Expect the urges to cluster in specific hours rather than spread evenly, expect day three to five to be harder than day one, and expect at least one evening where you're convinced this was a stupid idea. What you should not expect is a fixed timeline, because there isn't one that any research supports, however precisely the internet states it.
Start here: nobody has published the timeline
Search this and you'll find day-by-day charts. Day 3 is irritability, day 7 is the flatline, day 14 is clarity, day 90 is a different man. They're written with total confidence and they're not measurements of anything. There is no retrievable study establishing a stage-by-stage timeline of abstinence from pornography.
That matters practically, not just pedantically. If you're given a chart and your fortnight doesn't match it, you'll conclude you're doing it wrong, and "I'm doing it wrong" is one of the most reliable ways to end a first attempt in week two.
So what follows is a shape, not a schedule. Yours will be its own shape and that will be fine.
What is actually established
Short list, and short is the honest length.
The behaviour pattern being interrupted is real and recognised. Compulsive sexual behaviour disorder is in ICD-11, and part of its definition is [repeated unsuccessful efforts to significantly reduce the behaviour](https://www.findacode.com/icd-11/code-1630268048.html). If you've tried before and it didn't hold, that's part of the described picture rather than a personal failing.
Treatment helps more than nothing. A 2025 meta-analysis of 20 studies and 2,021 participants found psychotherapy, mostly cognitive behavioural and acceptance and commitment therapy, produced improvements ["on PPU, frequency/duration of pornography use, and sexual compulsivity, with large effect sizes"](https://pubmed.ncbi.nlm.nih.gov/40126561/), while flagging a high risk of bias in the underlying studies. Something works. Nobody has published what the second Tuesday feels like.
And distress that comes entirely from disapproving of yourself isn't the disorder. The ICD-11 entry says so directly. Keep that near you in week one, because two weeks of monitoring your own thoughts can turn a person into their own prosecutor.
The shape of the fortnight
**The first two or three days** are often easier than people expect. There's momentum in a decision, and the environment changes you made are still novel. Some people have a genuinely good first weekend and read it as proof the problem was smaller than they thought. It isn't. It's day two.
**Days three to seven** is where most first attempts come apart. The novelty has gone, the ordinary triggers have come back around, and the reasons that felt so clear on Sunday are much less vivid at eleven on a Thursday. Sleep is often disturbed in this stretch and irritability is common, though both are also just what happens when you take away the thing you've been using to get to sleep and to regulate a mood.
**The second week** tends to be quieter with sharper spikes. Fewer bad hours, but the ones that arrive are more convincing, and they often follow something specific: a row, an empty evening, a piece of bad news, an unusually good day. Recovery from a good day catches people out constantly.
Across both weeks, notice the clustering. Almost everyone finds it concentrates in one or two windows, usually late and alone. Two weeks of noting the hour is the cheapest useful data you'll ever collect on yourself, and [when do you reach for porn](/guides/pornography/when-do-you-reach-for-porn) turns it into something you can act on.
The claims that circulate, sorted honestly
Not all of these are wrong. Some are simply unverified, and there's a difference worth keeping.
- **"Withdrawal symptoms".** Plenty of people report irritability, low mood, trouble sleeping and restlessness in the first days. What isn't established is that this is a withdrawal syndrome in the sense that word carries elsewhere. Note that the ICD-11 authors put compulsive sexual behaviour under impulse control disorders rather than with the addictions, precisely because [the field does not yet have definitive information](https://pmc.ncbi.nlm.nih.gov/articles/PMC5775124/) about whether the processes match. Treat the discomfort as real and the mechanism as unsettled.
- **"The flatline".** A period of flat mood and low sexual interest, widely described in online communities. No retrievable research establishes it as a documented phenomenon with a timing or a duration. If it happens to you, it's not a sign something has broken.
- **"Ninety days rewires your brain".** Ninety is a fellowship convention and a forum convention. There's no published evidence tying that number to any biological change. Counters can be motivating and they're also the main engine of the all-or-nothing crash described in [you watched again, what now](/guides/pornography/you-watched-again-what-now).
- **"Superpowers", confidence, magnetism.** Self-reported, in communities selected for wanting it to be true. Some people do feel noticeably better in a fortnight. Attributing that to anything other than better sleep and reclaimed evenings is a story, not a finding.
- **"Escalation is inevitable".** Not established. A 2024 network analysis across two samples of men found that quantitative tolerance, meaning simply needing more, [sat centrally in the network](https://pubmed.ncbi.nlm.nih.gov/38761685/) connecting other patterns to problematic use. That's a real finding about how the patterns cluster, but it's cross-sectional, so it cannot tell you that one thing leads to the next over time. [Escalation, and why it stops working](/guides/pornography/escalation-and-why-it-stops-working) takes this apart properly.
- **"Erectile problems will resolve at day X".** No. If you have erection problems, the NHS position is that occasional difficulty is [usually caused by stress, tiredness or drinking too much alcohol](https://www.nhs.uk/conditions/erection-problems-erectile-dysfunction/), and that recurring problems should be seen by a GP or a sexual health clinic because they can indicate a treatable condition. Waiting out a countdown instead of getting checked is a bad trade.
What to measure instead of days
Counting days makes the whole thing brittle. These are better, and they survive a bad Thursday.
- **Hours back.** Roughly how much of your week has returned, and where it went instead.
- **Sleep.** What time you actually went to sleep, not what time you went to bed.
- **The gap.** When something does happen, how long between the episode and telling someone. Shrinking that gap is real progress even in a week you'd otherwise call a failure.
- **Which route leaked.** Every episode used a specific device on a specific network at a specific hour. Close that route rather than adding a general resolution. [Blocking porn on every device](/guides/pornography/blocking-porn-on-every-device) has the layers.
- **Whether you told the truth to one person.** Weekly. That single question predicts more than a counter does.
Practical things that help in this fortnight
Give the two bad hours a job before they arrive, and make it something with another person in it where you can. Eat properly, because a surprising share of these evenings are hunger and tiredness in costume. Keep the passcodes with someone else. Don't take on anything else hard this fortnight if you can avoid it.
And decide now, while you're clear headed, what you'll do if it goes wrong, because deciding it afterwards never goes well. [Getting through the urge without watching](/guides/pornography/getting-through-the-urge-without-watching) is the ten-minute version to keep on your phone.
If any of this tips into something darker than frustration, Samaritans is on 116 123, free, any hour. In England, 111 and option 2 reaches urgent mental health support.
Common questions
**How long do porn withdrawal symptoms last?** There's no established answer and any specific number you're given is invented. People commonly describe the hardest stretch falling somewhere in the first week, with the difficulty concentrating in particular hours rather than being constant. If low mood or sleep problems persist beyond a couple of weeks, that's worth a GP appointment on its own terms.
**Is the flatline real?** It's widely described in online communities and it isn't documented in research that we could retrieve. That doesn't mean people are lying about how they feel. It means nobody has measured it, so treat any confident description of when it starts and ends as folklore.
**Will two weeks fix my sex life or my erections?** Two weeks is not a treatment for a medical problem. If erections are the concern, see a GP or a sexual health clinic rather than waiting out a counter, because recurring difficulty can point to something treatable. [Sex and intimacy after stopping](/guides/pornography/sex-and-intimacy-after-stopping-porn) covers what to expect and when to get checked.
**I got to day nine and it happened. Do I start again from zero?** Only if you decide to run it that way, and running it that way is what makes week three so dangerous. Nine days of changed behaviour happened. They don't unhappen. Note which route leaked, close it, and carry on from today.
Sources
- [ICD-11 compulsive sexual behaviour disorder entry](https://www.findacode.com/icd-11/code-1630268048.html). Repeated unsuccessful attempts to reduce the behaviour, and the exclusion for distress arising only from moral disapproval.
- [Compulsive sexual behaviour disorder in the ICD-11, World Psychiatry](https://pmc.ncbi.nlm.nih.gov/articles/PMC5775124/). Why it sits under impulse control disorders rather than with substance use and gambling.
- [Ince et al., Addictive Behaviors, 2024](https://pubmed.ncbi.nlm.nih.gov/38761685/). A cross-sectional network analysis of escalating use patterns in two samples of men.
- [NHS, erection problems](https://www.nhs.uk/conditions/erection-problems-erectile-dysfunction/). Common causes of occasional difficulty and when to see a GP.
