Chapter 10: Following the Crowd

Two Food Stalls

Tendai is twenty-two and has come with two friends to an outdoor music festival. It is early evening, the music has not yet started and they are hungry. Along the edge of the field, there is a row of food stalls. Two of them sell almost the same thing: grilled flatbreads with vegetables and sauce. The prices on the boards are the same. The smell from both is wonderful.

In front of the first stall, there is a long queue, perhaps twenty people. In front of the second stall, there is nobody at all. A young man behind the counter is waiting with his hands on his hips, looking hopefully at people as they walk past.

Tendai and his friends hardly discuss it. They join the long queue. "If that many people are waiting, it must be better," says one friend. Twenty-five minutes later, they reach the front, buy their flatbreads and find that they are good, though not remarkable. On the way back to the stage, Tendai notices that the second stall now also has a queue. A few people tried it, others saw them waiting and joined them, and soon it too looked popular.

Afterwards, Tendai wonders about the choice. Was the first stall really better? Perhaps it was. Perhaps the first people to arrive knew something. But perhaps the first stall had simply opened a few minutes earlier, so that its queue started first, and everyone after that, including Tendai, followed the queue rather than the food. This chapter is about how strongly we are guided by what others do, when that guidance is wise and when it leads a whole crowd in the wrong direction.

Social Proof

When we are unsure what to do, one of the most natural things we do is look at other people. If many people are doing something, we take it as evidence that it is a good thing to do. This tendency is called social proof: taking what others do as a sign of what is right, true or good.

Social proof is not foolish. In fact, it is often an excellent shortcut. Other people may have information we do not have. If a restaurant is full, it may really be good. If everyone on a train platform is suddenly moving to the other side, they may have heard an announcement you missed. If the local people in a neighbourhood avoid a certain street at night, they may know something about it. For most of human history, copying what the people around you did was a sensible way to learn which foods were safe, which paths were dangerous and which tools worked best.

Social proof is especially strong in certain situations. It is strongest when we are uncertain, because then the behaviour of others is our best clue. It is strongest when the other people seem similar to us, because their choices seem more relevant. And it is strongest when there are many of them, because many people doing the same thing seems hard to argue with.

Researchers often distinguish between two reasons we follow others. The first is to be right. We follow the crowd because we think it knows something, as Tendai did at the food stall. This is called informational influence. The second is to fit in. We follow the crowd because we want to be accepted, liked or at least not stand out. This is called normative influence. In everyday life, the two often work together, and it can be hard to tell which one is moving us at a given moment.

You can often feel the difference between the two kinds of influence if you pay attention. Informational influence feels like learning: "Oh, so that is how it is done." It usually changes what we privately believe, not only what we do. Normative influence feels more like pressure: "I had better do what they do, or they will think I am strange." It can change what we do in public while leaving our private opinion untouched. Both are normal. But knowing which one is at work helps you decide whether to follow. Learning from others is often wise. Going along only to avoid discomfort is sometimes wise and sometimes not.

The Lines on the Card

In the 1950s, a psychologist carried out a series of experiments that became some of the most famous in the history of the subject. Small groups of people sat around a table and were shown cards. One card showed a single line. Another card showed three lines of different lengths. Each person in turn was asked to say aloud which of the three lines was the same length as the first. The task was easy. When people did it alone, they almost never made mistakes.

But in the experiment, only one person at the table was a real participant. The others had been told in advance to give the same wrong answer on certain rounds. The real participant, sitting near the end of the row, heard person after person confidently name a line that was clearly wrong. Then it was their turn.

Many participants went along with the group at least once, naming the wrong line even though their own eyes told them otherwise. Some later said that they began to doubt their own judgment. Others said they knew the group was wrong but did not want to stand out. These results are often described as proof that people will conform blindly.

But there is an important part of the story that is often left out. Across the experiments, most of the answers given by real participants were correct. Many people never went along with the group at all, and most people stood by their own judgment on most rounds. The experiments also found that when even one other person at the table gave the correct answer, the pressure to conform dropped sharply. A single ally made it much easier to trust one's own eyes.

Later researchers repeated versions of these experiments in many places and at many times. The amount of conformity varied between groups and between periods, which suggests that it is not a fixed feature of human nature but depends on the situation and the social setting. The fair summary is this: the pressure of a unanimous group is real and can push people to say things they do not believe, but most people resist it much of the time, and one other dissenting voice can make a great difference.

What Most People Do

Social proof can be used to change behaviour on purpose. One way is to tell people what most others like them are doing. Information about what most people do is called a descriptive norm.

In a well-known type of programme, households received regular letters or reports comparing their energy use with that of similar homes nearby. Households that used more than their neighbours were encouraged by the comparison to use less. Large trials of such programmes have found that they do reduce energy use, on average, by a modest but real amount, and that the effect can last for some time when the reports continue.

These programmes also revealed a problem. Households that were already using less energy than their neighbours sometimes began to use slightly more after receiving the comparison. Learning that they were below average seemed to give them permission to relax. When the reports added a simple sign of approval for low users, such as a small smiling face, this unwanted effect became smaller. This detail shows that descriptive norms pull people toward the middle in both directions, unless something else, such as approval, is added.

Descriptive norms also explain why some well-meaning messages backfire. A poster that says "Too many people drop litter in this park!" is trying to discourage littering. But it also tells readers that littering is common, which can make it seem normal. A message that says "Most visitors take their litter home" uses the same social pull in a more helpful direction. When you want to encourage something, it is usually wiser to show that many people already do it than to complain that many people do not.

Descriptive norms are not only used by organizations. Families, friends and schools use them all the time, often without noticing. A parent who says, "In our family, we always say thank you to the bus driver," is describing a norm. A teacher who says, "Most students in this class hand in their homework on time," is using one too. Simple statements like these can be powerful because they tell people what is normal here, among people like them. They work best when they are true, because people quickly notice when a claimed norm does not match what they see.

When the Crowd Is Wise, and When It Is Not

In the early twentieth century, the scientist Francis Galton described a guessing competition at a country fair. Visitors paid a small fee to guess the weight of an ox after it had been slaughtered and prepared. Hundreds of people took part, some of them farmers and butchers, most of them not experts at all. When Galton looked at all the guesses together, he found that the middle value of the crowd's guesses was remarkably close to the true weight, closer than most individual guesses.

This kind of finding has been repeated in many forms. When many people make independent guesses about a quantity, their individual errors often cancel out, and their average can be surprisingly accurate. This is sometimes called the wisdom of crowds. But notice the word "independent". The crowd at the fair was wise because each person wrote down their own guess without knowing the others' guesses. Their errors pointed in different directions.

When people copy each other instead of judging independently, the crowd loses this wisdom. Imagine that the first few people at the fair had shouted out their guesses, and everyone after them had been influenced by those first numbers. Then the errors of the first few would spread through the whole crowd instead of cancelling out. This is what happened at the food stalls. The queue may have begun for a good reason or for no reason at all, but once it began, later people followed the queue rather than their own judgment of the food.

When many people follow the choices of those before them, without adding information of their own, the result is sometimes called an information cascade or, more simply, herding. Herding can make a restaurant famous, a song popular or an idea fashionable for reasons that have little to do with its quality. It can also cause sudden rushes, in which many people buy, sell or abandon something at the same moment because others are doing so. It is worth knowing that herding is one reason why prices of popular things can sometimes rise and fall sharply.

So the crowd can be wise or foolish depending on how it forms its view. A crowd of people who each think for themselves, from different backgrounds and with different information, is often a good guide. A crowd of people who are mostly copying one another is a much weaker guide, however large it is.

Three Lives in the Crowd

Wanjiru is fifty-nine and belongs to a lively group of friends who meet every week. Recently, several of them have started buying a new kind of expensive dietary supplement, which a well-known personality has praised online. They talk about it enthusiastically, and Wanjiru feels left out. She is about to buy some when she remembers that her own doctor has always advised her to ask before taking anything new, because of a medicine she already takes. She asks at her next appointment, and the doctor explains that this supplement is not suitable for her. Wanjiru tells her friends, and two of them decide to ask their own doctors too. The friends' enthusiasm was sincere, but it was based on the same single source, copied from person to person, rather than on independent information.

Lukas is fifteen. At his school, a new online challenge has become popular: students film themselves doing a risky stunt and share it. Many of his classmates have done it, and the videos have many views. Lukas feels the pull strongly. If everyone is doing it, how dangerous can it be? But he also notices that he has never seen a video of someone getting hurt, and he realizes that people would hardly share those. His older cousin, who works in a hospital, tells him that she has treated injuries from exactly this kind of stunt. Lukas decides not to do it and, to his surprise, discovers that two of his friends had been feeling the same doubt and were relieved when he said it aloud.

Svetlana is forty-four and works on a team planning a new project. In a meeting, the team leader proposes a plan, and the first two people who speak praise it. Svetlana has serious doubts about the timeline, but she sees everyone nodding and says nothing. After the meeting, a colleague quietly tells her that he had the same worry. They agree to raise it together at the next meeting. When they do, two other team members admit that they had doubts too. The plan is improved. Like the people at the table with the lines on the cards, Svetlana found that one ally made it much easier to speak.

In each of these lives, the crowd was not simply right or wrong. Wanjiru's friends meant well, Lukas's classmates were not foolish and Svetlana's team was not careless. But in each case, many people were following one another rather than thinking independently, and a single voice that added new information or honest doubt changed the direction of the whole group.

The Courage to Be First

One of the most useful lessons from research on social influence is how much difference one person can make. In the line experiments, one ally reduced conformity sharply. In Svetlana's meeting, one quiet conversation led to a better plan. In Lukas's class, one honest doubt freed others to share theirs.

This works in emergencies too. You may have heard of the bystander effect: the idea that people are less likely to help someone in need when many others are present, because each person assumes someone else will act or takes the inaction of others as a sign that nothing is wrong. Studies in laboratories have found this effect. But later research, using real recordings of public conflicts, found that in many real situations, at least one bystander does step in, and that larger groups are often more likely to contain someone who helps. The picture is more hopeful than the famous version suggests.

Still, the first step is the hardest, and practical advice from many public safety guides is the same. If you see someone who needs help, do not wait for the crowd to decide. If you need help yourself in a crowded place, speak to one specific person directly, for example, "You, in the green jacket, please call for help," rather than calling out to everyone. A direct request breaks the spell of waiting for others.

Being first does not require you to be loud or dramatic. It can be as simple as saying, "I am not sure about this. Can we think about it again?" or "I will try the stall with no queue." Often, the people around you are waiting for someone to say exactly that.

It also helps to remember that the people who first said "Wait" in groups are often remembered with gratitude afterwards, once the group can see that the doubt was justified. In the moment, speaking up can feel lonely. Over time, it tends to build trust, because people learn that you will tell them the truth when it matters.

Practical Steps

First, ask whether the crowd is informed and independent. When you notice yourself following what many others are doing, ask: Do these people have real information? Did they decide separately, or are they copying each other? A crowd of independent, informed people is a good guide. A crowd copying a single source is not.

Second, form your own view before you look at others'. Before reading ratings, reviews or comments, take a moment to decide what you think or what you are looking for. Then look at what others say and compare. This keeps your own judgment in the picture.

Third, look for reasons, not just numbers. A large number of people choosing something tells you that it is popular. It does not tell you why. Read a few reviews or ask a few people for their reasons, and pay attention to whether those reasons apply to you.

Fourth, look for what is missing. Crowds tend to show their successes and hide their failures. Lukas never saw videos of injuries, and enthusiastic friends rarely mention the purchases they regret. Ask yourself what you would not see, and try to find it.

Fifth, be the ally. If you have a doubt in a group, there is a good chance that someone else has it too. Saying it calmly, or quietly finding one other person who shares it, can help the whole group think better.

Sixth, use social proof for good. If you want to encourage a helpful behaviour in your family, school or workplace, show that many people already do it, and add a sign of approval. Avoid messages that complain about how many people behave badly, since they can make the bad behaviour seem normal.

A Short Exercise: Independent First

This exercise shows you, in a small way, how independent judgments differ from copied ones. It works best with a group of four or more people, such as a family, a class, a club or a team at work. It takes about fifteen minutes.

Choose a question with a number for an answer that nobody in the group knows exactly, but that can be checked later. It might be the number of sweets in a jar, the height of a nearby tree or building, the distance between two familiar places or the weight of a heavy bag. Give each person a small piece of paper and ask them to write down their guess privately, without talking.

Collect the papers and work out the middle value of all the guesses. Then, in a second round, ask people to say their guesses aloud one at a time, in order around the room, before writing down a final guess. Work out the middle value of the second round too. Finally, check the true answer.

In many groups, the first round, where guesses were independent, comes closer to the truth than most individual guesses. In the second round, the guesses often cluster around whatever the first few speakers said, and the group loses some of the advantage of its different views. Not every group shows this pattern clearly, but the discussion afterwards is usually lively.

Talk together about what happened. Did hearing others change your guess? Did the first speaker have more influence than the last? How might the same pattern appear in your meetings, family decisions or online reading? Many teams find it helpful to adopt a simple rule from this exercise: before discussing an important question, everyone writes down their own view first.

If you do not have a group, you can do a version alone. Before you read the reviews of something you are thinking of buying, write down your own guess of how good it is likely to be and why. Then read the reviews and notice how much they move you, and whether they gave you new information or simply new pressure.

Questions People Often Ask

"Is it wrong to want to fit in?" Not at all. Wanting to belong is a deep and healthy human need. Following the customs of the people around you is often a form of respect and a way to build trust. The question is not whether to follow others, but when. In matters of taste, manners and small daily choices, fitting in is often perfectly fine. In matters of safety, honesty, health or important decisions about your life, it is worth checking whether the crowd's choice is really right for you.

"How can I trust online ratings?" Online ratings can be useful, but they are shaped by herding. Early ratings tend to influence later ones, people with very strong feelings are more likely to write reviews and some ratings may not be genuine. It helps to read several reviews in detail, especially those of medium length that describe specific experiences, and to look at a range of sources rather than only the overall score.

"Why is it so hard to disagree with a group?" Because disagreeing with a group has real social costs, and our minds feel them strongly. We may worry about being seen as difficult, about losing the group's goodwill or about being wrong in front of others. These worries are natural. It often helps to disagree with a question rather than a statement, such as "What would happen if the timeline slipped?", and to find one ally before a meeting. Over time, groups that welcome honest disagreement usually make better decisions, and people in them feel more respected.

Chapter Summary

Social proof is the tendency to take what others do as a sign of what is right. It is often a sensible shortcut, especially when we are uncertain, and we follow others both to be right and to fit in. Classic experiments showed that the pressure of a unanimous group can lead people to say things they do not believe, but also that most people resisted much of the time and that one ally reduced the pressure sharply.

Information about what most people do can change behaviour, often in modest but real ways, and messages work better when they show the helpful behaviour as common. Crowds are wise when people judge independently and foolish when they copy each other, which can lead to herding.

Asking whether the crowd is informed and independent, forming your own view first, looking for reasons and for what is missing, being the ally and using social proof for good all help. In the next chapter, we turn from the crowd to ourselves, and to the surprising confidence we place in our own judgments and plans.