‹ Back to blog

The if-then sentence, and why writing one down does anything

July 10, 2026 · 6 min read

The problem with ordinary intentions

I am going to order less food this month.

That sentence has a large hole in it: it says nothing about when and nothing about what. It is a wish about an average, and there is no specific moment in the month at which you are obliged to do anything.

So when the real decision arrives — 1am, tired, phone already in hand — the sentence is not there. It was never attached to anything that could summon it.

What an implementation intention is

The psychologist Peter Gollwitzer proposed a different form of intention, which he called an implementation intention. Its structure is fixed:

If situation X occurs, then I will do action Y.

That is all. One conditional clause, one specific action, both written or spoken in advance, before the situation shows up.

Research on implementation intentions generally finds that this form is associated with higher follow-through than goals stated in general terms. This is a line of behavioural psychology research, not a clinical claim, and the size of the effect varies considerably by behaviour, by context and by person.

Why it behaves differently

The usual explanation runs like this: when you attach an action to a specific cue, you move the decision from the moment it is needed back to a moment when you were calm.

At 1am you are not in good condition to weigh options. You are tired, and every choice costs more than it usually would. If that is when you first have to invent a plan, the easiest plan wins, and the easiest plan is always the app your thumb already knows.

But if the plan was chosen earlier, while you were alert and nothing was pressing, then at 1am there is nothing left to decide. You only have to recognise the situation and follow what you already wrote.

An if-then sentence converts a decision into a recognition. Recognition is far cheaper than deciding.

Writing one that works

A good sentence has three properties.

  • The situation is specific enough that you notice it while it is happening. Not "when I feel like snacking" but "when I pick up my phone after 11pm and go to open a delivery app".
  • The action is specific enough that no further thinking is required. Not "I will hold back" but "I will open Table for None and place an order there".
  • It is short. If you cannot recall it at 1am, it does not work, no matter how well it was written.

A few that are concrete enough to run:

  • If I open a delivery app after 11pm, then I open Table for None first and place an order there.
  • If I finish an episode and reach for my phone, then I place a simulated order before doing anything else.
  • If I catch myself scrolling a menu without being hungry, then I switch over to the Table for None tray.

This is not willpower

The most important point, and the one most often misread: implementation intentions do not work by making you more determined.

They work by reducing the number of decisions you have to make at precisely the hour when you are least equipped to make good ones. That is why they do not require you to become a more disciplined version of yourself before you are allowed to start.

If one night you do not follow the sentence, the sentence is not broken. It is still sitting there for the next night. No streak was lost, because there is no streak.

Inside the app

Table for None has one place to write this sentence, and it is the only thing in the app that resembles a goal. You write it, it sits there. No reminder notification, no day count, no prompt asking whether you managed it.

We do not check whether you followed through, because we have no way of knowing and no business knowing. That data lives on your device and goes nowhere. The sentence is yours, and whether it works does not depend on anyone watching.

A sentence written in advance is not stronger than resolve. It is just present at the hour when resolve is not.