How to Reduce Screen Time: 6 Ways, Ranked by the Evidence

Advice on how to reduce screen time is abundant and almost entirely unranked. Every list gives equal billing to turning off notifications, leaving the phone in another room, buying a $60 timer box and "being more mindful", as though the evidence behind them were comparable. It is not.

This article ranks six ways to reduce screen time and use your phone less by how much controlled evidence sits behind each, says how big the effect was, and says where the evidence runs out. One widely repeated tip is demoted at the end, because the meta-analysis of it came back close to zero.

Two things are true of almost everything below. The effects are real and measured in a fortnight, and the mechanism is always the same: making the automatic thing require a decision. Nothing here is about wanting it more.

1. Cut the phone off the internet entirely

Evidence: a randomised controlled trial, the largest effects in this article, and the worst adherence.

Researchers randomised 467 adults to block all mobile internet on their own phones for two weeks, calls and texts still working, with a delayed-intervention group blocking in weeks three and four instead. A blocker app verified objectively whether the block stayed on.

In the group that blocked first, average daily screen time fell from 314 minutes to 161, measured from the iPhone Screen Time screenshots participants uploaded. Mental health improved with an effect size of 0.56, which the authors point out is larger than the meta-analytic effect of antidepressants and comparable to cognitive behavioural therapy. Subjective well-being improved at 0.45. Objectively measured sustained attention improved at 0.23, which the authors liken to undoing about ten years of age-related decline. Over 90 per cent of participants improved on at least one of the three.

Screen time went back up afterwards, to 265 minutes, still below the 314 they started at.

Now the number that changes how you should read all of it. Of 467 people who volunteered, 272 installed the blocker and 119 kept it running for the required ten days out of fourteen. That is a quarter of a sample that had explicitly signed up to reduce their phone use. The authors' own conclusion is that "less restrictive approaches may be beneficial."

So: the biggest effect in the literature, from the intervention three-quarters of motivated volunteers could not complete. If you want to know what your ceiling looks like, this is it. If you want to know what you will still be doing in March, read on.

2. Change ten settings and change nothing else

Evidence: a randomised trial against a control group, effects holding six weeks, and everything in it is free.

A McGill team tested a bundle of ten nudges — disabling non-essential notifications, switching the display to greyscale, and eight others of that size. Participants picked which to follow and ran them for two to six weeks.

The first study (51 people, before and after) found less problematic smartphone use, less screen time and fewer depressive symptoms after two weeks. The second (70 people) compared the bundle against a control group who did nothing but monitor their own screen time, and found the intervention group lower on problematic use and screen time, and better on sleep quality. Scores stayed at normal levels for at least six weeks.

Two things to take from it. The first is the best effort-to-effect ratio in this article: an evening of settings changes, holding for a month and a half. The second is what the control condition was. Watching your own screen time number is the thing the intervention had to beat, and it beat it. Tracking is a diagnostic, not a treatment.

The caveats are real: small samples, self-selected, and because people chose their own strategies you cannot tell which of the ten did the work.

3. Take one specific hour out

Evidence: a randomised trial in workplaces, with a two-week follow-up.

A German study assigned 278 employees to one of four groups for a single week: reduce daily smartphone use by one hour, add thirty minutes of physical activity, do both, or change nothing. Measures were taken at baseline, at the end, and two weeks later.

The two groups that cut phone use reported better mental health, better work-life balance, higher work satisfaction and motivation, less sense of work overload, and fewer symptoms of problematic smartphone use. Notably, every group asked to change something reported fewer depressive symptoms and more sense of control, including the exercise-only group, which is a useful check on attributing all of it to the phone.

The instruction here is unusually concrete, and that is probably the point. "Use your phone less" is not actionable. "One hour less per day, starting Monday, for a week" is, and a week is short enough to actually finish. All outcomes were self-reported, which is this study's main weakness.

4. Batch your notifications — do not kill them

Evidence: a randomised field experiment, and the result most summaries get backwards.

A field experiment with 237 people compared three conditions: normal notifications, notifications batched into three deliveries a day, and no notifications at all.

Batching won. Those people felt more attentive, more productive, in better moods and more in control of their phones, with lower stress and fewer interruptions.

The group that received no notifications at all got few of those benefits and reported higher anxiety and more fear of missing out. Cutting yourself off from information about your own life turns out to have a cost that cancels much of the gain.

This is the single most useful correction in this article, because "turn off all notifications" is the most repeated instruction in the genre and it is not what the experiment supports. Three scheduled deliveries a day is the finding. Most phones can approximate it: iOS has Scheduled Summary, Android has notification snoozing and per-app controls, and the notification setup is a job of its own.

5. Put friction in front of the app you keep opening

Evidence: a very large observational dataset, honest about its own decay.

Analysis of 1,318,331 interventions from 1,039 users of the friction app one sec, over an average of 13.4 weeks, found people backed out of 33 per cent of app openings when a short delay was inserted first. Over 80 per cent of the frictions people set up targeted social media, mostly Instagram.

Then the finding that matters for anyone planning to rely on this. Both curves fell over time: app-open attempts declined, and so did the proportion of attempts the friction managed to stop. The authors read the second as openings becoming more intentional, and their interviewees back that up. The same data is equally consistent with plain habituation to an obstacle you see fifty times a day.

Also: 26.9 per cent of users took at least one break from the intervention, 3.46 breaks on average among those who took any, with about two days the most common length.

This is not a controlled trial, it comes from a commercial app's own user base, and one sec's founder is a co-author. Take it as good evidence about the shape of friction over months, not as a verdict on a product. The shape is: works immediately, works less later, gets switched off periodically, and that is normal rather than a personal failure.

6. Leave the phone in another room

Evidence: a famous study, a failed replication, and a meta-analysis that lands near zero.

In 2017, Ward and colleagues reported that merely having your own smartphone on the desk, switched off and face down, reduced available working memory compared with leaving it in another room. It became one of the most-cited findings in this field and the basis for the advice you have read a hundred times.

It has not held up cleanly. A 2020 replication using the same tasks and conditions found no difference between phone-location conditions on either measure.

The meta-analysis is the one to know. Pooling 56 studies and 7,093 participants, working memory showed a negative effect of smartphone presence, but every other cognitive domain came back null, and the overall combined effect was not statistically distinguishable from zero — indeed, against conservative equivalence bounds, likely equivalent to zero. Where the working-memory effect did appear, it was far smaller than the original paper suggested. The authors flag substantial methodological heterogeneity and generally poor statistical power across the whole literature.

This does not mean the advice is bad. Putting the phone in another room removes the thing you pick up, and every other item in this article is about exactly that kind of access change. It means the specific claim — that a silent phone in your eyeline is draining your cognitive capacity — is not supported at the strength it is usually stated. Do it because you will not reach for what is not there, not because of the brain drain.

The tips every list repeats

Most lists on how to reduce screen time lead with the same few suggestions. Here is where each stands.

• Greyscale. It was one of the ten nudges in the McGill bundle, so it shares that trial's result without being isolated by it. Tested alone, greyscale cut daily screen time by 37.9 minutes in a study of 161 undergraduates, and a 2024 study found shorter sessions but no fewer unlocks. Free, reversible and modest. The greyscale guide has the setup for iOS and Android.
• No phone in the bedroom. This is the other-room advice applied at night, and it rests on access rather than on the cognitive claim. Logged phone data from 815 young adults found 41 per cent never went six consecutive hours without touching their phone during their own reported sleep, an association rather than proof of cause. The phone-free bedroom guide deals with the alarm and the emergency call.
• Screens off before bed because of blue light. The blue light is the weak part of this one. A randomised trial of 167 young adults found Night Shift made no meaningful difference to sleep compared with an ordinary screen. The case for a cut-off is the scrolling, not the colour of the light, which the sleep guide goes into.
• Replace it with a hobby. None of the trials above tested substitution directly, so it gets no rank here. The closest is the German workplace study, where the group asked only to add thirty minutes of exercise also reported fewer depressive symptoms and more sense of control.

What this adds up to

Ranked by evidence, not by how good it sounds:

1. Block the internet for two weeks if you want to know what the ceiling is. Largest measured effects of anything here, and three-quarters of volunteers could not complete it, so treat it as a diagnostic rather than a plan.
2. Change ten settings in one evening. Best return for the effort, held for six weeks against a control group, costs nothing.
3. Cut one specific hour, for one specific week. Concrete enough to finish, and it moved work satisfaction and mental health in a randomised trial.
4. Batch notifications into three deliveries a day. Do not switch them all off: that arm of the experiment came out worse on anxiety.
5. Add friction to your one worst app, and expect it to stop working as well after a few weeks. Plan the reset instead of being surprised by it.
6. Leave the phone in another room because of the access, not the folklore. The cognitive claim behind it is close to zero once the whole literature is pooled.

And the thing that is not on the list: monitoring your screen time. In the one trial here that used it as a control condition, it was the thing the real intervention had to outperform, and a field study of 242 people found that tracking raised awareness of usage without reducing it. Knowing the number is where this starts, not what fixes it.

The honest summary of this whole literature is that restriction works while it is running, and the difficulty is never the fortnight. It is the fifth week, which is roughly when every study above stops measuring.