Wind-Down Routines
Caffeine Cutoff: How Early Is Early Enough
By Nora Vale · August 13, 2025 · 7 min read
It’s 3:50 on a Tuesday, the afternoon has gone soft at the edges, and the coffee order feels less like a beverage than like rescue. I say that as a loyal customer, but the thing caffeine sells you isn’t energy. It’s a loan against tonight, and the question of caffeine before bed isn’t really about the hour you lie down. It’s about when you paid back the last cup.
The honest answer to “how early is early enough” comes from arithmetic, not vibes, and the arithmetic is set by your bedtime rather than by the clock on the wall or by whatever the person ahead of you in line just ordered. The caffeine cutoff is also one of the two moves, alongside dimming the house, that carry the most weight in the whole evening, and the wider routine it belongs to lives in the wind-down routine guide.
Key takeaways
- The Sleep Foundation guideline is no caffeine within eight hours of bedtime, ten or more if sleep is already poor.
- A typical adult half-life runs five to six hours, so a 200 mg cup at 4 p.m. leaves about fifty milligrams circulating at 2 a.m.
- Falling asleep fine is not evidence that the cutoff doesn’t matter; the damage shows up as lighter, fragmented sleep.
- The cutoff only counts if it counts everything: tea, cola, chocolate, energy drinks, and decaf all carry a remainder.
- Run the two-week experiment, changing nothing else, and your own log will hand you the personal number.
The half-life math that sets the cutoff
Half-life is the number that turns all of this from theory into a schedule, and for a typical adult it runs about five to six hours, meaning that’s how long your blood takes to clear half of whatever caffeine it’s holding. The Sleep Foundation notes the honest range is wide, anywhere from two to twelve hours depending on the person, but start with the average, because it’s ugly enough.
Run the clock on a two-hundred-milligram large coffee ordered at 4 p.m. By 9 or 10 that night, half of it, a hundred milligrams, is still on board while you wind down. By 2 a.m. a quarter of the original cup, about fifty milligrams, is still circulating, and fifty milligrams is not nothing, it’s a respectable cup of tea arriving at the exact hours your sleep is supposed to be at its deepest and quietest. That’s the arithmetic. It doesn’t care that you fell asleep fine.
The guideline follows straight from that math. Their advice is no caffeine within eight hours of bedtime, and ten or more for anyone whose sleep is already poor, which puts the cutoff for a ten o’clock bedtime at around two in the afternoon, and earlier if you turn in earlier. Eight hours sounds fanatical right up until you do the subtraction yourself. Then it sounds obvious.
Why falling asleep fine doesn’t clear the suspect
Here’s the piece that keeps this problem alive for years. Caffeine can lighten and fragment your sleep even on the nights you fall asleep without any trouble at all, so the person lying awake at 2 a.m. never once suspects the 3 p.m. espresso, because as far as they can tell, coffee has never interfered with their sleep. Falling asleep fast isn’t the test. Staying down in deep, unbroken sleep is the test, and it’s the one the afternoon cup keeps failing.
The book tells the story of Alex, a tax manager who ordered a large latte, an extra shot in the bad weeks, every afternoon at four. He’d tell you sincerely that caffeine has never once interfered with his sleep, because he’s unconscious within minutes of his head hitting the pillow. He even ran informal experiments, skipping the latte one frazzled Friday and sleeping badly anyway, so case closed, coffee was innocent. What that Friday never controlled for was the week it sat inside, the wine with dinner, the way the Thursday latte was still paying out when he lay down. One data point, badly kept, and the suspect walks free for years.
The deeper mechanism is the receipt. Caffeine doesn’t give energy, it hides the receipt, and the receipt is still coming, carrying every bit of sleepiness that piled up while the doors were blocked. Deep sleep is when the brain clears the day’s adenosine backlog, and a caffeinated night can spend hours in the shallow end without clearing much of it, so you wake with the bill mostly unpaid and reach for the morning cup before you’re out of bed. The loan rolls over. By Thursday you’re not drinking coffee, you’re servicing debt.
The cutoff only works if it counts everything
A cutoff experiment dies quietly if the cutoff only counts coffee, because coffee is the honest carrier and everything else lies to you. The FDA’s published numbers make the case better than I can: a cup of brewed tea runs 40 to 120 milligrams, a 12-ounce cola about 30 to 40, a bar of dark chocolate somewhere between 25 and 70, an energy drink anywhere from 50 to 300, and the decaf on the restaurant table still carries 2 to 15 milligrams per cup. None of those are zero. Some of them are a latte in a costume.
Run a quick audit on a normal Tuesday and the arithmetic gets uncomfortable fast. Tea at four because coffee feels too heavy, a square or three of dark chocolate around nine because it feels almost virtuous, decaf after dinner because it’s basically nothing. The running total for an evening you’d call caffeine-free lands somewhere that explains a lot of 2 a.m.’s. The post on hidden sources of caffeine goes through the full list, and whether decaf can keep you awake covers the decaf remainder specifically.
Finding your personal cutoff
Sensitivity genuinely varies, and it isn’t a character trait. The liver enzyme that clears caffeine, which researchers call CYP1A2, runs fast in some people and slow in others, pregnancy roughly doubles the half-life, and smoking roughly halves it, so two colleagues splitting a 3 p.m. pot can be having completely different nights. Genetic tests for the enzyme exist. You don’t need one. Two weeks of your own data will tell you the same thing for free.
The experiment is two moves. First, write down the time of your last caffeine every day, because memory will fudge that number by an hour in coffee’s favor, and the seven-day log in the book’s appendix has a column for exactly this. Second, change nothing else for the two weeks, because if you also move dinner, dim the house, and start a bath routine in the same fortnight, you’ll never know which lever did the work. One variable at a time; it’s slower, and it’s true.
About the afternoon slump, because the cutoff doesn’t delete it. The two-to-four dip arrives on schedule whether or not a latte comes to meet it, and the difference is that without caffeine you actually feel the thing, plain and unedited, for a week or two. It shrinks. Most people find the real dip was a fraction of the crash the latte cycle had been manufacturing, and the morning light habit is the tool that trims whatever’s left of it.
The honest limits
The cutoff is one of the cheapest levers in this entire book, and it still has edges. It does nothing about the wine, the phone, or a body clock that’s already drifted late, so a perfect cutoff can coexist with a mediocre night for a while. It’s a necessary habit, not a sufficient one.
And the timing rule has real exceptions that belong to real schedules. Night-shift workers run a different clock entirely, and the cutoff belongs to their target sleep window, not to the sun’s. If your evenings run late by necessity, the honest move is counting backward from when you actually sleep, not from when the textbooks say you should. The cutoff follows the bedtime; it doesn’t lead it.
Frequently asked questions
How many hours before bed should I stop drinking caffeine?
The Sleep Foundation’s guideline is no caffeine within eight hours of bedtime, and ten or more for anyone whose sleep is already poor. For a ten o’clock bedtime that puts the cutoff around two in the afternoon, counted from the last sip, with tea and chocolate included.
Why does afternoon coffee still affect sleep if I fall asleep fine?
Falling asleep fast isn’t the test. Caffeine can lighten and fragment your sleep even on nights you drift off without trouble, so the person lying awake at 2 a.m. never suspects the 3 p.m. espresso. Staying down in deep, unbroken sleep is the test the afternoon cup keeps failing.
Does the cutoff apply to tea, chocolate, and decaf?
Yes. The FDA’s numbers make the case: brewed tea runs 40 to 120 milligrams, a 12-ounce cola about 30 to 40, dark chocolate 25 to 70 per bar, and even decaf carries 2 to 15 milligrams per cup. None of those are zero, and some of them are a latte in a costume.
Why does coffee hit my partner differently than me?
The liver enzyme that clears caffeine, CYP1A2, runs fast in some people and slow in others, pregnancy roughly doubles the half-life, and smoking roughly halves it. Two colleagues splitting a 3 p.m. pot can be having completely different nights while each swears the other is exaggerating.
Will the afternoon slump get worse without caffeine?
For a week or two you’ll feel the dip plain and unedited, and then it shrinks. Most people find the real dip was a fraction of the crash the latte cycle had been manufacturing, and morning light is the tool that trims whatever’s left.
Count backward from your bedtime, set the cutoff in writing, and let the experiment hand you the verdict. The coffee will still be there tomorrow morning, and it behaves better in the morning anyway. The full system, cutoff included, is in the book. Get the Quiet Core Reset book.
This guide is educational, not medical advice. If sleep problems persist for months or arrive with symptoms you can’t explain, a doctor is the right door.
