Do app blockers work? What the research actually shows

Every app blocker's website says it works, ours included. There is a small amount of solid research on whether tools like these change behaviour, and it is more encouraging and more modest than the marketing. Here are the three studies we find most useful, with their weaknesses left in.

Study 1: limits you set yourself and cannot instantly override

Who: Economists Hunt Allcott, Matthew Gentzkow and Lena Song. Published in the American Economic Review in 2022.

What they did: A randomised experiment with 1,933 US Android users over twelve weeks. One group was given an app that let them set their own daily time limits on apps. Unlike the limits built into phones, these could not be dismissed on the spot.

What happened: Use of six target apps (Facebook, Instagram, Twitter, Snapchat, YouTube and the browser) fell by 22 minutes a day, about 16%. The effect was still there in the final weeks. 78% of people set limits that actually constrained them and kept using them. About half the time saved moved to other apps.

What it tells us: People want to restrict themselves and will keep doing it when given a tool that holds. The authors estimate that self-control problems account for about 31% of social media use. That figure comes from their economic model and is not a direct measurement.

Caveats: Android only, recruited through ads, run during 2020.

Source: Allcott, Gentzkow and Song, "Digital Addiction"

Study 2: a pause before the app opens

Who: David Grüning, Frederik Riedel and Philipp Lorenz-Spreen. Published in PNAS in 2023.

What they did: Analysed six weeks of data from 280 people using One Sec, an app that makes you take a breath before a chosen app opens and then asks if you still want to continue.

What happened: In 36% of attempts, people closed the app again after the pause. Over the six weeks they also tried to open those apps less often. Put together, actual openings of the target apps were 57% lower by week six. A follow-up experiment with 500 people found that being offered the choice to back out did most of the work. A message asking people to reflect did nothing by itself.

What it tells us: A lot of phone use is reflex, and a few seconds of friction interrupts a good share of it.

Caveats: These matter. One of the authors, Frederik Riedel, is the developer of One Sec. The 280 people had chosen to install the app, and there was no control group. The 57% is a fall in app openings, not in total screen time.

Source: Grüning, Riedel and Lorenz-Spreen, 2023

Study 3: blocking mobile internet entirely

Who: Noah Castelo and colleagues. Published in PNAS Nexus in 2025.

What they did: A preregistered randomised trial with 467 iPhone users in the US and Canada. For two weeks, an app blocked all internet access on their phones. Calls and texts still worked, and they could go online on a computer.

What happened: Daily screen time in the blocked group fell from 314 minutes to 161. On average, people's reported well-being and mental health improved, and so did their performance on an objective test of sustained attention.

What it tells us: Cutting the phone back to a phone has measurable benefits, at least over a fortnight.

Caveats: Only 119 of the 467 participants, about a quarter, kept the block on for at least 10 of the 14 days. The headline results include everyone assigned to the block, whether or not they stuck to it. A total internet block is also much harsher than locking a few apps, which may be why so many gave up.

Source: Castelo et al., 2025

What the three have in common

Blocking changes behaviour, by a moderate amount. Nobody's phone use fell to zero. The realistic prize is tens of minutes a day and fewer reflexive opens.

The exit matters most. In the first study the limits worked because they could not be waved away. The paper explicitly contrasts them with the "flexible limits" in tools like iOS Screen Time. A limit with a "one more minute" button is a suggestion.

Too strict fails as well. Three quarters of people in the third study did not keep a total block going. A tool you abandon in week one does nothing in week two.

Time leaks. Some of what you save goes to other apps and other screens.

A fourth, smaller study is worth a mention because it tests strictness directly. Researchers compared a standard lockout with a gentler approach that made taps and swipes slightly sluggish once a daily allowance ran out. Among 42 Android users over five weeks, the gentler approach cut use further than the lockout, and people were more willing to accept it. (Lu et al., 2024) It is a small study, and it is a useful reminder that people stick with tools they can tolerate.

How to use this when choosing

The research does not crown a product. It does suggest what to look for.

1. An exit that costs something. Not a button on the same screen. A passcode held by someone else, a strict mode, or a physical key somewhere else.
2. Strictness you will keep. Lock the few apps that cause the trouble, for the hours that matter, and leave the rest of the phone alone.
3. A safety valve. Some limited way back in for real emergencies, so that one bad day does not make you delete the whole thing.
4. A plan for the freed time. A book by the sofa does more than an empty hour.

Where physical blockers fit

We could not find a published trial of NFC pucks or cards, so we will not claim one. The case for them is an argument from the findings above, and you should weigh it as an argument: the exit costs a walk (point 1), you choose which apps and when (point 2), and most have emergency unlocks (point 3).

moss is our attempt at that design: tap the stone to lock chosen apps, tap again to unlock, three emergency unlocks a month, and the option for a household to lock at the same time. Other products make different trade-offs, and we compare them in detail here. If you want the cheapest test of whether friction helps you, start with a free pause app or a strict Screen Time setup and see what a fortnight does.