Why do I keep relapsing on day 2 or 3?

Most attempts to quit end on day two or three, not day ninety. The usual causes are deciding how to handle an urge in the middle of it, and watching the count instead of the day. Neither is a willpower problem, and both have a fix with some evidence behind it.

Why day three and not day thirty

Because the first few days decide most attempts, and very little advice is aimed at them.

In what people write when they ask for help with this, one sentence keeps coming back in different words: I keep failing after two or three days. Hardly anyone describes losing their way in week eight or getting bored around day forty. It is two or three days, over and over, sometimes since they were fourteen.

That makes the wall a feature of the problem rather than a personal flaw. Almost everything written about quitting skips it and starts at day sixty.

The loop, and which part of it is the engine

The relapse starts the loop. The hour afterwards keeps it going.

The sequence people describe is consistent. Something sets it off, they give in, and then they feel awful. Feeling awful makes everything else hard, so the studying does not get done, the gym does not happen and the day gets written off. A written-off day has nothing in it, and the next day starts from there.

The wanting sets it off. What someone tells themselves afterwards keeps it running, and that is the half with room to act on. It is the difference between one bad night and a bad month.

The part where thinking about it is what breaks it

Watching the count can cause the relapse. This cuts against how most apps in this area work, ours included.

One person working out their own pattern noticed why they kept failing on day three. They were thinking about the count constantly: about the number, and about losing it. A day spent guarding a number is a day spent thinking about the thing you are trying not to think about.

The most agreed-with advice on this lands in the same place from the other side. How you feel about the number does not matter. What you choose today does.

So treat the count like a tally on the back of an envelope. It records what you have done. When it starts to feel like something to defend, move your attention back to the next few hours.

What to do on day two

Decide your response before the urge arrives, and write it down while deciding is easy.

It is the least exciting advice here and has the best evidence. Choosing a response in advance, in an if-this-then-that form, has been tested on drinking and on smoking. Malaguti and colleagues pooled 21 trials in Drug and Alcohol Dependence in 2020 and found it helped with both, at around g = 0.31 each. The effects are small, the authors say so, and nobody has run the same test on this. It beats deciding in the moment. It does not do the work for you.

In practice that means specific sentences, even slightly embarrassing ones, because embarrassing and specific beats dignified and vague. When the urge hits at night I will get out of bed and go to the kitchen. When I have been on my phone for an hour I will leave it in another room. The more exactly a plan names the moment, the more use it is when the moment arrives.

Write it where you will see it. Harkin and colleagues, pooling 138 studies and nearly 20,000 people in Psychological Bulletin in 2016, found that keeping track of progress helps people reach goals, at d+ = 0.40, and that it helped more when people wrote things down instead of only thinking about them. That covers health behavior in general, such as eating and exercise, not this specifically. It is the reason to put the plan on paper and mark the day. What that looks like inside the minute itself is in the first sixty seconds of an urge.

The step that does not depend on being sharp at 1am is removing the easy path in advance. Your iPhone has a content filter built into Settings that you can switch on yourself in about five minutes, for free. It is the highest-value step on this list, because it works when nobody is at their best. The steps are in how to block porn on an iPhone.

What to do when it happens anyway

Treat it as one day. As far as anybody has measured, that is what it is.

Gardner, Lally and Wardle, following people building simple daily habits, summed it up in the British Journal of General Practice in 2012, and that paper is free to read. Missing the occasional day "did not seriously impair the habit formation process", and the gains resumed afterwards.

The limits sit next to that. Not seriously impaired is not the same as no effect. One missed glass of water is not this. And nobody has run the equivalent test on quitting porn. It is the closest evidence there is that one day is one day, and it is not a free pass.

The part that does the real damage has a name, from work on people quitting smoking. Curry, Marlatt and Gordon described it in 1987: what predicted whether one slip turned into a full return was not the slip but what the person decided the slip meant about them. I am weak. I always do this. I was never going to manage it. Those sentences do more damage than the night did. It is a model rather than a measured effect, it was built in smoking and drinking research, and nobody has tested it here.

So the useful question the morning after is a small one: what am I doing in the next two hours? Day-sized questions are the ones anybody can answer.

What gets you to day ten

Three things, and trying harder is not one of them.

Take the easy path away while calm, so the decision is not waiting at midnight. Choose the response in advance and write it where you will see it. And keep marking the days, bad ones included: a month with four bad days in it is still a month with twenty-six good ones, and that is invisible if the page gets thrown away every time.

If the last app you tried came off your phone inside a week, that is a separate problem: why porn blockers stop working. How long this takes, with the measured median and range, is in how long it takes to break a habit. If the thing you keep quitting is the plan rather than the porn, that is the majority, and the one trial that measured it is in most people quit the programme, not the porn.

What the next three days are for

Asked why they are doing this, people almost never say "to be a better person". The answer is specific, and usually something they used to have: the grades they used to get, a plan they keep not starting, something they were good at before their attention went elsewhere.

The next three days are for that, not for day ninety.