Almost everyone has heard it: drink eight glasses of water a day. It gets printed on water bottles, repeated by well-meaning colleagues, and treated as settled science. It is not. When a kidney physiologist went looking for the evidence behind the rule, he could not find any. That does not mean hydration is unimportant, it means the specific number you have been told to hit was largely invented. Here is the honest version: where the rule came from, what the real guidance actually says, and the simple signals that tell you whether you are drinking enough.
The short version
- There is no solid scientific evidence for the "eight glasses a day" rule. A 2002 review by a kidney physiologist found none.
- Official guidance is about total water, not glasses of plain water: roughly 2.7 litres a day for women and 3.7 litres for men, from all drinks and food combined.
- About 20 percent of your water comes from food, and tea, coffee, and milk all count toward the rest.
- For most healthy people, thirst and urine colour are good enough guides. You do not need to count.
- You genuinely need more in heat, illness, and hard exercise, and it is possible, though rare, to drink dangerously too much.
Where the eight-glasses rule came from
In 2002, Heinz Valtin, a physiologist who literally wrote textbooks on the kidney and water balance, published a review with a blunt title: "Drink at least eight glasses of water a day. Really?" He set out to find the science behind the advice and came back with a striking answer. There was no scientific study supporting the idea that healthy adults need eight glasses of plain water on top of everything else they drink and eat.
So where did it come from? The most likely origin is a 1945 recommendation that people take in about 1 millilitre of water for each calorie of food, which for a typical diet works out near 2 to 2.5 litres. The catch is the next sentence, which noted that most of that water is already contained in prepared foods. That crucial part got dropped in the retelling, and a recommendation about total water intake mutated into a command to drink eight extra glasses.
"Total water" includes the water in your food and every drink you have, not just plain water. Miss that word and a sensible total turns into an exhausting daily quota of glasses you were never meant to add on top.
What the actual guidance says
The most cited modern figures come from the US National Academies, which looked at what adequately hydrated people actually consume. The numbers are given as total water from all sources, food included.
Read those correctly and the picture relaxes a lot. If a fifth of your water comes from food, and tea, coffee, milk, and juice all count, the amount of plain water you personally need to sip can be well under eight glasses. The National Academies were also clear that these are population averages for guidance, not a target everyone must hit, and that healthy people can let thirst do most of the work.
It is also worth saying that your number is genuinely your own. A tall, heavily built person who trains outdoors in summer needs far more than a smaller person working in an air-conditioned office. Body size, activity, temperature, altitude, and even a salty meal all move the figure around from day to day. That variability is exactly why a fixed count of eight glasses was always a blunt instrument, and why a signal that responds to your actual state beats a number printed on a bottle.
Large surveys of what people actually drink suggest most healthy adults are getting enough water without counting anything. If you are not unusually thirsty and your urine is pale, you are almost certainly fine.
Food counts, and so does your coffee
Two persistent myths make hydration feel harder than it is. The first is that only plain water counts. The second is that coffee and tea dehydrate you. Both are wrong for everyday amounts.
Caffeine is a mild diuretic, but the fluid in a normal cup of coffee or tea more than makes up for the small extra loss, so these drinks give you a net gain of water. And food is a genuine source: many fruits and vegetables are more than 80 to 90 percent water, and dishes like soup, stew, porridge, yoghurt, and beans all carry water into your day.
| Source | Roughly how much water | Notes |
|---|---|---|
| Cucumber, lettuce, courgette | About 95% water | Salads quietly hydrate you |
| Watermelon, strawberries, oranges | About 90% water | Fruit counts, not just drinks |
| Soup, broth, stew | Mostly water | A warm way to top up in cold weather |
| Tea and coffee | Net water gain | The diuretic effect is small at normal intakes |
| Beans, lentils, cooked grains | Absorb water when cooked | Add to the day's total |
None of this means water is pointless, it is still the simplest, cheapest, calorie-free way to top up. It just means you can stop treating hydration as a plain-water-only accounting task. There is one sensible caveat: sugary drinks and juices carry water but also a lot of sugar and calories, so leaning on them to hydrate is a poor trade. Water, plain or sparkling, unsweetened tea and coffee, milk or a fortified plant milk, and water-rich foods do the job without the downside.
How to actually tell if you are hydrated
Your body has a very good hydration monitor built in, and for healthy people it is more reliable than any glass count. Instead of chasing a number, watch two simple signals.
- Let thirst guide most of it. Thirst is a genuine, well-calibrated signal in healthy adults. Drinking when thirsty, and a bit with meals, covers most people most of the time.
- Check the colour of your urine. Pale straw is the target. Consistently dark yellow suggests you could drink a little more. Completely clear all day can mean you are overdoing it.
- Watch the harder days. Heat, exercise, alcohol, illness with fever, vomiting, or diarrhoea all raise your needs. On those days, drink proactively rather than waiting for thirst.
- Mind the exceptions. Older adults feel thirst less reliably, and some medical conditions and medicines change fluid needs. If that is you, follow your doctor's advice over any general rule.
The idea that "if you are thirsty you are already dangerously dehydrated" is overstated for everyday life. Thirst is an early, normal prompt, not an emergency alarm. It is fine to drink in response to it rather than ahead of it.
Can you drink too much water?
Yes, though it is uncommon and usually takes real effort. Drinking a very large amount of water in a short time, faster than your kidneys can clear it, can dilute the sodium in your blood. This is called hyponatraemia, and in severe cases it is dangerous. It has happened in endurance events and in water-drinking challenges, where people force down litres well beyond thirst.
The practical takeaway is simple and calming. Forcing yourself to hit a high water target you do not want is not virtuous, and chugging enormous volumes "to be safe" is the one scenario with real downside. For everyday life, drinking to thirst, plus a bit with meals and more on hot or hard days, is both safe and sufficient.
Hydration matters. The eight-glasses rule does not. Aim for pale urine, drink when thirsty, count food and all your drinks, and push fluids in heat, illness, and exercise. That covers the science without the mythology.
Questions people ask
So is eight glasses of water a day wrong?
Does tea and coffee count toward my water intake?
How do I know if I am drinking enough?
Can drinking too much water be harmful?
References
- Valtin H. American Journal of Physiology - Regulatory, Integrative and Comparative Physiology. "Drink at least eight glasses of water a day." Really? Is there scientific evidence for "8 x 8"?. 2002. doi.org/10.1152/ajpregu.00365.2002
- US National Academies of Sciences, Engineering, and Medicine. Report Sets Dietary Intake Levels for Water, Salt, and Potassium. 2004. nationalacademies.org/news/report-sets-dietary-intake-levels-for-water-salt-and-potassium-to-maintain-health-and-reduce-chronic-disease-risk
- Rosinger A, Herrick K. NCHS Data Brief, Centers for Disease Control and Prevention. Daily Water Intake Among US Men and Women, 2009-2012. 2016. cdc.gov/nchs/products/databriefs/db242.htm
Last reviewed 18 July 2026. We check health-condition articles against current guidelines and update the date above when we do.
KF.Social Editorial