At the 238 WPM research benchmark, 500 words take 2 min 6 sec to read, 1,500 words take 6 min 18 sec and 2,500 words take 10 min 30 sec. The table below gives the reading time for eleven word counts from 100 to 10,000 at three speeds, and the sections after it explain how to choose a column and what to do when your word count falls between rows.

Reading time by word count at three reading speeds
Word countCareful (180 WPM)Benchmark (238 WPM)Brisk (300 WPM)Label at 238 WPM
10033 sec25 sec20 sec1 min read
2501 min 23 sec1 min 3 sec50 sec2 min read
5002 min 47 sec2 min 6 sec1 min 40 sec3 min read
7504 min 10 sec3 min 9 sec2 min 30 sec4 min read
1,0005 min 33 sec4 min 12 sec3 min 20 sec5 min read
1,5008 min 20 sec6 min 18 sec5 min7 min read
2,00011 min 7 sec8 min 24 sec6 min 40 sec9 min read
2,50013 min 53 sec10 min 30 sec8 min 20 sec11 min read
3,00016 min 40 sec12 min 36 sec10 min13 min read
5,00027 min 47 sec21 min 1 sec16 min 40 sec22 min read
10,00055 min 33 sec42 min 1 sec33 min 20 sec43 min read

How to read the table

Each row is a word count. The three middle columns show the exact reading time, to the second, at three speeds:

  • Careful (180 WPM): a round planning figure for dense or unfamiliar text.
  • Benchmark (238 WPM): the average silent reading rate for English non-fiction among adults, from the meta-analysis by Brysbaert (2019).
  • Brisk (300 WPM): a round planning figure for easy text read quickly.

The last column is the label a site would show at the benchmark speed, such as “5 min read”. It is the exact time rounded up to the next whole minute, never down, with a minimum of one minute. That rule explains two rows that look odd at first. A text of 100 words takes between 20 and 33 seconds across the three speeds, yet its label is “1 min read”, because a label cannot say zero. And 5,000 words takes 21 min 1 sec at the benchmark, yet its label is “22 min read”: that extra second pushes it into the next minute.

Exact times suit planning. Labels suit display. If you are working out whether you can finish something before a meeting, use the exact columns.

Which column to use

Pick the column by the text and the reason for reading it, not by how fast you would like to be.

Use the benchmark when you are estimating for a general adult audience reading ordinary non-fiction in English: news, blog posts, reports written in plain language. It is the only one of the three that is a published average.

Use the careful column when the material is technical, legal or academic, when it is full of unfamiliar terms, or when the reader needs to retain it. As a planning assumption it is also the safer choice when you do not know the audience, since under-estimating someone’s time costs more goodwill than over-estimating it.

Use the brisk column for light, familiar material: a chatty newsletter, a subject the reader already knows well. Brysbaert (2019) points out that 300 WPM, often quoted as the average, is higher than the evidence supports, so treat this column as the quick end of normal and not as a default.

The gap between the outer columns is large. At 2,000 words, careful reading takes 11 min 7 sec and brisk reading takes 6 min 40 sec. Neither is wrong; they describe different situations. When the difference matters, quote both ends as a range.

Two common cases are not in the table. Fiction averages 260 WPM in the same meta-analysis, a little faster than non-fiction, which the paper attributes largely to fiction using shorter words. And your own speed may sit anywhere in or outside the typical 175 to 300 WPM band. For either, type the speed into the reading time calculator as a custom value. If you do not know your speed, the reading speed test measures it in a few minutes, and average reading speed explains how to interpret the result against the research.

Word counts that are not in the table

The table has eleven rows, and real documents rarely land on one. There are three quick ways to fill the gaps without a calculator.

Add rows together

Reading time is proportional to length, so rows add. A 4,000-word essay is a 3,000-word row plus a 1,000-word row. At the benchmark that is 12 min 36 sec plus 4 min 12 sec, or 16 min 48 sec. For one length across many speeds, including reading aloud, see the 1,000-word reading time guide.

Use words per second

Two of the three speeds reduce to easy numbers. 180 WPM is exactly 3 words a second, and 300 WPM is exactly 5 words a second. So an 800-word piece takes 800 ÷ 3 = 267 seconds at the careful speed (4 min 27 sec) and 800 ÷ 5 = 160 seconds at the brisk speed (2 min 40 sec).

The benchmark is less tidy at just under 4 words a second. A workable shortcut is 25.2 seconds per 100 words: 800 words is 8 × 25.2 = about 202 seconds, or 3 min 22 sec.

Scale a row

Double the words, double the time. This is where a small caution applies. The table shows times rounded to the nearest second, and scaling a rounded figure scales the rounding too. Doubling the 3,000-word benchmark entry gives 25 min 12 sec for 6,000 words, while the direct calculation gives 25 min 13 sec. A one-second gap is harmless for planning, but it is the reason hand-typed reading-time tables on the web often disagree with each other in the last digit. For anything you intend to publish, calculate directly.

The underlying arithmetic is one division, set out step by step in how reading time is calculated.

Going the other way: from minutes to words

Sometimes the fixed quantity is time. You have a 20-minute train ride, a 15-minute slot before a call, or an editor who wants “a five-minute read”. The reverse formula is words = minutes × WPM.

How many words fit into a reading session
Time available180 WPM238 WPM300 WPM
5 min900 words1,190 words1,500 words
10 min1,800 words2,380 words3,000 words
15 min2,700 words3,570 words4,500 words
20 min3,600 words4,760 words6,000 words
30 min5,400 words7,140 words9,000 words
45 min8,100 words10,710 words13,500 words
60 min10,800 words14,280 words18,000 words

Read across a row to see how much the speed assumption matters. A 15-minute slot holds 2,700 words at the careful speed and 4,500 at the brisk speed, a difference of 1,800 words, which is a substantial article in its own right.

Two practical uses:

  • Choosing what to read. With 20 minutes and an ordinary pace, you can finish about 4,760 words. A 6,000-word feature will not fit; a 4,000-word one will, with a few minutes to spare.
  • Setting a length. If a piece must carry a “5 min read” label at the benchmark, it can be at most 1,190 words. Anything longer rounds up to six.

For times that are not in the table, multiply directly. Twelve minutes at 238 WPM is 2,856 words. The calculator’s reverse mode does the same thing for any speed.

What the numbers do not capture

The table assumes one reader moving through plain prose at a steady rate, start to finish. Several things break that assumption.

Reading is not one activity. Carver (1992) describes separate reading processes, which he calls gears, each with its own typical rate among the college students he studied: roughly 600 WPM for scanning, 450 for skimming, 300 for ordinary reading, 200 for reading to learn and 138 for memorising. These are his estimates for his population, not universal constants, but the pattern holds generally: someone studying a 5,000-word chapter and someone skimming it for one fact are not doing the same job, and no single column describes both.

Only prose is counted. A 2,000-word tutorial with six code samples and a 2,000-word opinion column have the same row in the table. The tutorial will take longer. Charts, tables, equations and images all cost time that a word count does not register.

Silent reading is assumed. Reading aloud is slower. Brysbaert (2019) puts the adult average for oral reading at 183 WPM, close to the careful column, so a script or speech needs a different estimate. The comparison is worked through in reading time versus speaking time.

Long texts are not read in one sitting. The 10,000-word row says 42 min 1 sec at the benchmark. That is continuous reading time. Add breaks, interruptions and the minute it takes to find your place again, and elapsed time is longer. For book-length work the question becomes one of sessions and days, which how long it takes to read a book covers.

The research is about English-speaking adults. The 238 WPM figure should not be carried over to other languages, younger readers or second-language readers without adjustment.

Treat every figure here as a planning estimate with a margin around it. The margin is smallest when the text is plain and the speed is one you have measured yourself.