Five settings, about fifteen minutes: autoplay of the next episode off, autoplay of previews off, a sleep timer on the television, a bedtime or downtime setting on the phone, and the remote living somewhere other than the arm of the sofa. None of these stops you watching. Each one puts a small pause where the machine currently runs without you.
Where the switches live
Menu names move between app versions and devices, so treat the middle column as a direction rather than a map. If the wording doesn't match your screen, search the settings for "autoplay" or "playback".
| Where | What to change | Roughly where it lives | What it leaves open | |---|---|---|---| | Netflix | Autoplay next episode, and autoplay previews while browsing | Account settings, playback settings, set separately for each profile | It's per profile, so a household of four needs four passes | | YouTube | Autoplay toggle, plus the break and bedtime reminders | The toggle sits on the player itself, reminders under general settings | Shorts, which runs on its own feed and its own rules | | Prime Video, Disney+, and the rest | Autoplay next episode | Playback or player settings, either in the app or on the account page on the web | Wording differs by service, and sometimes by which device you're on | | Your television | Sleep timer, and auto power on for the streaming stick | System, power or general settings on the TV, not in the streaming app | The phone in your hand at 1am | | iPhone | A sleep or downtime schedule, and the Clock app's stop playing timer | Focus and Screen Time, and the timer at the bottom of the Clock app | The television, which knows nothing about your phone | | Android | Bedtime mode, and the Clock app's sleep timer | Digital Wellbeing, and the Clock app | Same gap. Phone settings do not reach the TV | | The room | Where the remote and the charger spend the night | The hallway, a drawer, the kitchen | Nothing much. This is the strongest one on the list |
The sleep timer does more work than the autoplay switch
Autoplay off gives you a decision at the end of every episode. Useful, and also a decision you get to make five times in a row while getting steadily more tired.
A sleep timer makes the decision once, at the start, when you're still sensible. Set it for what you meant to watch plus twenty minutes. When the screen goes dark mid-scene it feels slightly rude, and that's the point. Rude is what gets people off the sofa. Most televisions have one buried in the settings, and both phone platforms have one in the clock app that stops playback rather than just muting it.
Decide before you sit down
Browsing is where the evening goes. You open the app with twenty minutes of intent, spend eleven of them on a wall of thumbnails designed to keep you scrolling, and start something because starting something is easier than closing the app.
So pick at lunchtime. One title, one number of episodes. Then when you sit down, you're collecting a thing you chose rather than shopping while tired.
The 2017 focus group work on binge-watching found participants describing their watching as context dependent, and every one of them "agreed having trouble recognizing themselves as truly being an 'addict.'" That's a hint about where the lever is. If the behaviour rides on context, change the context. You don't have to accept a label first, and nobody is asking you to.
Give the evening an ending
"When it finishes" is not an ending, because it never finishes. Something has to happen at a fixed time that isn't the credits rolling.
Pick something small and physical. The kitchen gets tidied at half ten. The dog goes out. The cue works because it's a different activity in a different room, so choose one you'd actually do.
Then plug the phone in outside the bedroom on your way up. That move is covered properly in [how to stop binge-watching](/guides/streaming/how-to-stop-binge-watching), and it's the one that protects the sleep.
The household part
A rule that annoys everyone else in the room lasts about four days.
Watching is usually shared, so negotiate rather than announce. Netflix settings are per profile, so yours can be the strict one without touching anyone else's. A TV sleep timer is a household decision and worth asking for out loud: "can we set the telly to go off at eleven on work nights." The remote living in a drawer is fine if everyone knows which drawer.
The conversation goes better if you say what you want rather than what's wrong with you. "I keep losing my mornings and I want to stop" is easier for someone to help with than a confession. If the answer is no, take the settings that are yours alone and leave the shared ones for another week.
Does any of this work?
Honestly: nobody has run a trial on streaming autoplay settings, and it's worth saying so rather than implying otherwise.
Part of the reason is that there's nothing formal to study. The DSM-5-TR is explicit that "gambling disorder is the only behavioral addiction (as opposed to chemical substance use disorders)" it identifies, and the proposed research condition for gaming "is limited to gaming and does not include problems with general use of the internet, online gambling, or use of social media or smartphones". Television isn't in the manual at all.
What does have evidence behind it is friction on a device in general. A randomised trial that blocked mobile internet on people's phones for two weeks reported that "the intervention improved mental health, subjective well-being, and objectively measured ability to sustain attention; 91% of participants improved on at least one of these outcomes". That's a far heavier intervention than an autoplay toggle, on a different device, for a different behaviour. The direction looks right. It does not prove that turning off previews will do anything for you, and this piece is not going to borrow its authority.
The sleep half is on firmer ground. A review of 67 studies found "screen time is adversely associated with sleep outcomes (primarily shortened duration and delayed timing) in 90% of studies", with the authors noting the limitation every study here carries: "causal association not confirmed". Associations, in children and adolescents, in studies that can't establish cause. Enough to justify moving a charger. Not enough to justify a lecture.
If you want the friction to be harder to undo than a settings toggle, that's what [Renovyn's Shield](/shield) does, blocking and delay across phone and laptop with someone else in the loop. It doesn't treat anything, and it won't decide what you watch.
Common questions
**How do I turn off autoplay on Netflix?** It's in your account settings under playback, set separately for every profile, so changing yours leaves everyone else's alone. Two switches: autoplay the next episode, and autoplay previews while you browse. Turn off both.
**Does turning off autoplay actually stop you binge-watching?** It stops the machine deciding for you, which is real, and it hasn't been tested in a trial. Most people find the sleep timer harder to argue with, because it lands mid-scene rather than at a natural stopping point.
**Is there a sleep timer for Netflix or YouTube?** Not reliably inside the apps. Use the timer built into your television, or your phone's clock app, which can stop playback rather than lowering the volume. Set it before you start.
**What if I keep turning the settings back on?** That's information rather than failure. It usually means the watching is doing a job for you in the evening, and [what you're avoiding when you press play](/guides/streaming/what-you-are-avoiding-when-you-press-play) is the piece to read next.
**I can manage one episode. Why can't I manage two?** The second starts while you're still inside the first one's ending, before your judgement catches up. [Stopping after one episode](/guides/streaming/stopping-after-one-episode) goes through that gap and what to put in it.
Sources
- [Blocking mobile internet on smartphones, 2025](https://pubmed.ncbi.nlm.nih.gov/39967678/). The randomised two week trial and the 91% figure, on device friction generally rather than streaming.
- [Screen time and sleep among school-aged children and adolescents, 2015](https://pubmed.ncbi.nlm.nih.gov/25193149/). The 67 study review, the 90% finding, and the causality caution.
- [Internet Gaming Disorder, American Psychiatric Association](https://www.psychiatry.org/patients-families/internet-gaming). Gambling disorder as the only behavioural addiction in DSM-5-TR, and the limits of the proposed gaming condition.
- [Toward a qualitative understanding of binge-watching behaviors, 2017](https://pubmed.ncbi.nlm.nih.gov/29025269/). Context dependency, and participants not recognising themselves in addiction language.
