Where The Time Actually Goes: A Three-Day Count
Nobody has a moral opinion about your hours. There is only the number, and what you would rather have done with it.
Of everything on this site, this is the cheapest thing to find out and the one that changes most people's minds without any argument at all. Not "is it a problem". Not "what does it say about me". Just: how many hours, in an ordinary week, and what would you otherwise have done with them?
The reason this works is that time is the one dimension of this subject with no moral content whatsoever. Nobody has an opinion about your hours. There is no debate to be had about whether four hours is sinful. There is only the fact of four hours, and whatever you would rather have done with them.
Measure it properly, which means measuring it badly at first
Most people's estimate is wrong in a specific direction: it counts the deliberate sessions and misses everything around them. A realistic count includes the fifteen minutes of scrolling beforehand, the time spent looking for something rather than watching it, the recovery period afterwards where you are not really doing anything, and the second and third episodes on the days there was more than one.
A three-day count, done honestly
- For three days, note the time it started and the time you were properly doing something else again. Not a spreadsheet. A note on your phone or a scrap of paper.
- Write down what you were doing in the ten minutes before it started. This is the single most useful column and almost nobody fills it in.
- At the end, add it up and multiply by a week. Do not adjust the number down because the days felt unrepresentative. They rarely are.
What people usually find
Two patterns come up repeatedly. The first is that the total is roughly double the estimate, almost entirely because of the surrounding time. The second is that it is heavily concentrated: not spread evenly across the week, but clustered into two or three specific situations. Late at night, alone, tired. Immediately after a stressful thing. A particular empty slot in a particular day.
The concentration is good news, though it rarely feels like it. A behaviour spread thinly across every hour would be genuinely hard to change. A behaviour that clusters into two situations gives you two places to intervene, and you can design for two situations.
Now convert the hours into something real
Six hours a week is a training programme. It is a language at a slow pace. It is most of a novel. It is two evenings you could have spent with someone. The point of the conversion is not guilt — guilt is counterproductive here and always has been. The point is that "six hours" is abstract and "the reason I never get round to X" is concrete, and concrete is what actually motivates a change.
It is also worth doing the conversion honestly in the other direction. If the count comes back at forty minutes a week and it is not displacing anything you care about, then the time dimension is not your problem, and you can stop worrying about that one and look at whether something else — control, escalation, how you feel afterwards, a partner's reaction — is what is actually bothering you.
The measurement is also the intervention
There is a well-documented effect where simply recording a behaviour reduces it, without any other change. Do not be surprised if the three days you are counting are quieter than usual. That is not you cheating the exercise. That is attention doing what attention does: a behaviour running on autopilot becomes harder to run when you are watching it.
Which means you have already found the first lever. Anything that converts autopilot back into a decision — a note, a count, ten seconds of friction — does real work, and it does it without requiring you to be a more disciplined person than you were yesterday.
What to do with the number
Do not set a target. Targets on this subject tend to become streaks, and streaks become something to break. Instead, take the two situations where the time clustered, and change something about those two situations specifically. That is a much smaller ask than "use less porn", and it is the ask that actually gets done.
Why estimates are wrong in a predictable direction
Self-reported time is unreliable for every behaviour that runs on autopilot, and it is unreliable in one consistent direction: people count the part they decided to do and forget the part that happened around it. There are three specific blind spots.
- The approach. The scroll before it starts. This is frequently longer than the episode and is almost never counted, because it does not feel like part of it.
- The search. Time spent looking rather than watching. On any recommendation-driven platform this can be most of the session.
- The tail. The flat twenty or thirty minutes afterwards where you are not really doing anything. It is real time, it is caused by the episode, and it belongs in the total.
Add those three and a "twenty minutes, a few times a week" estimate frequently becomes four or five hours. Nobody is lying to themselves on purpose. The parts simply do not feel like they belong to the thing being counted.
How to count without turning it into a project
The best method is the one you will actually use for three days. A note on your phone works. A piece of paper by the bed works better, because it does not require picking up the device that is part of the pattern.
Record four things: when it started, when you were properly doing something else again, where you were, and what you were doing ten minutes before. That fourth column is the one people skip and the one that turns out to matter most — it is where your trigger is hiding.
Do not adjust as you go. Do not skip a day because it was unusual; unusual days are part of the average. And do not extend the exercise past a few days, because a measurement you get bored of is worse than a short one you complete.
Converting hours into something you can feel
An abstract number does not motivate anybody. A concrete trade does. Six hours a week is roughly:
- Two full evenings with someone you have not seen in a while.
- A structured training block, three times a week, with time left over.
- Forty minutes a day toward something that compounds — an instrument, a language, a side project.
- An extra hour of sleep on the nights it happens latest, which is usually the more valuable trade.
Notice which of those produced a reaction when you read it. That reaction is the actual motivation, and it is a great deal more durable than resolve.
When the answer is "this is fine"
Sometimes the count comes back small and displaces nothing. That is a real and common result, and it deserves to be taken as seriously as a large one. If the time is not the issue, stop working on the time. Look instead at whether what is bothering you is control, direction of travel, secrecy, or how you feel afterwards — those are four different problems with four different responses, and working on the wrong one is how people spend months making no progress.
What to do with the number
Do not set a target. A target becomes a streak, and a streak becomes something to break. Instead take the two situations where the hours clustered and change one thing about each. That is a far smaller ask than "use less", and it is the ask that gets done.
Re-count in a month. Not to grade yourself — to find out whether the change you made moved anything, or whether the behaviour simply relocated to a different evening. Relocation is common, it is not failure, and you cannot detect it without counting twice.
This is general education for adults, not medical or psychological advice, and it cannot take account of your situation. If something here is persistent or distressing, a qualified professional can do what a website cannot.