When a list of tips is suggested at the end of a meal, the suggested numbers shape what people leave. This is not always a trick. Suggested amounts can be helpful when people have no idea what is normal. But they are anchors all the same.

Anchors also appear in estimates of time and effort. If a colleague says, "This will probably take a week," that number shapes everyone's thinking about the project, even if nobody has checked it. If a doctor's waiting room shows that the average wait is twenty minutes, patients may feel more patient at fifteen and more annoyed at twenty-five. And in our own lives, the first figure we hear for a house, a wedding or a holiday often becomes the standard we measure every later figure against.

Even judgments that do not look like numbers can be anchored. A teacher who has just marked an excellent essay may mark the next ordinary essay more harshly than she would otherwise. A person who has just seen a very beautiful house may find their own home looking smaller than it did that morning. The first thing we see becomes the measuring stick for what follows.

Experts Are Not Immune

It would be comforting to think that anchoring affects only people who do not know much about a subject. If you really knew what a vase was worth, surely a seller's number would not move you. Unfortunately, the evidence suggests otherwise.

In studies with people who have professional experience, such as people who value homes for a living, researchers have found that a suggested or listed price still influenced their estimates. Interestingly, many of these experts said afterwards that the listed price had not affected them at all. They believed that they were relying only on their knowledge, but their numbers told a different story.

Studies of judgments in legal settings have raised similar concerns. In some experiments, the amount requested at the start of a case seemed to pull later judgments about appropriate amounts. These are laboratory studies, and real cases are much more complex, so we should not draw strong conclusions about any particular courtroom. But they suggest that training and experience reduce anchoring less than we might hope.

The lesson here is not that experts are foolish. It is that anchoring works below the level of awareness. Knowledge helps, but it does not switch the effect off. The best protection is not simply knowing a lot, but having a habit of forming your own estimate before you hear anyone else's.

Experts do have one real advantage, though. Because they have a strong sense of the usual range, they are better at noticing when an anchor is wildly unrealistic. A seller who asks an experienced buyer for ten times the normal price will usually be laughed at. Anchors are most powerful when they fall within the range that seems possible, slightly high or slightly low, which is exactly where most real-life anchors are placed.

Four Lives and Their Anchors

Tariq is thirty-one and has been offered a new job. On the phone, the person hiring him asks, "What salary are you expecting?" Tariq has not prepared for this. He remembers that his current salary is modest, and he names a figure only a little above it. The person agrees immediately, which tells Tariq too late that he could have asked for more. His own past salary had become his anchor. For his next job search, Tariq does some research first. He learns the usual range of pay for similar work from public information, decides on his own figure in advance and writes it down. In the next conversation, he does not start from his old salary. He starts from what the work is worth.

Mariam is forty-six and teaches at a secondary school. One evening she marks thirty essays in a row. She notices that the essay she marks right after a brilliant one always seems weak, and the one after a very poor one always seems better than it is. Her judgment is being pulled by whatever she has just read. Mariam changes her method. She reads all the essays quickly first, without marking, to get a sense of the whole range. Then she marks them using a written list of what makes a strong answer, so that her measuring stick comes from the list rather than from the essay before.

Kwabena is sixty-nine and needs a new roof on his house. A friend tells him over dinner that his own roof cost a certain amount. The next week, a builder gives Kwabena a quote that is close to that amount, and Kwabena feels it must be fair. Then his niece suggests that he ask two more builders. Their quotes are noticeably lower, for the same work. Kwabena realizes that his friend's number, for a different house with a different roof, had quietly become the price he expected to pay. He chooses one of the lower quotes after checking that the builder is reliable.

Sofia is fifteen and has started making earrings from beads and wire, which she sells to classmates and neighbours. When she first showed a pair to an aunt, the aunt said kindly, "I would pay a little for those," and named a very small amount. Sofia priced every pair at that amount for months, even as her skills improved and people asked for more designs than she could make. Her aunt's casual guess had become the price of her work. When a friend's mother, who sells crafts at fairs, looked at the earrings and asked how long each pair took to make, Sofia worked out the cost of her beads and time for the first time. She raised her price, and her customers kept buying.

In each case, the first number was not chosen on purpose as a trick. Tariq's old salary, the brilliant essay, the friend's roof and the aunt's kind guess simply happened to come first. That is what makes anchors so common. Many of them are not set by clever sellers. They are set by chance, by memory and by the order in which things happen.

Practical Steps

First, make your own estimate before you hear anyone else's. Before you look at a price, ask for a quote or enter a negotiation, take a moment to decide what you think something is worth or how long it should take. Write the number down. Then, when you hear the other person's number, you have an anchor of your own to compare it with.

Second, gather a range, not a single number. One quote, one price or one opinion becomes a strong anchor. Three or four give you a sense of the real range. For larger decisions, such as repairs, rents or wages, reliable public information about typical ranges is one of the best protections against an anchor that is far too high or far too low.

Third, think of reasons for the opposite. When someone gives you a number, deliberately ask yourself why the true answer might be very different. If a seller says a vase is worth a hundred, ask yourself why it might be worth only twenty. Studies on reducing anchoring suggest that actively considering the opposite is one of the more helpful methods, because it brings different evidence to mind.

Fourth, be careful with "was" prices. When you see an original price crossed out, cover it with your thumb and ask: "Would I want this at the price shown, if I had never seen the higher number?" The higher number may be real, but it is not information about what the item is worth to you.

Fifth, think about who speaks first. In some negotiations, naming the first number can help you, because it anchors the discussion. This works best when you have done your research and know what is reasonable. If you know little, it may be wiser to let the other side speak first and then compare their number with the range you have found. In either case, do not let a number you hear feel more important just because it came first.

Sixth, refresh your measuring sticks. When you judge a series of things, such as job applications, essays or flats to rent, use a written list of what matters before you start. This keeps your judgment tied to your own standards rather than to whichever example you saw last.

When Anchors Help

Anchoring is not always harmful. Like most mental shortcuts, it exists because it is often useful. When we really know nothing about a quantity, starting from a sensible number and adjusting is a reasonable way to think. A rough figure from a trusted source is usually better than a wild guess.

Anchors can also help us keep good habits. People who set a clear amount to save each month, a time to go to bed or a number of pages to read each evening are, in a sense, giving themselves a helpful anchor. The number becomes a reference point that guides later choices. This works best when the number was chosen carefully, at a calm moment, for good reasons.

Some people also use anchors to protect themselves. A family that agrees in advance on a firm limit for a large purchase, before visiting any shop, gives itself an anchor that a seller's numbers cannot easily move. A person who decides before a celebration how much they will spend on gifts is setting a helpful starting point for every later choice that evening. In both cases, the anchor comes from a calm conversation at home rather than from the pressure of the moment.

The aim, then, is not to fight every number you hear. It is to notice where your starting numbers come from. Did this number come from careful research, from my own experience or from a trusted source? Or did it come from a seller, a passing comment or simply the first thing I saw? Good anchors are worth keeping. Accidental ones are worth replacing.

A Short Exercise: Your Own Number First

This exercise practises the most useful protection against anchoring: forming your own estimate before you hear anyone else's. It takes about fifteen minutes and can be repeated whenever you face a decision involving numbers.

Think of something you will need to buy, pay for or plan in the coming months. It might be a repair, a gift, a trip, a piece of furniture or a service. Before you look at any prices or ask anyone, write down three numbers on a piece of paper. The first number is the lowest price you think would be reasonable. The second is the highest price you think would be reasonable. The third is the price you expect to pay. Write a sentence next to each number explaining where it comes from: your past experience, something a friend said, an advertisement you saw or simply a guess.

Now look at the sources. Which of your numbers come from real information, and which come from a single comment or a single advertisement? If your expected price comes mainly from one source, that source may be acting as an anchor.

Next, gather information from at least three independent places, such as different shops, public price lists or people who have recently bought the same thing. Write down these numbers too. Compare them with your first three. Did your first guesses fall inside the real range? Were they pulled toward one particular source?

Finally, make a decision rule for yourself before you speak to any seller. For example: "I will not pay more than this amount, and if the first price I hear is higher, I will thank them and check elsewhere." Write it down and keep it with you.

You can practise the same skill in a playful way with friends or family. Before looking up the answer, each person writes a guess for some quantity, such as the height of a tall building, the age of an old tree in a park or the length of a well-known river. Then one person reads out a number, either high or low, and everyone writes a second guess. Compare how much the second guesses moved. It is a light-hearted way to see that nobody is free from the pull of a number spoken aloud.

Questions People Often Ask

"If anchors are so powerful, should I always name the first number in a negotiation?" Not always. Naming the first number can help when you know the usual range well, because your number sets the starting point. But if you know little, you might name a number that is far too low or far too high, and then you have anchored the discussion in the wrong place yourself. A safer rule is this: do your research first. If you are well informed, consider speaking first with a number at the favourable end of the reasonable range. If you are not, let the other side speak first, and then compare their number calmly with the range you found.

"How can I tell whether a 'was' price was ever real?" Often you cannot tell, and that is exactly why it is wise to ignore it. What matters is whether the price you will actually pay is fair for what you are getting, compared with other places and with what the item is worth to you. If you have checked a few other sources and the price is good, buy it with a calm mind. If you have not, the crossed-out number gives you no reliable information.

"Do anchors work on time as well as on money?" Yes. If you are told that a task usually takes an hour, you will tend to expect it to take about an hour, whether that is realistic or not. This is why first estimates in projects and plans deserve care. A rough guess spoken early in a meeting can quietly become everyone's expectation. When someone gives an early estimate, it can help to say out loud, "That is a first guess. Let us check it," so that the number stays a draft rather than becoming an anchor.

Chapter Summary

Anchoring is the tendency for a number we hear first to pull our later estimates toward it. We adjust away from the anchor, but usually not enough. Anchors that seem relevant, such as asking prices, first offers and quotes, have a strong and well-documented pull; completely random anchors sometimes have an effect too, but less reliably.

Anchors appear in prices, menus, negotiations, suggested amounts and estimates of time, and even experienced professionals are affected, often without noticing. The best protections are to form your own estimate first, gather a range rather than one number, think of reasons for the opposite and use written standards when judging a series of things.

Not every anchor is harmful; a carefully chosen number can guide good habits. In the next chapter, we look at a related power: how the same facts, described in different words, can lead us to quite different choices.