There is definitely an ethical component to nudging. A good and ethical nudge is designed to help someone make a decision they want to make themselves. A bad and unethical nudge makes it harder for a person to do what they want.

Unethical nudges used to be called “shoves”, but the term gaining more steam is “dark patterns“. Examples abound, from tiny companies you’ve never heard of to places like LinkedIn and Amazon.

The rule of thumb for determining whether a nudge is “dark”, is by simply thinking whether you would be happy if your local supermarket made you do the same thing. Would you be happy if whenever you went grocery shopping, the manager put items she thinks you might like in your basket when—you weren’t looking? What if they made you walk through a tedious and deceptive maze every time you tried to leave?

Don’t nudge anyone with methods you don’t want to be nudged with yourself.

Indecision is a problem everyone faces. Recently, some have taken to call it “choice paralysis,” with the explanation being that it’s “too much choice” that leaves us paralyzed and afraid to make a mistake.

My explanation is that choice paralysis has two separate causes: first, a lack of information about the products; and second, a lack of information about our own preferences.

For an example of the first case, think of the familiar situation of deciding what to get at a Chinese restaurant. A thick menu, written in a foreign language, leaving you unsure what anything actually is. You know you want something that isn’t spicy and doesn’t have fish guts in it. But the menu is not helping you decipher this information.

A high-end coffeeshop might face the same problem. While experienced customers might confidently ask for a tall-blonde-mocha-chai-latte-with-two-pumps-of-vanilla, someone new to the experience might try to individually decipher all 85,000 different combinations of roasts, toppings, and sizes.

The telltale sign of this sort of indecision is that the decision-maker is asking a lot of questions. Questions that are not answered simply by looking at the menu. The solution here is simply to design a more informative menu. The key here is to break up the decision making process into a series of steps (“first, choose a size. Second, choose a roast. Third, choose how much foam,” or “first, choose your meat. Second, choose your veggies. Third, choose how spicy,” etc.).

The second type of indecision, which is caused by the decision-maker not knowing what they want, can be witnessed in fast food restaurants when someone is struggling between the Coke or the Cherry Coke. They know what both are, but they can’t decide which one they want at this moment. The reason for this is because they actually value both things equally. 

Think about it. If the choice was between Coke and cyanide, would they struggle as much with the decision? No, because a thirsty person would get a lot more out of life with the Coke than the cyanide. So the only logical reason a person would struggle with such a decision is because both choices are equivalent.

Thus in this situation, any choice will do. So if you recognize someone in this situation, you can try to create some dichotomy between the two options. For example, if you know the person likes to change things up, you can remind them of what they picked last time so they choose the opposite thing this time.

When you give people good news in spite of bad news, this lessens the impact of the bad news.

One example from Rory Sutherland: when a plane doesn’t pull up to the gate, this is generally considered to be bad news because you now have to take a bus, and who wants to sit in a bus after hours of being in a train?

Back to Sutherland’s story: a plane he was riding stopped of the gate. Instead of simply directing people to get on the bus, the pilot added new information: that the bus will take the passengers directly to passport control, so that now they don’t have to walk all that way while carrying their luggage.

But it’s always been true that the bus takes you to passport control directly. The pilot simply reminded the passengers of the story, and this way of framing the situation really helped the passengers feel better.

How often have you been waiting around for a train, or a cab, or a friend, and you became extremely anxious as you waited? Why are they late? How late are they going to be? Are they in danger? Are they incompetent? And so on.

Your anxiety isn’t being derived from the lateness itself, but the uncertainty that surrounds it. You have no idea whether you’ll be waiting five minutes or an hour. It’s this uncertainty that drives the fear and frustration. That’s why, as a business, you should try to limit not only your lateness, but also to give as much information about progress as possible.

Take, for example, the Toronto subway system. Up until a few years ago, you would get to the station, and you had no idea if you just missed a train, or if the next train is just around the corner. You start worrying about whether you’ll make it to your appointment on time. This is anxiety. Anxiety is bad.

But the Toronto Transit Commission solved this problem with a simple clock that estimates the time of the next train. While you might be unhappy to learn that the next train is running late, but you’ll still be happier than if you were left completely out in the dark.

Another way to think of this is to go back to your high school math tests. The more you show your work, the more brownie points you’ll earn from your customers.

Surely being logical has many benefits. Much of science and human progress owes its successes to the disciplined application of logic. Choosing logic tends to be a winning formula.

But should you ever choose to be illogical? I say yes. Consider the following.

Logic has universal rules. Aristotle wrote down a lot of them. The economist Ludwig von Mises even said that all human minds have the exact same logical structure, and to even imagine superlogic or semilogic is impossible.

So logic has rules, and all humans think about these rules in the same way. You can see where this would be helpful: for example, accounting. If someone started calculating revenues and costs in radically different ways than the norm, that would probably lead to a lot of problems. Similarly, if engineers suddenly decided that the logic of gravity doesn’t always apply, probably more bridges would fall down.

So when can we break the laws of universal logic? When it comes to dealing with people. Conventional business logic says that lower prices will lead to more sales. But it’s also possible that low prices lead to fewer sales: it all depends on how people perceive what the price says about the products in question.

Rory Sutherland points out that if an accountant tried to develop a product to compete with Coca-Cola, he would probably recommend something that would taste better and costs less. But in fact the most successful competitor to Coke has been Red Bull, which tastes awful and is much more expensive.

The only logic that applies to human decision making is that people will try to act reasonably to get what they want. But how people decide what they actually want is fundamentally, in the words of Ludwig von Mises, irrational. Trying to use logical to influence a fundamentally irrational decision is not an obviously best strategy.

If you’re trying to get ahead of the competition, using the same logical thought processes as everyone else will mean you’ll have same the same outcomes of everyone else. The only way to really get ahead of the pack is to through logic out the window.

So here is a simple flowchart of when you’ll want to employ conventional logic and when you’ll want to try to be a bit silly:

Is what you’re doing directly concerned with people’s opinions?

If no, use conventional logic just like everyone else.

If yes, try throwing conventional logic out the window and do something silly.

For more on this topic, see my (much shorter) post on math solutions vs. people solutions.

Economist Shengwu Li, currently at Harvard, has a new theory called “Obviously Strategyproof Mechanisms“.

The basic idea is this: there are many situations where “in theory” we expect different methods to result in identical outcomes. For example, if you want to auction off a car, theory says that having a so-called “English” auction (where the highest bidder wins) and a “second-price sealed bid” auction (where all the bids are secret, and the highest bidder wins and pays the second highest price) should both result in the same person winning.

But in practice, we don’t see this equivalence. What is the cause of this discrepancy? According to Professor Li, it is the fact that one strategy is “obviously better” better than the other.

What does an “obviously better” strategy mean? It means that the worst you can do under one strategy is still better than the best you can do under any other strategy.

Think about how the worst that your business has to offer is still better than the best alternatives for your customers.