Three things today, and none of them is a fresh set of rules. Write down the hour you reinstalled and what happened just before. Restore the one lock that was missing rather than adding four new ones. Then tell whoever you told last time, today, while it is still small.
What it actually proves
Nothing about your character. It is the single most common thing that happens in this whole category, and the only reason it feels rare is that people do not post about it.
The pattern is well documented across digital habits. Real posts read like this: "I've 'uninstalled' like 100 times by now and I keep relapsing", and "dealing with some extremely rapid cycles of quit/relapse". Almost every article stops at "delete the app" and leaves the next chapter unwritten, which is why the next chapter feels like a personal failing rather than the standard second step.
The UK's gambling guideline is the nearest clinical wording anyone has published on this shape of thing, and it is worth borrowing whatever the compulsion is: "relapse is not shameful and it may be part of a recovery journey or learning event", and "relapse does not indicate individual failure".
Do the forensics while it is fresh
You have about a day before the details blur into a general sense of having messed up, and the details are the only genuinely valuable thing here.
- **The hour.** Write the actual time. Most people find they have one window that accounts for nearly every reinstall.
- **The feeling just before.** One word. Bored, flat, rejected, wired, humiliated, wide awake. Not "I wanted to".
- **The last human contact before it.** Often the answer is a message that did not get answered, or a day with no conversation in it at all. That is the mechanism, more often than craving is.
- **The lock that was missing.** Almost always one specific thing. The site was not blocked in the browser, only the app was deleted. You knew the password. The passcode on the screen time settings was one you set yourself.
- **How long it took.** From opening it to closing it. This is the number to improve, and the next section explains why.
Shorten the round trip, do not chase the streak
The instinct after going back is to reset a counter and try to beat the old number. That instinct is what turns one evening into a fortnight, because once the streak is broken there is nothing left to protect until the new one is long enough to matter.
Track something else. How quickly you stop.
An evening is a smaller event than a week, and a week is smaller than a month. If your last cycle was three weeks back on and this one was two days, that is progress by the only measure that reflects what actually happened, even though a streak counter would show zero both times.
This also removes the reason people hide it. There is nothing to confess if there was no streak to break, only a duration to report.
Restore one lock, not five
The reflex after going back is to build a stricter regime. It fails reliably, for a simple reason: rules made in the aftermath of a bad night are made by someone furious with themselves, and they are always too harsh to keep. Then breaking those is a second failure on top of the first.
Fix the specific gap instead.
- **If the site was the route,** block it in the browser and in your phone's screen time settings, not only the app.
- **If you knew the password,** change it to something long and random you do not write down, and log out everywhere.
- **If you set your own passcode,** ask someone else to hold it. This is the one that changes outcomes most, and it is the one people avoid because it requires telling somebody.
- **If it was the bed,** the phone charges in another room from tonight. The CMOs' instruction is unambiguous: "Leave phones outside the bedroom when it is bedtime."
For what it is worth on the evidence, friction is real and modest. In a two-week trial that blocked mobile internet, "91% of participants improved on at least one" of mental health, wellbeing and sustained attention. For consumer app blockers specifically, no efficacy trial could be found. They buy you ninety seconds. Ninety seconds is often enough and it is not immunity.
Tell someone today
The gap between going back and telling someone is where this grows.
Tell the same person you told before, today, in one sentence, without a full account. "Went back on Tuesday, out again now." Nothing more is required.
If you never told anyone in the first place, that is the missing piece rather than the willpower. [Telling someone about your chatbot use](/guides/ai_chatbot/telling-someone-about-your-chatbot-use) has the words and the order to put them in.
When the cycles are getting faster
There is a version of this that is not a wobble, and it is worth being able to spot it.
If the round trips are getting shorter and more frequent, if you are back within hours rather than weeks, or if the evenings without it are genuinely distressing rather than dull, that is a signal to get a person involved rather than a better blocker.
Peer meetings exist for compulsive internet use and they are free. Samaritans is **116 123**, free, 24 hours a day, if it has got heavier than a habit. [Where to get support for compulsive chatbot use](/guides/ai_chatbot/where-to-get-support-for-compulsive-chatbot-use) is the honest inventory of what is out there, including what is not.
Common questions
**How many times can I do this before it means something?** There is no number and nobody has one to give you. What is worth watching is direction. Shorter cycles and faster recoveries are one story, longer and heavier ones are another.
**Should I delete the account this time instead of the app?** If you have reinstalled more than once, yes, because deleting the app leaves the history intact and the history is what you go back for. Expect it to hurt more, and [deleting a companion app, and the grief that follows](/guides/ai_chatbot/deleting-a-companion-app-and-the-grief-that-follows) is written for that.
**I went back and it was fine. Does that mean I was overreacting?** Possibly, and one calm evening is not much evidence either way. Check it against what your use displaced over a month rather than against how one night felt.
**Should I start the whole plan again from day one?** No. Keep the parts that were working and change the one that failed. Starting from zero throws away everything you learned, which is the actual asset.
**Why does it feel so much worse than it is?** Because you framed it as a test of character and then scored yourself. It was an evening. Shame is the thing most likely to send you back tomorrow, which makes going easy on yourself the practical choice rather than the soft one.
Sources
- [NICE NG248, gambling-related harms](https://www.nice.org.uk/guidance/ng248/chapter/recommendations). That relapse is not shameful and does not indicate individual failure.
- [Blocking mobile internet access for two weeks, PNAS Nexus](https://pubmed.ncbi.nlm.nih.gov/39967678/). 91% improved on at least one outcome, from friction rather than abstinence.
- [UK Chief Medical Officers' commentary on screen time and social media](https://www.gov.uk/government/publications/uk-cmo-commentary-on-screen-time-and-social-media-map-of-reviews). Phones out of the bedroom at bedtime.
- [Longitudinal randomised study of extended chatbot use (arXiv, 2025)](https://arxiv.org/abs/2503.17473). That heavier voluntary use went with worse outcomes, as an association rather than a cause.
