Reading Time Calculator
Paste content and estimate reading time, rounded to the minute. English-type text is timed at an adjustable 238 words per minute, Chinese and Japanese by character at stated, adjustable rates. Word, sentence and paragraph counts included.
Reading Time Calculator
238 words a minute is Brysbaert's 2019 mean for adults aged 18–60 reading non-fiction silently in their native language, from studies in Latin-alphabet languages, most of them English; most adults fell between 175 and 300. Chinese 255 and Japanese 357 characters a minute are reading-aloud rates, from the IReST study (Trauzettel-Klosinski & Dietz 2012: native readers aged 18–35, counting characters without punctuation). Thai, Lao, Khmer and Burmese are counted but not timed.
How to use the Reading Time Calculator
Paste your content
Paste your article, newsletter, script or any text. The tool counts words, sentences and paragraphs as you type, and strips HTML tags first, so you can paste from a CMS preview. Script and style code is skipped. Chinese and Japanese are timed by character; Thai, Lao, Khmer and Burmese are counted as words but not timed, because none of our sources gives a rate for them.
Set the reading speed
The default, 238 words per minute, is the mean in Marc Brysbaert's 2019 meta-analysis for adults aged 18–60 reading non-fiction silently in their native language — studies in Latin-alphabet languages, most of them English; most adults fell between 175 and 300. Fiction averaged 260 in a separate study of 48 book readers. Chinese and Japanese have their own rates, in characters a minute; they were measured with readers reading aloud, so raise them if your readers read silently and you know their pace. Lower the rate for younger readers, older readers, readers of English as a second language, or dense technical text — the study found all of those read slower.
Read the estimate
You get the estimate rounded to the nearest minute ("Under 1 min" below half a minute), the unrounded figure beneath it, and word, sentence and paragraph counts. For English-type text the result also gives the time at 175 and at 300 words a minute. Treat it as the time for a typical reader: people differ widely, so one reader's five minutes is another's seven.
Let the subject set the length
Write to the length the subject needs. Google's own guidance on helpful content asks whether you are writing to a particular word count because you heard Google has a preferred one — and answers "No, we don't."
How the estimate is worked out
A reading-time estimate is words divided by a reading rate. Which rate, and measured how, is most of the question — and many tools use a rate higher than the evidence supports.
Where 238 words per minute comes from
In 2019 Marc Brysbaert published "How many words do we read per minute? A review and meta-analysis of reading rate" in the Journal of Memory and Language. From 190 studies and 18,573 participants he estimated the average silent reading rate for adults at 238 words per minute for non-fiction. The studies are of native readers aged 18 to 60 in languages written in the Latin alphabet, most of them English; the English rows alone average 236. His 260 for fiction comes from a separate study of 48 people each reading a book. He found that "most adults fall in the range of 175–300 wpm" for non-fiction. Reading aloud, from 77 studies, averaged 183. Crucially, he concluded that these estimates "are lower than the numbers often cited in scientific and popular writings" — the familiar 250 or 300 wpm figures overstate how fast people read.
Adults read English non-fiction silently at about 238 words a minute — slower than the figures usually quoted, not faster.
Who reads slower
The same study reports that reading rates are lower for children, for older adults and for readers with English as a second language, and that within any group there are reliable individual differences. It also explains the fiction–non-fiction gap by word length: non-fiction uses longer words. For an audience reading in a second language, or for technical material, lower the rate and the estimate becomes more honest.
Chinese and Japanese are timed by character
Chinese and Japanese are written without spaces, so there is no word count to divide. The tool counts Chinese characters, and Japanese kanji and kana, leaving punctuation out, and divides by a characters-per-minute rate: 255 for Chinese and 357 for Japanese by default. Those are the IReST norms (Trauzettel-Klosinski and Dietz, Investigative Ophthalmology & Visual Science, 2012), measured with native readers aged 18 to 35 who read standard passages aloud against a stopwatch. A line containing any kana is treated as Japanese. Mixed text is the sum of its parts: English words at the words-a-minute rate plus the characters at theirs.
Silent reading is faster than reading aloud — Brysbaert's review gives Chinese 260 words a minute silently against 152 aloud — but there is no agreed silent rate in characters: his estimate depends on how many characters make a word, and we found no silent-reading figure for Japanese at all. If you know how fast your readers read, set the rate. Thai, Lao, Khmer and Burmese are also written without spaces; your browser's word segmenter counts them, but none of these sources gives a rate, so the tool says "Not estimated" rather than guess.
What we changed on 18 October 2026
Chinese, Japanese and Thai text used to show a reading time of "0 sec" beside a Words count in the thousands, because only space-separated words were timed. Chinese and Japanese are now timed as described above, and text that cannot be timed says so. The estimate is rounded to the minute, as our own FAQ advises, rather than given to the second. A reading speed of 0, a negative number or a non-number now gives a message instead of silently becoming 238, and the code inside <script> and <style> tags is no longer counted as words. We also removed a heading claiming that reading-time labels lift engagement, which nothing we found supports, and speed hints of 180 for "slow" and 500 for "skim", which no source we opened gives.
What we changed on 30 September 2026
Words are now counted by the same code as our PDF Word Counter. Before, a paragraph in Chinese, Japanese or Thai — scripts written without spaces — counted as a single word; those words are now found with your browser's word segmenter and reported, though not timed, because the 238 wpm rate is measured for English words. We also removed advice that had no source: a claimed 10–15% fall in bounce rate from reading-time badges, an "average top-ranking page of 2,000 words", and a sharp drop in retention "past 12 minutes".
Length and search
Google's guidance on helpful content includes a self-assessment question: "Are you writing to a particular word count because you've heard or read that Google has a preferred word count? (No, we don't.)" A reading-time estimate is for your readers, to help them decide whether they have time now — not a target.
Brysbaert's 238 wpm, 183 aloud, and why popular figures run high
Adults reading non-fiction silently in their native language average 238 words per minute (Brysbaert 2019: 190 studies, mostly English, 18,573 participants); fiction averaged 260 in a separate study of 48 book readers.
Most adults read English non-fiction silently at between 175 and 300 words per minute; fiction, 200 to 320.
Reading aloud averages 183 words per minute, from 77 studies and 5,965 participants.
Brysbaert found his estimates "lower than the numbers often cited in scientific and popular writings" — the common figures overstate reading speed.
Non-fiction reads slower than fiction mainly because its words are longer, by the same analysis.
Reading rates are lower for children, older adults and readers with English as a second language.
In Norvig's analysis of Google Books, the average English word is 4.79 letters long.
Google's helpful-content guidance: are you writing to a word count "because you've heard or read that Google has a preferred word count? (No, we don't.)"
In the IReST study, native readers aged 18 to 35 read Chinese aloud at 255 characters a minute and Japanese at 357; this tool uses those as its default character rates.
At 238 wpm a 1,000-word article is about 4 minutes 12 seconds; at 175 wpm it is about 5 minutes 43.
Frequently Asked Questions
-
Because 238 is what the largest review of the evidence found. Brysbaert's 2019 meta-analysis of 190 studies — native readers aged 18 to 60, in Latin-alphabet languages, most of them English — put silent reading of non-fiction at 238 words per minute, and concluded that commonly quoted figures are too high rather than too low. Use a lower rate for audiences the study found read more slowly: children, older adults and second-language readers.
-
By character, not by word: Chinese at 255 characters a minute and Japanese (kanji plus kana) at 357, the IReST rates for native readers reading aloud. Punctuation is not counted. Silent reading is faster, but no single silent rate in characters is agreed, so change the rate if you know your readers. English words in the same text are added at the words-a-minute rate. Thai, Lao, Khmer and Burmese are counted but not timed.
-
If it helps your readers decide whether to start now. We found no reliable published figure for its effect on engagement, so we don't quote one; round the minutes, and don't show it for pieces under a minute.
-
Google does not describe reading time as a ranking factor, and its helpful-content guidance says it has no preferred word count. Write for the reader; the estimate is for them.
-
It is right for an average reader and wrong for most individuals: in Brysbaert's data most adults read non-fiction at 175–300 wpm, so the same 1,000 words take between about 3 minutes 20 seconds and 5 minutes 43. Topic knowledge and difficulty move it further.
-
As long as the subject needs. Google says it has no preferred word count. A reading-time estimate tells a reader what they are committing to; it is not a length target.
-
Brysbaert's review found reading rates in line with maximum listening speed, and most adults below 300 wpm for non-fiction. Much faster "reading" is skimming, which covers more text and retains less.
-
Paste it as it is: the tool removes HTML tags before counting. What to include — headings, captions, footnotes — is your choice; count what a reader actually reads.
-
No. All analysis runs in your browser via JavaScript. Open DevTools → Network and confirm zero outbound requests. Content text + word counts stay on your device. Safe for unpublished drafts, confidential reports, or pre-release content analysis.
-
Comparing time on page with the estimated reading time is a rough sign of whether people finish a piece, but analytics tools measure time differently and a reader's pace varies widely, so treat any ratio as a clue rather than a verdict.
-
The PDF Word Counter counts documents with the same rules; the SEO Title Tag Checker and Meta Description Checker measure the title and snippet a search result shows.
Method & sources
How it computes
Strips HTML (script/style bodies too). Words in spaced scripts / wpm, default 238 (Brysbaert 2019: silent, non-fiction, native readers 18-60, Latin-alphabet studies, mostly English); Chinese characters / 255 and Japanese kanji+kana / 357 per minute (IReST 2012, read aloud), punctuation excluded; mixed text summed; Thai, Lao, Khmer, Burmese counted, not timed. Rounded to the minute. RecaReadingTime kernel, tested.
What this tool implements
- Brysbaert (2019), 190 studies, 18,573 participants: 238 wpm non-fiction, 260 fiction; most adults 175-300 (non-fiction); 183 wpm aloud; the estimates are lower than figures often cited. The page had called older 200 wpm figures too low and put reading aloud at 150-160.
- Until this review a Chinese paragraph counted as one word. Removed unsourced figures: Medium badges cutting bounce 10-15%, top pages averaging 2,000 words, retention dropping past 12 minutes, L2 readers 30-40% slower.
- Google's helpful-content guidance: writing to a word count because Google prefers one? "(No, we don't.)"
- Corrected 2026-10-18: Chinese, Japanese and Thai text had shown "0 sec" beside a non-zero word count. Chinese and Japanese are now timed by character at the IReST reading-aloud rates (Table 2: Chinese 255, Japanese 357 characters/min, native readers aged 18-35), labelled as aloud; Thai, Lao, Khmer and Burmese say "Not estimated". The estimate is rounded to the minute; 0, negative and non-numeric rates give a message; script/style contents are not counted. 238 is labelled as the mean for native readers aged 18-60 in Latin-alphabet studies (144 of 190 English); 260 for fiction is from a separate 48-reader study. Removed: the "lifts engagement" heading and the 180 "slow" / 500 "skim" hints.
Sources
- Brysbaert M. How many words do we read per minute? A review and meta-analysis of reading rate. J Mem Lang. 2019;109:104047. https://doi.org/10.1016/j.jml.2019.104047
- Google Search Central. Creating helpful, reliable, people-first content. https://developers.google.com/search/docs/fundamentals/creating-h…
- Norvig P. English Letter Frequency Counts: Mayzner Revisited. https://norvig.com/mayzner.html
- Trauzettel-Klosinski S, Dietz K. Standardized assessment of reading performance: the new International Reading Speed Texts IReST. Invest Ophthalmol Vis Sci. 2012;53(9):5452-5461, Table 2. https://doi.org/10.1167/iovs.11-8284
Pick up where you left off
Stored only in this browser — never sent to our servers.