How late is too late for coffee? Stop 9 hours before bed
A review of 24 caffeine studies found about 45 minutes less sleep, and a small trial found lost sleep even 6 hours before bed. Four ways to set a coffee cutoff.

An afternoon coffee feels harmless. You drink it at 3pm and still fall asleep at 11pm like any other night. But feeling sleepy is a poor test of whether caffeine has left your system, and the research suggests the cutoff is earlier than most of us assume.
Two studies, one pooled review and one small trial, point the same way. Neither is the last word, but together they give you a practical rule to start from.
What the studies found
Gardiner and colleagues pooled 24 controlled studies of healthy adults aged 18 to 65. Across the trials, caffeine was linked to about 45 minutes less total sleep time and about 7% lower sleep efficiency, with about 9 more minutes to fall asleep.
The review then asked when caffeine stops mattering. Its model suggests a cup of coffee (107 mg) should be finished at least 8.8 hours before bed to avoid a significant drop in total sleep time. A standard pre-workout dose (217.5 mg) needed about 13.2 hours. For a single cup of black tea (47 mg), the model found no clear cutoff.
A smaller trial shows what this looks like in practice. Drake and colleagues gave 12 healthy adults 400 mg of caffeine, or a placebo, in pill form at 0, 3 or 6 hours before their usual bedtime. They slept at home with a portable sleep monitor. At every timing, total sleep time dropped by between 1.1 and 1.2 hours compared with placebo, including the dose taken 6 hours before bed. Only the sleep monitor picked up the 6-hour effect. The volunteers’ own sleep diaries did not show a clear change, so you may lose sleep without noticing.
The caveats
These numbers are estimates. Drake’s trial was small: 16 volunteers qualified and four were dropped for breaking the protocol, leaving 12. The review covered healthy adults with low to moderate caffeine habits, and its authors note the results may not hold for teenagers, older adults or heavy caffeine users. The cutoffs are based on statistical significance, not on how you feel the next day.
Caffeine also clears at different speeds. The review notes its half-life varies from about 2 to 10 hours between healthy adults, so treat the cutoff as a starting point and adjust to how you sleep.
1. Set a cutoff about 9 hours before bed
Take your usual bedtime and count back nine hours. For an 11pm bedtime, that is 2pm. For 10pm, it is 1pm, which matches the review’s own example of 1:12pm.
Put the time somewhere you will see it, such as a phone alarm labelled “last coffee”. A clear time removes the daily decision about whether this cup is the exception.
2. Count the hidden caffeine
Coffee is not the only source. Cola, chocolate and energy drinks carry caffeine too, and the review’s typical pre-workout dose (217.5 mg) was about twice its typical cup of coffee (107 mg). That dose needed around 13 hours.
If you train in the evening, check the label on your supplement. Black tea is milder: the review found no clear cutoff for one cup.
3. Swap the afternoon cup
The afternoon cup is often about the break, not the caffeine. Keep the break and change the drink: decaf, or a short walk outside. Neither is something the studies tested, so treat this as a practical swap, not a proven one.
Tie it to the moment you would usually reach for the kettle, so the new routine has a cue.
4. Keep the morning coffee
You do not need to give up coffee. The review’s model found no clear loss of sleep for cups finished before the cutoff, so your first cup is not the problem.
Let the cutoff protect your evening rather than ban your morning.
Start with one time
Pick your cutoff tonight, set the alarm, and leave your usual morning coffee alone. Sleep regularity matters too, and an earlier cutoff only helps if your bedtime stays steady.
Related reading: Sleep regularity may matter more than how long you sleep.
Sources
-
Gardiner, C., Weakley, J., Burke, L. M., Roach, G. D., Sargent, C., Maniar, N., Townshend, A., & Halson, S. L. (2023). The effect of caffeine on subsequent sleep: A systematic review and meta-analysis. Sleep Medicine Reviews, 69, 101764.
-
Drake, C., Roehrs, T., Shambroom, J., & Roth, T. (2013). Caffeine effects on sleep taken 0, 3, or 6 hours before going to bed. Journal of Clinical Sleep Medicine, 9(11), 1195–1200.





