How Long Should You Read a Day? What the Research Actually Says
Twenty to thirty minutes a day is the honest answer. Here is the evidence behind it, and what 10, 20, 30 and 60 minutes compounds to in books a year.
By Daniel Daneshi August 5, 2026 9 min read TL;DR: Aim for 20 to 30 minutes a day. That is the band where the health research is strongest, and at an average adult reading speed of about 250 words a minute it compounds to roughly 20 to 30 books a year. Below that, 10 minutes a day still beats an hour once a week, because a daily habit survives a bad month and a weekly one does not. Almost none of this evidence is causal, so treat the numbers as a target, not a prescription.
Search “how long should I read a day” and you will get a wall of confident numbers: 20 minutes, 30 minutes, an hour, six minutes to cut your stress by 68 percent. Most of them trace back to two or three studies that have been quoted so many times the original meaning has worn off. So here is the actual evidence, with the sources, along with honest arithmetic on what each amount of daily reading turns into over a year.
The short answer: 20 to 30 minutes
Three separate lines of evidence converge on roughly the same band.
Longevity. The most cited study is a Yale School of Public Health analysis of 3,635 older adults in the Health and Retirement Study, published in Social Science and Medicine in 2016. Over up to 12 years of follow-up, book readers had about a 20 percent lower risk of dying than non-readers after adjusting for age, sex, race, education, wealth, self-rated health, comorbidities, marital status and depression. There was a dose response: people reading up to 3.5 hours a week had a hazard ratio of 0.83, and people reading more than 3.5 hours a week 0.77. The authors framed the benefit around readers averaging about 30 minutes a day. Two caveats matter. It is observational, so it cannot prove reading caused the difference. And the headline “23 months longer” number that circulates online is the unadjusted figure; once covariates are controlled the survival advantage in the paper is much smaller.
Cognitive aging. Wilson and colleagues at Rush, publishing in Neurology in 2013, followed older adults to autopsy and found that frequent mentally stimulating activity in late life (reading, writing, games) was associated with a rate of memory decline about 32 percent slower than average, while infrequent activity ran 48 percent faster. Again observational, and reading is bundled with other activities. But it points the same direction.
Sleep. This one is actually randomised. The People’s Trial, an online pragmatic trial of 991 people published in Trials in 2021, asked half the participants to read a book in bed before sleep for a week. Of those who reported back, 42 percent of readers felt their sleep improved versus 28 percent of non-readers. It is self-reported and unblinded, but it is the rare piece of reading research with a control group.
Nobody has run a trial assigning adults to 10 versus 30 versus 60 minutes a day for a decade, and nobody will. So “20 to 30 minutes” is a reasonable target drawn from correlational data, not a dose confirmed by experiment.
Where the famous “20 minutes a day” number comes from
The chart you have seen on a classroom wall (20 minutes a day equals 1.8 million words a year) is real, but it is not a recommendation. It comes from Anderson, Wilson and Fielding’s 1988 study in Reading Research Quarterly, in which 155 fifth graders kept diaries of how they spent time outside school. The researchers then ranked them. Here is the table as it is usually reproduced:
| Percentile rank | Minutes of book reading per day | Words read per year |
|---|---|---|
| 98th | 65.0 | 4,358,000 |
| 90th | 21.1 | 1,823,000 |
| 70th | 9.6 | 622,000 |
| 50th | 4.6 | 282,000 |
| 30th | 1.8 | 106,000 |
| 10th | 0.1 | 8,000 |
So “20 minutes a day” is a description of what a 90th-percentile ten-year-old did in 1988, not a dose anyone prescribed. It is also correlational: as literacy researcher Timothy Shanahan points out, better readers plausibly read more because they are better readers, and parental education could drive both. What the table does give you is a genuinely useful sense of scale. The gap between a median reader and a top-decile reader is about 17 minutes a day.
Volume does look like it matters, though. Mol and Bus’s 2011 meta-analysis of 99 studies covering 7,669 people found print exposure explained 12 percent of the variance in oral language skills in preschool, rising to 19 percent in middle school and 30 percent in high school. Cunningham and Stanovich found reading volume predicted vocabulary even after controlling for general ability and decoding skill. More reading is associated with a bigger vocabulary in a way that is hard to explain away entirely.
What each amount actually compounds to
Here is the arithmetic, stated honestly. Assumptions: 250 words a minute, which sits between the 238 wpm for non-fiction and 260 wpm for fiction that Brysbaert’s 2019 meta-analysis of 190 studies and 18,573 participants found for adults reading English prose; and a 90,000 word book, the standard average for an adult novel, which is roughly 360 pages at 250 words a page.
| Minutes a day | Words a year | Pages a year | Books a year (every day) | Books a year (5 days out of 7) |
|---|---|---|---|---|
| 10 | 912,500 | 3,650 | 10 | 7 |
| 20 | 1,825,000 | 7,300 | 20 | 15 |
| 30 | 2,737,500 | 10,950 | 30 | 22 |
| 45 | 4,106,250 | 16,425 | 46 | 33 |
| 60 | 5,475,000 | 21,900 | 61 | 44 |
Two things fall out of that.
The rule of thumb is embarrassingly clean. At these assumptions, books per year is roughly equal to minutes per day. Twenty minutes a night is twenty books a year. That is a far better mental model than any target number, and it puts you well past the average person’s annual book count.
Your mileage varies a lot. A slow reader on dense non-fiction at 175 wpm and a 150,000 word book gets less than half these numbers. A fast reader on 70,000 word thrillers gets nearly double. Plug your own book in with our reading time calculator rather than trusting the table.
Note too that 20 minutes a day gives you 1.825 million words a year, almost exactly the 1.823 million the 90th-percentile fifth grader read. The viral number is not wrong about the scale. It is just describing an avid reader rather than setting a floor.
How much do people actually read?
Almost none of it. Analysis of the American Time Use Survey by the Humanities Indicators project found that only 14.6 percent of Americans aged 15 and over read for more than 20 minutes on a given day, down from 22.3 percent in 2003. And there is a huge age gap: in the 2024 survey, adults 75 and over averaged 46 minutes a day while teenagers averaged nine. More numbers like this are in our reading statistics roundup.
Which is the reassuring part of this whole article. Twenty minutes a day puts you in the top sixth of adults by reading time. You are not behind.
About that “six minutes reduces stress by 68 percent” figure
You will see this everywhere, and it is worth being precise about. The claim traces to a 2009 piece of research by Dr David Lewis at Mindlab International, University of Sussex, commissioned by a chocolate brand. It was reported widely in the UK press and is still repeated as established fact, but the only citation anyone gives is “Lewis, D. (2009), Galaxy Stress Research, Mindlab International” with no peer-reviewed paper, no sample size and no published method behind it. Reading probably does calm you down. The specific 68 percent is a press release, not a finding, and we would not build a habit around it.
What to actually do tonight
The research supports a target, not a rule. Here is how to turn 20 to 30 minutes into something that survives February.
- Set the floor low and the target high. Make your non-negotiable minimum 10 minutes and your aim 30. The floor exists for your worst day. The whole logic behind this is in our guide to building a reading streak.
- Make it a minutes goal, not a books goal. Books-per-year targets punish you for picking long books. Minutes reward the behaviour you control. We walk through choosing the right unit in how to set a reading goal.
- Attach it to a fixed cue. The before-bed slot has the randomised evidence on its side, and it replaces a screen with something better.
- Time it, do not estimate it. People are terrible at guessing how long they read. This is what the booksense timer is for: start a session, read, and the minutes land in your streak, your goal and that book’s stats automatically. Readers have logged 480K+ minutes this way across 12K+ finished sessions.
If you keep hitting the target and still feel like you are not getting anywhere, the problem is usually retention rather than time, and how to remember what you read is the better next read. If you cannot get started at all, try getting out of a reading slump first, then come back to the clock. And for the full system around the habit, our pillar guide on how to read more books covers the rest.
Twenty minutes tonight. That is twenty books this year. Download booksense free on the App Store, set a daily minutes goal, and let the timer count it.
FAQ
How long should I read a day?
Twenty to thirty minutes a day is the best-supported target. It is the level the longevity research describes (book readers averaging about 30 minutes a day had the largest survival advantage in the Yale analysis of the Health and Retirement Study), and at an average adult reading speed it compounds to roughly 20 to 30 books a year. If 20 minutes feels out of reach, 10 minutes daily still beats an hour once a week, because the daily version is the one that survives a busy month.
Is 20 minutes of reading a day enough?
Yes, for almost everyone. At 250 words a minute, 20 minutes a day is about 1.8 million words a year, or roughly 20 average novels. It is also close to the level of the most avid readers in the classic school study the viral 20 minute figure comes from. Twenty minutes is a genuinely strong daily habit, not a minimum you should feel bad about.
How many books a year is 30 minutes of reading a day?
About 30 books a year if you read every day, or about 22 if you read five days out of seven. That assumes 250 words a minute and a 90,000 word book. Dense non-fiction or long fantasy will pull the number down, and short or fast-paced books will push it up.
Does reading really help you live longer?
The best-known study, a Yale analysis of 3,635 older adults in the Health and Retirement Study, found book readers had about a 20 percent lower risk of dying over 12 years of follow-up than non-readers. It is observational, so it shows an association rather than proof that reading caused it. Treat it as encouraging, not as a prescription.