Why this question is so hard to answer#
You are probably asking this at the worst possible moment to answer it. It is late, you are alone, and the only evidence available is your memory of an evening you spent trying not to pay attention to.
So you reach for the explanation closest to hand, which is almost always some version of something is wrong with me. It arrives fast, it feels true, and it gives you nothing you can act on.
There is a more practical problem with it. You are being asked to recall detail you never recorded. What time did the last four start? What was your stress like those days? Were you alone, or had you just come from being around people? Had you slept? Most people cannot answer those questions about last Tuesday, let alone across months. Not because their memory is bad, but because there was no reason to be taking notes at the time.
That gap is the whole difficulty. The question is answerable. You just do not currently hold the information the answer is made of.
The answer is not that you are weak#
Try it from the other end. If willpower were the thing deciding this, relapses would land more or less at random. Some Monday mornings. Some Saturday afternoons. Whenever your reserves happened to run low.
That is not what people find once they start writing it down. Relapses bunch up. Particular hours, particular days, particular states, and usually a narrower window than expected. Someone certain it was constant finds it is mostly two evenings a week. Someone who assumed stress was the driver finds their worst nights are the quiet ones.
Things that cluster have a cause that clusters with them. A character trait does not switch on at eleven on a Thursday night and switch off again by Saturday lunchtime. A situation does.
That is better news than it sounds. There is nowhere to go with “I am weak.” It is not actionable, and mostly it just hurts. A specific recurring situation is a thing with edges. It has a time, a lead-up, and a set of conditions that show up before it does. Those can be planned around, and planning around a known situation is a completely different task from trying to become a better person.
The situations that actually carry the risk#
Relapse-prevention research has been describing this for decades, mostly in the context of alcohol and drugs. The model most of it traces back to is Marlatt's, and its central claim is plain: relapse is not a random event or a moral failure. It is what happens when a high-risk situation meets someone with no ready response to it.
Three things do most of the work.
The situation itself. Some circumstances carry far more risk than others. Being alone with an unstructured evening. A particular room, a particular device, a particular hour. Low mood, or the hour after an argument.
Whether you have a response ready. This is the part that gets underrated. Meeting a high-risk moment with something specific to do is a different experience from meeting it and having to invent an answer while the urge is already running. The response does not have to be clever. It has to already exist.
What you expect it to do for you. In the moment, the behaviour is not competing with your long-term goals. It is competing with whatever else is available in the next ten minutes, and it promises to work faster than any of them.
Two caveats, both of which matter. That model comes out of substance-use research rather than work on pornography specifically, so what it supports is the mechanism and not a prediction about you. And it says nothing here about what happens afterwards, which has its own well-documented trap and is covered in our main guide on quitting.
The decisions before the decision#
Ask someone when they decided to relapse and they point at the last moment. Opening the browser. Picking up the phone.
That is almost never where it was decided.
Marlatt's model has a second half that gets far less attention than the first. Before the obvious high-risk moment there is usually a chain of smaller choices, none of which look like they have anything to do with it. He called them seemingly irrelevant decisions, and the name is exact. Each one is defensible on its own terms, and together they walk you into the room.
Here is what that looks like in practice. You are tired, so you skip the gym. Reasonable. With the evening now free you decide to catch up on work instead. Also reasonable. Working late means eating at your desk, so you are still up at ten. Everyone else has gone to bed. You are tired, you feel like you have earned something, and you are alone with the device you do this on.
At no point did you decide to relapse. You decided to skip the gym.
By the time you reach the last step you are choosing under conditions you set hours earlier, and you are choosing badly because those conditions were always going to produce that. The urge did not come from nowhere. It came out of a specific evening you built, one sensible decision at a time.
This is why try harder in the moment fails so reliably. It loads everything onto the worst point in the chain, the one where you have the least room and the least clarity. Anywhere earlier, the same decision would have cost you almost nothing.
So the useful question is not why you gave in at eleven. It is what the four hours before eleven looked like, and how often they look like that.
The load you are not counting#
Some conditions do not trigger anything by themselves. They just lower the bar for everything else.
Marlatt's term for this is lifestyle imbalance, and it covers the ordinary background of a life: how much you are carrying, how much of it you chose, and how long it has been going on. In practice it usually shows up as some mix of accumulated stress, sleep debt, and a stretch of being more alone than usual.
None of those make you relapse. What they do is make every other factor worse. The same free evening is a different proposition on four hours of sleep than on eight. The same trigger lands differently in week three of something difficult than it did in week one.
This is the part people miss when they go hunting for a trigger, because they are looking for an event and this is not an event. It is a level. It moves slowly, it is close to invisible from the inside, and it explains the otherwise baffling weeks where nothing in particular happened and everything was harder anyway.
It also explains the case that confuses people most, which is relapsing when things are objectively going well. A good stretch lowers vigilance. Nothing feels like it needs managing, so nothing gets managed, and then the first genuinely empty evening in a month arrives with no plan attached to it.
The reasons that are general, and the reasons that are yours#
Everything above is general. It is also, roughly, half of the answer.
The other half is not general at all. It is which hours, for you. Which days. What state you have to be in. What the four hours before look like on your worst nights, and how those differ from your best ones. Whether stress genuinely drives yours, or whether that is just the explanation you inherited from somewhere and never checked.
Nobody can write that half for you, because it is not the same for any two people. The general model tells you what kind of thing to look for. It cannot tell you where yours is.
Which brings us to the honest reason you have not been able to answer your own question. It is not that you have not thought about it hard enough. You have almost certainly thought about it more than about anything else in your life. It is that thinking is the wrong instrument. You are trying to detect a pattern spread across months, from memory, using the one tool guaranteed to be unreliable about it: your recollection of evenings you were not paying attention during, filtered through how you felt the next morning.
The pattern is there. It has just never been written down.
How to find yours in two weeks#
None of this requires buying anything. It requires two weeks and a small habit.
- Record every relapse and every urge you get through. Both. The ones you survive are data too, and they are the ones that show you what already works.
- Note four things each time: the time, your stress from one to ten, whether you were alone, and what you were doing in the hour before. Four fields. If it takes longer than a few seconds you will stop doing it by day four.
- After two weeks, lay them out and look for the bunching. Which hours. Which days. Which state. It is usually narrower and more specific than people expect, and often not what they assumed going in.
- Pick your two worst windows and decide in advance what you will do when you are in one. Not a general intention. A specific action attached to a specific moment. Deciding ahead of time rather than in the moment is the part of this with real research behind it (Gollwitzer and Sheeran, 2006).
- When you relapse, record it and change nothing else. Do not reset anything. Do not start over. It is one more line of data, and it is worth more than the line before it.
Two weeks of that will tell you more about your own pattern than two years of thinking about it.
Where Firmhold comes in#
We built Firmhold to do steps 1 through 3, because those are the ones that fall apart when you do them by hand. A log is easy to keep for nine days. Then you miss one, and then it is not a record any more.
Logging an entry takes under twenty seconds and needs no writing. Once there is enough data to say something honest, Firmhold shows you the conditions your relapses actually cluster around, and tells you when you are entering one of your own high-risk windows while you are still in it.
What it deliberately does not do: no streak counter, no blocking, no community, no accountability partner, no AI. Patterns are computed by plain deterministic rules on our own servers, and your data is never sent to a language model or any third party.
Logging, the calendar, your daily risk status and the panic button are free. The full pattern breakdown is the paid part.
We are not going to tell you how well it has worked for other people. We are new, we do not have that data, and inventing it is what most of this category does.
Start with one entry tonight
Logging takes under twenty seconds and needs no writing. No card, and the panic button stays free permanently.
Start free, no card neededCommon questions#
Why do I keep relapsing on porn even though I want to stop?
Wanting to stop and being able to stop in one specific moment are different things, and the gap between them is not a measure of how much you want it. Relapses cluster around particular hours, states and situations. If wanting it enough were the variable, they would fall randomly across the week, and they do not.
Is it normal to relapse this often?
There is no normal frequency, and any specific number you are given is invented. The more useful question is whether the frequency is trending down across months, rather than whether it sits above or below some figure.
Why do I relapse when everything is going well?
Relapse is not only a response to stress. A drop in vigilance, an unusually empty evening, or the quiet sense that the problem is handled are all common conditions, and they are easy to miss precisely because nothing feels wrong at the time.
How long before the pattern becomes clear?
Two to four weeks of consistent logging is usually enough to see the shape of it. Firmhold deliberately shows no pattern at all before fifteen entries, because before that it would be guessing.
Does knowing the pattern actually change anything by itself?
On its own, no. What it gives you is the ability to decide your response in advance for a specific window rather than in general, which is the part with research behind it.
What actually counts as a porn relapse?
Whatever you would call one if you were being honest with yourself and nobody else was watching. There is no official line, and hunting for a technical definition is usually a way of arguing yourself out of logging something. Firmhold treats the question as yours to answer, and logs whatever you call a relapse the same way it logs an urge you got through: as one more line of data, not a verdict.
Sources#
- Larimer, M. E., Palmer, R. S. & Marlatt, G. A. (1999). Relapse Prevention: An Overview of Marlatt's Cognitive-Behavioral Model. Alcohol Research & Health.
- Gollwitzer, P. M. & Sheeran, P. (2006). Implementation intentions and goal achievement: A meta-analysis of effects and processes. Advances in Experimental Social Psychology, 38, 69–119.
This is general information rather than medical advice. If this is causing you real distress or doing real damage to your life, a qualified therapist will be worth more to you than any app, including ours.