Behaviour design: find the field that blocks

Why the classic techniques work, and how to match the technique to the barrier

Most behaviour design is a list of tricks: social proof, defaults, scarcity, and so on. But there is one mechanism underneath all of them, and once you can see it they turn into a single picture. An action happens when it wins an internal race. It fails to happen when one of four fields makes it too costly: safety, meaning, ability, or effort. So the skill is not knowing twenty tricks. It is working out which field is blocking, and then using the technique that works on that field.

The one idea

Every action runs as a race between routes, and the first usable route wins. A technique works by doing something to that race: making the wanted route cheaper, making the unwanted one more expensive, or shifting which field decides.

The four fields are four ways an action can be too costly:

• Safety — does it feel safe to do?
• Meaning — can the person see why?
• Ability — do they know how, and can they?
• Effort — is it easy enough?

Diagnosis comes first

The most common mistake is reaching for your favourite trick without asking which field is actually in the way. An initiative that assumes an ability problem and delivers a course, when the real barrier is safety, solves the wrong friction. It is the same point as the compliance page: more information only helps if knowledge was what was missing.

So start by finding the barrier. Do people hesitate because it feels unsafe? Ignore it because they cannot see the point? Want to but do not know how? Or know exactly what to do, but it is too much hassle? Once you can name the field, it points to the technique itself.

The techniques, sorted by field

Safety: make it safe to act

• Social proof — "others did it and it went fine" lowers the uncertainty. It works because we forward-model other people as a source of what is safe; the weight we give the source is a reliability estimate, not a contradiction.
• Authority and expertise — a credible sender lowers the doubt about what the right thing to do is.
• Guarantees and easy returns — make the action safe to undo, so it does not feel like a leap.
• Familiarity — the known reads as low-risk, so hesitation drops.

Meaning: give the action a reason

• Identity and consistency — when the action fits who you are, or something you already said, the route wins more easily.
• A real "why" — a reason, not just an instruction, strengthens the route toward the wanted action.
• Reciprocity and fairness — social and moral obligations pull, because they run as friction between people.
• Stories and concrete examples — a lived example is a shorter race than pure theory, so it lands faster.

Ability: make the next step clear and doable

• Small steps — a step small enough to be taken makes the route wider; a big step that never gets taken does nothing.
• Help right when it is needed — a short checklist in the moment beats a thorough course given in advance, because it triggers one specific route exactly when it is needed.
• Salience and feedback — make the right route visible in the moment; that steers which race even opens.

Effort: lower the price of the right thing

• Defaults and fewer steps — the pre-chosen option wins because it is cheapest; remove a step and more people reach the end.
• Clear the obstacles — the most overlooked improvement is removing hassle, not adding motivation.

Timing: the first thing and the scarce thing

• Anchoring — the first number or word shapes the rest of the race.
• Scarcity and loss — a loss weighs more than an equivalent gain, so "this is going away soon" raises the price of not acting now.

Goal shape: make the goal so small it cannot fail

An all-or-nothing goal says the week either succeeds or fails. You set out to run three times, miss once, and then you don't run at all. It looks like a failure of will, but it is the shape of the goal. Once the first run was missed, "success" was out of reach for the rest of the week, and then there is no gain left to collect by running tomorrow while the effort is exactly the same. The route stops winning, and the feeling that it may as well not matter comes afterwards; it describes what happened, it is not the reason.

The fix is to make the goal a daily minimum instead of a weekly number. Not "three times a week", but "something every day, even if it is only ten minutes". Set that low, it can hardly be failed, because the only thing that counts as failing is doing nothing at all, and you cannot go lower than that. Then there is always a route that still counts, and a bad day cannot close the week. It is the same amount of training, but it cannot collapse.

Watch one thing: the rule "no sweets" is not the same. There the first piece of chocolate does not remove the gain, it removes the opponent. The rule was the only thing making the eating-route lose, and once it is broken the eating-route wins every time after that. A minimum does not help here; what fits is another short route to reach for instead, and a limit on the episode that is not the rule itself. The phenomenon is well known: all-or-nothing thinking is one of Aaron Beck's cognitive distortions, and the collapse after the first break is called, in relapse research, the abstinence-violation effect.

Friction the other way: designing against the unwanted

Friction is not always the enemy. To make a wrong or irreversible action rarer, make it more expensive on purpose: a confirmation step before deleting, a small obstacle before an impulse buy, an extra click before something is sent. The rule is that friction against the unwanted should feel like care, not bureaucracy.

We weigh what is there, and miss what is absent

A pattern runs through several of the techniques: what is present counts, and what is missing does not. We build a coherent story out of what we can see, and rarely weigh in what we were not told. Daniel Kahneman called it "what you see is all there is" (Thinking, Fast and Slow, 2011). It is not carelessness. A route can only run toward something that exists; an absence is not a route that can lose, it is one that was never set up.

That explains several of the techniques above with one reason. Defaults work not only because they are easy, but because the alternative is never presented as something to choose against. Salience and cues work because they make a route present at the moment the choice is made; an option that is not visible right now does not compete. And social proof works partly for the same reason: other people's action is visible, their hesitation is not.

It points to the technique that is easiest to miss: make the absent present. Show what is being turned down, not only what is chosen. Show the ones who dropped out, not only the ones who succeeded. Put a number on the denominator, not only on the numerator. It is the cheapest way to move a decision, because it does not require anyone to try harder, only that the other option is in the race at all.

Why this is not just another list of tricks

Most of the techniques are not mine. Social proof, authority, reciprocity, scarcity and consistency come from Robert Cialdini's work on influence. Anchoring and the fact that a loss weighs more than an equivalent gain come from Kahneman and Tversky. Defaults and removing effort are the core of Thaler and Sunstein's nudges. What is new is that they get one shared explanation, and that each one points to a field, so you can choose by what is actually blocking instead of guessing. And several of the techniques turn up again in language models, where you can measure them directly: anchoring, fairness, and that pushing too hard backfires (see What language models reveal about humans).