Screens & Sleep
A One-Week Experiment: No Screens After 9pm
By Nora Vale · November 23, 2025 · 7 min read
Most screen advice fails because it’s a forever rule with no test date. This one is a week, which changes everything: a week is short enough to keep honestly and long enough to learn something true. The experiment is simple. No screens after 9pm, seven evenings, everything else as normal as you can keep it. Log what happens, because memory will tell you that you barely scrolled and barely woke, in the same breath, and be wrong about both. The reasoning behind the cutoff is in my screens and sleep guide.
Key takeaways
- Seven evenings, screens dark after 9pm, everything else held steady.
- Log each night: screen status, bedtime, wake-ups, morning verdict.
- Plan the empty hour in advance, or the phone gets rehired by Thursday.
- Swaps are fine, gaps are not. Move a night instead of canceling it.
- A null result is a good result: it clears a suspect with real data.
Before day one: set the table
An experiment needs a fixed variable, so decide the details before 9pm on day one, not during it. Decisions made at 11 p.m., phone already warm in your hand, resolve about as reliably as decisions to have one more drink. Decide at nine where the phone sleeps, and the eleven o’clock version of you never gets a vote. The practical version: buy or borrow a real alarm clock, pick the phone’s new home, the kitchen counter, a hallway shelf, anywhere the reflex can’t reach, and tell the house. A rule nobody explained reads as a mood, and moods get negotiated with at midnight.
Then fill the hour you’re about to free, because a hole with nothing planned for it is how the phone gets rehired. One low-effort thing for your body, one for your mind, one for the room. A shower, a stretch, tomorrow’s list on paper, a few pages of a paperback with a cracked spine, same order each night. It sounds like nothing. It is nothing, and that’s the point. The full menu is in what to do the hour before sleep.
The week, night by night
Nights one and two are the itchy ones. Expect to reach for the phone around 10:15 out of pure reflex and find a lamp. Sit with that. The first evening usually feels longer than it should, the same hour in the same chair with nothing arriving, and that faintly bored feeling isn’t a problem. It’s the wind-down slot announcing itself, the one the feed had been quietly occupying for years. Bored is what an evening looks like when it has stopped being medicated.
Nights three and four are where the bed gets boring, in the good way. A bedroom with no feed in it turns out to be a room where a tired person can finish the thought called sleeping, because nothing in it is asking questions. If your evenings ran on streaming instead of scrolling, this is also the moment to notice auto-play: it’s a setting, not weather, and turning it off is two free minutes that protect every night from here on.
Nights five through seven are the data. By the second half of the week most people can already see the trade: the feed returned time they didn’t know they were spending, and the question is whether sleep got any of it. Somewhere around day five or six you may wake at some dark hour, reach out of habit, and close your hand on nothing. That nothing is the whole trick, because the worst possible company at 3 a.m. is a feed.
The log, and how to read it
Keep it stupid-simple: date, screens after 9pm yes or no, lights-out time, wake-ups, and a one-line morning verdict. The pairing of screen column and wake column is the entire point. After seven days, look for one of three patterns. If evenings went faster and mornings improved, you’ve found a thief with real evidence. If evenings improved but the middle of the night didn’t, the screens were stealing time and something else owns the wakings, which points you toward the 3 a.m. mechanics in why do I wake up at 3am every night. If nothing moved, that’s a clean null result, and an experiment that clears a suspect is still a good experiment.
Two rules keep the week honest. Everything else stays as steady as you can manage: if you simultaneously move dinner, start a bath routine, and dock the phone, you’ll have no idea which lever did the work. And the calendar rule: if a birthday lands on an experimental night, move the night instead of canceling it, because the experiment is a pattern, and a pattern tolerates swaps but not gaps.
If a full week sounds like more than you can promise, the stair-step is legitimate: make the final hour screen-free first, then stretch to 90 minutes, then claim the whole evening. A rule you keep beats a rule you admire. And a note on the content itself, since the cutoff is about feeds more than photons: a paper book passes, an e-reader with the front light low passes, audio with a sleep timer passes. A phone displaying a book fails, because the entire arcade is one swipe away, and at 10:40 you know exactly which one wins. If your evening ran on shows rather than feeds, the same experiment works with auto-play off, which is two free minutes tonight and the second-best setting change in the whole chapter.
One more expectation worth setting, because nobody warns you: the freed hour is a vacancy, and vacancies make you nostalgic. Around day four or five you’ll catch yourself reasoning that one small check wouldn’t hurt, and the reasoning will feel fresh every time. It isn’t. It’s the same negotiation the feed ran the week before, wearing a different outfit, and the answer is the same as it was on night one: the phone sleeps where you decided it sleeps, and the eleven o’clock version of you never got a vote.
The book behind this chapter, with the full evening protocol, is Sleep, Finally. Pick your start date, dock the phone, and let the week report honestly.
Frequently asked questions
Is one week enough to see a difference?
Usually yes for the evenings themselves, because removing the feed returns time immediately and the first quiet nights are noticeable. Sleep habits compound over weeks, but a week is enough to learn whether screens were stealing hours, quality, or neither, and any of those answers is useful.
Does 9pm count as early enough?
For most bedtimes, a 9pm screen cutoff buys a full wind-down hour or more before lights out. If you sleep at midnight, shift the rule rather than stretching the gap: no screens for the last 90 minutes of your day. The line that matters most is no infinite scroll in the final hour.
What do I do with the empty evening?
Plan it before the experiment starts: one low-effort thing for your body, one for your mind, one for the room, same order each night. A hole with nothing planned for it is how the phone gets rehired, usually by Thursday.
What if I fail one night?
Lose the hour and start over the next evening without treating it like a moral event. A pattern tolerates swaps but not gaps, so if a birthday lands on an experimental night, move the night instead of canceling the experiment.
This guide is educational, not medical advice. If a week of cleaner evenings doesn’t touch persistent sleeplessness, that’s a conversation for a doctor rather than a failed experiment.
