Reading time is one division: words ÷ words per minute. Everything interesting is in the second number, and most calculators pick it without saying so. This one shows several rates side by side and names each, because “5 min read” means nothing until you know who is doing the reading.

The rates, and where they come from

The most cited measurement is Marc Brysbaert’s 2019 meta-analysis of 190 studies covering some 18,000 readers. Its headline figures for English:

• 238 words per minute — silent reading of non-fiction. The default here.
• 260 words per minute — silent reading of fiction, which moves faster.
• 183 words per minute — reading aloud.

Speaking is slower again. Presenters land around 130 wpm once pauses are counted, and audiobook narration sits near 150 to 160. Skimming — running your eye down a page for one specific answer — reaches 400 and above, but it is not reading and should not be quoted as a reading time.

Two things move a real reader off these numbers: subject, because technical or unfamiliar material is read more slowly and re-read, and language, because a second-language reader is typically 30 to 40 per cent slower. The old “200 wpm” convention that most blog plugins use is not wrong so much as conservative — it quietly assumes a slower reader than the average.

What counts as a word

This tool counts a word as a run of characters containing at least one letter or digit, so a dash or an ellipsis on its own is not a word. It also reports sentences and paragraphs, and it does not end a sentence on a common abbreviation — “Dr. Smith arrived.” is one sentence, not two. When you fetch a URL, navigation, headers, footers and forms are removed first: counting the menu would inflate every estimate on a template-heavy site.

Should you show reading time on the page?

The honest answer is that it depends on what you are asking the reader to do, and the evidence cuts both ways.

• It helps when the number is small. “3 min read” lowers the cost of starting. Medium popularised the label for exactly this reason.
• It can hurt when the number is large. “22 min read” tells a reader on a phone to come back later, which usually means never. For long pieces, a table of contents does the same reassurance work without the deterrent.
• It is not a ranking factor. Google does not read the label. What it can do is change behaviour on the page, and behaviour is what the rankings are eventually made of.

Adding it to a site

Most platforms have it built in — Medium, Substack and Ghost show it automatically, and WordPress themes usually have a toggle or a one-line snippet. If you are computing it yourself, the whole implementation is:

const words = text.trim().split(/\s+/).filter(Boolean).length
const minutes = Math.max(1, Math.round(words / 238))
// "5 min read"

Round up, never show “0 min read”, and count the article body only. If the page has images or embedded video, some sites add a few seconds per image; that is a judgement call, not a standard.

Reading time and length, together

Length follows the question, not the other way round. A definition deserves 300 words; a comparison of eight tools deserves 2,500 and will be skimmed anyway. What is worth watching is the shape: sentences past 25 words on average, or paragraphs past four sentences, slow a reader down more than the word count does. The keyword density and readability checker scores that side, and the meta tag checker shows how the page will be introduced in search before anyone starts reading.

Frequently asked

How many words is a 5 minute read?

About 1,200 words at 238 wpm, or 1,000 at the conservative 200 wpm most blog plugins use. A 10 minute read is roughly 2,400 words.

How long does it take to read 1,000 words?

Just over four minutes silently, about five and a half minutes read aloud, and around seven and a half minutes at presentation pace.

Is the text I paste sent anywhere?

No. The counting runs in your browser. The only request this page makes is when you ask it to fetch a URL, and then it fetches that page, not your text.

Why is this calculator’s estimate shorter than my blog’s?

Most blog plugins assume 200 wpm; the measured average for non-fiction is 238. Both are defensible. Use the slider to match whichever your audience resembles.