A social media break: what changes after 1 week, 2 weeks, 1 month

Published Last updated

Most accounts of a social media break are one person's story. There are two large randomised experiments that measured what changes, so this timeline is built from those rather than from anecdotes — with the caveat that an average across hundreds of people will never be a prediction for one.

What the experiments did

Two weeks, mobile internet blocked. 467 people had an app block mobile internet on their phone for two weeks. Calls and texts still worked. 91% improved on at least one of three measures: well-being, mental health and sustained attention. Daily screen time in the group that blocked first fell from 314 minutes to 161 minutes — about two and a half hours a day returned (PNAS Nexus, 2025).

Four weeks, Facebook deactivated. 2,844 Facebook users were measured, and a randomly assigned subset deactivated their accounts for four weeks before the 2018 US midterm election. Deactivation reduced online activity and increased offline activity, including watching TV alone and socialising with family and friends; it reduced both factual news knowledge and political polarization; it increased subjective well-being; and it caused a large, persistent reduction in Facebook use after the experiment (American Economic Review 110(3), 629–76, 2020).

Keep both results in view. The second one includes a cost — people knew less about the news — which is the kind of finding that gets left out of break-your-phone-habit articles.

Download Unrot on the App Store

A timeline to compare yourself with

This is not a promise of what happens to you. It is a frame for reading your own week.

Days 1 to 3: the reach

The habit is in your hand before your intention arrives. You will pick up the phone, unlock it, and find there is nothing to open. Expect to feel restless rather than free, and expect to feel it most at the times of day you usually scroll — after work, on the sofa, in bed.

Nothing in the research measures this part, but it is the stage people quit at.

Week 1: the time appears

The clearest effect is the one you can count. In the two-week experiment, the time returned was about two and a half hours a day. That is a lot of evening, and the common mistake is not planning for it. People who have one thing ready — a book, a run, a friend to call — do better than people waiting to feel inspired.

Week 2: attention and mood

This is where the measured changes showed up: well-being, mental health and sustained attention all improved for most participants in the two-week experiment. If you are going to notice something other than the free time, this is roughly when.

Week 3 to 4: the shape of the day

In the four-week deactivation, time moved to offline activities and well-being rose. Also, news knowledge fell — a real trade, worth replacing on purpose if you care about it, with a newsletter or a paper rather than a feed.

After it ends

The useful finding. In both experiments, use did not snap back to where it started: Facebook use stayed substantially lower long after the four weeks, and in a separate randomised experiment on social media use, temporary incentives to use less had persistent effects (American Economic Review 112(7), 2424–63, 2022).

A break is not only a break. It resets what normal feels like.

How to run one that survives the first week

Remove access, not accounts. Both experiments took away access while leaving everything else in place. Closing an account adds a cost you will resent by Thursday, and it makes going back an all-or-nothing choice.

Expect the first three days to be the loudest. The reach for the phone arrives before the intention does, and it is strongest at the times of day you usually scroll. If you can tell yourself in advance that days 1 to 3 are noise rather than evidence, you will still be going on day 4. Our list of apps to stop doomscrolling covers the tools that hold during that stretch.

Decide the replacement first. The gap arrives at the time of day you are least inventive. What to do instead of doomscrolling has six options to pick from in advance.

Keep the parts you use. Messages to people you know are not the thing you are taking a break from. Being contactable makes the break easier to defend to everyone else.

Set an end date, then decide again. Two weeks is enough to get a result. Deciding at the end, with evidence, beats an open-ended promise.

And if you were planning a weekend reset instead, does a dopamine detox work covers what two days does and does not achieve.

Where Unrot fits

The two experiments above both work by removing access for a while. Unrot does the same thing permanently and in small pieces: the apps you choose are locked, and they open for a limited time when you spend coins earned by completing a habit. It is less dramatic than a month off, and it does not end on a Sunday night.

Get Unrot on the App Store

Sources

  • "Blocking mobile internet on smartphones improves sustained attention, mental health, and subjective well-being", PNAS Nexus 4(2), 2025. Randomised experiment, 467 participants, two weeks.
  • Allcott, H., Braghieri, L., Eichmeyer, S., & Gentzkow, M., "The Welfare Effects of Social Media", American Economic Review 110(3), 629–76, 2020. 2,844 Facebook users measured; four-week deactivation, randomly assigned.
  • Allcott, H., Gentzkow, M., & Song, L., "Digital Addiction", American Economic Review 112(7), 2424–63, 2022.

Frequently asked questions

What happens when you quit social media for a month?

In a randomised experiment where 2,844 Facebook users were measured and a subset deactivated for four weeks, well-being went up, time moved to offline activities, factual news knowledge went down, and use stayed lower long after.

Is two weeks long enough to notice anything?

Yes. In a two-week experiment that blocked mobile internet, 91% of 467 participants improved on at least one of well-being, mental health and sustained attention.

Will I go back to normal afterwards?

The experiments suggest not entirely. Both found that use stayed meaningfully below the starting point after the break ended.

Do I have to close my accounts?

No. Both experiments took away access rather than asking anyone to erase an account, and a lock you cannot dismiss does the same job without losing your messages or photos.