If you are wondering why caffeine stopped working, the research points to a few separate causes. You get used to a regular dose, and a lot of a daily drinker's morning lift is relief from the gap since the last cup. Sleep, timing and what is really in your cup also change the result.
One part of the picture is still argued over, so we went through the studies on both sides.
Contents
- The short version
- What people mean when they say it stopped working
- Reason 1: you get used to the dose
- Reason 2: much of the lift is the end of withdrawal
- Does the performance edge fade too? The studies split
- Reason 3: sleep takes some of the shine off
- Reason 4: the cup is not the dose you think
- Reason 5: what is still in you from earlier
- What to do with this
The short version
Regular drinkers get used to caffeine's felt effects. In one trial, volunteers taking the equivalent of nine mugs of instant a day no longer noticed its effects after 18 days. Much of what a daily drinker feels as a lift is the end of withdrawal, which can start 12 to 24 hours after the last dose. Whether the performance edge fades as well depends on the study: a single dose works about the same in regular and occasional drinkers, while daily dosing for several weeks has often shrunk it. Sleep, timing and the real dose in your cup move the result too.
What people mean when they say it stopped working
Three different things get lumped together under that sentence, and the evidence is different for each.
- The kick you feel. This has the clearest evidence, and it all points the same way.
- Your alertness and mood through the day. For regular drinkers, much of the lift looks like withdrawal relief.
- Measured performance, such as cycling power or reaction time. Here the studies split.
"Caffeine does nothing for regular drinkers" and "habit makes no difference" are both overstatements. The research sits between them.
Reason 1: you get used to the dose
The cleanest test is also the oldest. In a 1992 trial, 32 healthy adults took either 300 mg of caffeine three times a day or a placebo for 18 days. Afterwards, caffeine still produced clear subjective effects in the placebo group and none in the caffeine group. The authors called it the clearest evidence of complete tolerance to a brain effect of caffeine in humans (Evans and Griffiths, 1992).
That dose is about nine mugs of instant a day. A later study used something closer to a normal habit. Sixteen regular users spent at least 14 days on 400 mg a day, about four mugs, and another 14 days on placebo. Tolerance was complete on the subjective measures. It was not complete on the physiological ones, which included blood flow in the brain and EEG (Sigmon et al., 2009).
So the feeling fades faster than the body's response to the drug. Researchers describe it as the brain adapting to caffeine being around. The result you feel is well documented. The exact changes underneath are less well mapped in people.
Within weeks of daily use, the same dose can feel like much less.
Reason 2: much of the lift is the end of withdrawal
A review of 66 studies on caffeine withdrawal found symptoms usually start 12 to 24 hours after the last dose and peak between 20 and 51 hours. Stopping from as little as 100 mg a day, about one mug of instant, was enough to cause symptoms (Juliano and Griffiths, 2004). An overnight gap is long enough for it to begin.
That matters for how a morning coffee feels. A 2013 study tested 369 people after a night without caffeine. Withdrawal left them sleepier, less alert and slower on reaction and memory tasks. Caffeine put that right in medium-high consumers. In the lighter-drinking group it reduced sleepiness but did little else, because it also made them more anxious and jittery, which cancelled the gain. The authors concluded that frequent use "fails to enhance mental alertness" (Rogers et al., 2013).
An earlier review went further. It argued that when studies are designed to account for withdrawal, the effects of caffeine on performance and mood are almost wholly down to reversing it (James and Rogers, 2005). That is one side of a live debate, and the next section gives the other.
For a daily drinker, the first cup of the day does a lot of its work by ending the dip that started overnight.
If your morning coffee feels like it only gets you back to normal, this is a likely reason.
Does the performance edge fade too? The studies split
Different kinds of study ask different questions, and they get different answers.
| The question | Study | What it found |
|---|---|---|
| Does a regular drinker still respond to one dose? | Meta-analysis of 60 studies (Carvalho et al., 2022) | Caffeine gave a small positive effect on exercise, and habitual intake did not change it |
| Same | 40 trained cyclists split by habit, 58, 143 or 351 mg a day (Gonçalves et al., 2017) | A 6 mg per kg dose, about 420 mg for a 70 kg adult, made them faster (29.92 against 30.81 minutes on placebo) whatever their habit |
| Same | 48 people, 217 or 20 mg a day (Haskell et al., 2005) | After a night without caffeine, 75 or 150 mg improved reaction time and alertness in both groups |
| What happens if you take it every day? | 18 low-habit drinkers for 28 days (Beaumont et al., 2017) | The exercise boost was no longer apparent after four weeks of daily caffeine |
| Same | 11 active adults for 20 days (Lara et al., 2019) | Peak cycling power rose about 4% for the first 15 days, then the effect lessened |
| What if withdrawal is controlled for? | 10 habitual users on 394 mg a day (Griest et al., 2023) | A 10 km time trial improved only when they had gone without caffeine beforehand, not when an earlier small dose had kept withdrawal away |
| Same | 21 people, 13 users and 8 non-users (Bell and McLellan, 2002) | The boost was greater and lasted longer in non-users |
The first block compares people with different habits, each given one dose. It says a regular drinker still responds. The second block follows the same people taking caffeine every day. It says the response can wear down over a few weeks. The third block asks whether the benefit is real once withdrawal is accounted for, and two small studies suggest the benefit is smaller for habitual users.
Those results can all be true together. The daily-dosing and withdrawal-controlled trials are small, 10 to 21 people each, and they mostly measure exercise, so they say less about alertness and focus. What you can say with confidence is narrow: studies disagree on whether daily use blunts the performance edge, and some find it shrinks over weeks.
A regular drinker can still respond to a dose, and a daily habit can still wear that response down over weeks. The studies are answering different questions.
Reason 3: sleep takes some of the shine off
A meta-analysis of 24 studies found caffeine cut total sleep time by 45 minutes and sleep efficiency by 7%, and reduced deep sleep by 11.4 minutes. To avoid losing sleep time, its authors put the cutoff for a coffee, about 107 mg per 250 ml, at least 8.8 hours before bedtime (Gardiner et al., 2023). For an 11pm bedtime, that is about 2pm.
A separate trial gave people 400 mg, about four mugs of instant, 6 hours before their usual bedtime. It measurably disrupted their sleep (Drake et al., 2013).
A short night makes the next day harder to get through. The loop between sleep and caffeine is in why your afternoon coffee is keeping you awake.
Reason 4: the cup is not the dose you think
Lab tests on 20 espressos from high street coffee shops found a six-fold range in caffeine. The strongest single espresso held 322 mg (Crozier et al., 2012).
The chains differ just as much. A 350 ml americano has 89 mg at Starbucks and 325 mg at Costa, more than three mugs of instant. Change chain, size or order and your dose can move a long way without a thing feeling different. We went through it drink by drink in how much caffeine is in high street coffee.
If your coffee used to be a double and is now a single, or the other way round, that alone can change how the day goes.
Reason 5: what is still in you from earlier
Caffeine has a half-life of about 6 hours for most adults, anywhere from 2 to 10. Every drink starts its own clock, so they stack. A cup at 3pm is still about half there at 9pm, and the one you had before it is still contributing.
So what you drank earlier in the day, and how late you drank it, changes how you feel and how you sleep. The arithmetic is in how long caffeine stays in your system, and the morning side is in the best time to drink coffee.
What to do with this
- Add up a real day, including tea, cola and energy drinks. The caffeine calculator does it in about a minute and shows how much is still in you at bedtime.
- Check the cup. At a chain, ask how many shots are in your drink.
- Look at your last drink of the day, and where it sits against your bedtime.
- Know what a break involves. Withdrawal usually lasts 2 to 9 days. Nobody has measured how many days it takes for sensitivity to return, and no trial has tested whether a break brings back a performance edge, so we do not promise one. What has been measured is in our guide to caffeine tolerance resets.
If you would rather follow a plan
The Uncaffed 30-day reset kit steps your caffeine down on a fixed schedule:
- Reset takes you from 160 mg a day to none over 20 days, five days at each step, with L-theanine alongside.
- Flow has no added caffeine and runs for all 30 days.
- For the 30 days, the capsules are your only caffeine. Decaf, herbal tea and rooibos keep the ritual going.
Heavy caffeine user? If you have more than about 400 mg a day from all sources, try to cut down a little before you start. Dropping straight to the kit's starting level can be a big change if you're not prepared for it. Not sure how much you have? The calculator adds it up. Everyone else can start straight away.
Preorders are open. Kits ship by 30 October, UK only. The quiz helps you find your starting point, and the science page explains how the steps were chosen. The kit is not for under-18s or anyone pregnant or breastfeeding.
