What's the Average Reading Speed? (Research from 190 Studies)

238 words per minute. That's the average for adults reading non-fiction, pulled from Marc Brysbaert's 2019 meta-analysis of 190 separate studies. For fiction the number rises to 260, because narrative carries you forward. For technical material it falls below 150, because vocabulary slows everything down.

What most people quote — 300, 400, even 500 WPM — comes from studies with no comprehension check, self-reported speeds (everyone overestimates), or marketing pulled out of the air. Brysbaert's number is the boring, peer-reviewed one. It's also the one to anchor on.

We run a free speed test on this site, and across the first thousand or so people who took it, the median came out at 247 WPM with 78% comprehension. That tracks Brysbaert almost exactly. People are surprised when they see it. They shouldn't be — they're just reading at the speed the language was designed to be read at.

The 238 problem

Here's the awkward part: "238 WPM" is correct, but not in the way that helps you. It's an average across native English speakers reading averagely difficult prose with reasonable comprehension. The minute you change any of those three variables, the number moves.

A few worked examples:

A neuroscience PhD reading their own field's papers will register somewhere around 180 WPM on first pass. Same person reading a thriller? 320. Same person reading a contract their lawyer drafted? 95. These aren't different readers — they're the same brain meeting text that demands different things from it.

This is also why "improve your reading speed" is a partly meaningless goal in the abstract. You can absolutely speed up at the kind of material you read every day. You can't really speed up at material you don't yet know how to read, because the bottleneck isn't your eyes — it's that your brain hasn't built the vocabulary scaffold yet.

Find out where you actually fall — 90-second test →

Why the number is lower than you'd guess

Three things hold most readers below 300 WPM, and we know this because Keith Rayner and his collaborators spent thirty years tracking eyes through reading sessions with high-speed cameras.

Subvocalization. Most adults silently pronounce each word as they read. Slowiaczek and Clifton's 1980 EMG study put electrodes on people's lips during silent reading and found measurable motor activity even when subjects swore they weren't moving anything. The voice is real. It caps your speed at roughly speaking rate — about 180 WPM. Train past it and the ceiling lifts; we cover the how-to in our techniques piece.

Regression. Adult readers re-read about 10–15% of words. Sometimes this helps. Mostly it's anxious habit — a feeling that you missed something you actually got the first time. A simple finger pacer eliminates most regression and is, hands down, the single most reliable reading-speed intervention with research support.

One-word fixation. Slower readers fixate on individual words. Faster readers fixate on groups of 2 to 4. Your peripheral vision picks up the surrounding words and your brain assembles the chunk as a unit — what reading researchers call the perceptual span. Trained readers have spans of 14 to 15 letters; untrained adults have spans closer to 9. The 50% difference shows up directly in WPM.

Those three together account for almost all the variation between "slow" and "fast" readers among adults. Genetics and intelligence are weak predictors. Practice and technique are strong ones.

Where you probably fall

Roughly 90% of adult native English speakers read between 180 and 320 WPM at comfortable comprehension. Some orientation:

• Under 180 WPM. Below average. Almost always reducible — usually heavy subvocalization plus regression. Six weeks of structured practice typically lifts this group by 60–80 WPM.
• 180 to 260 WPM. The middle. Comprehension is usually strong because the brain has time to process. The biggest unlock here is reducing subvocalization without losing accuracy.
• 260 to 360 WPM. Above average. You read a lot; your eyes have learned to chunk. Gains come from peripheral-vision drills and pacer-led overspeed practice.
• 360 to 500 WPM. Strong reader, already top 10%. You're closer to the biological ceiling than most. Further gains exist but require disciplined daily training to lock in.
• Above 500 WPM. You're skimming. Rayner's eye-tracking work is clear here: humans can't fixate fast enough to actually read above 500 WPM with full comprehension. The ceiling is in the visual system itself, not in any technique. The 1000-WPM claims are skimming dressed up in a costume, and we've ranted about them at length in our science piece.

If your number landed somewhere unexpected on the test, the first instinct is usually to question the test. Resist. Take it again tomorrow morning when you're fresh. Reading speed varies by 10–15% within a single person across a day; one measurement is noise, two within 20 WPM of each other is signal.

"But I read so much faster than that"

This is the most common question we get, and the answer is mildly uncomfortable. Subjective reading speed is, on average, about 30% higher than measured reading speed. You feel like you're reading at 350. The clock and the comprehension test say 260.

This isn't a personal failing — it's a quirk of how the brain monitors its own activity. The metacognitive estimate of speed is correlated with how easy something feels, not how fast you actually processed it. Familiar topics feel fast. The number is the same.

If you don't believe the test, try this: pick an article you've never read, time yourself for 60 seconds, count the words you got through, and immediately write down three things you can remember. Then check the article for accuracy. Almost everyone who does this for the first time comes away calibrated.

Take the test — it'll be lower than you expect, and that's fine →

So can you actually improve?

Yes, with a caveat I'm going to repeat in every article on this site because it's the single most important point: reading speed only counts if comprehension comes with it.

Realistic gains for daily 5–10 minute training over six weeks:

• 30 to 60% improvement in sustained reading speed
• At the same comprehension level you started with
• On material similar to what you trained on

That's the well-supported window. 60% sounds like marketing but isn't — it's the median outcome in actual intervention studies. A reader going from 240 to 380 WPM is the same brain, faster eyes, better-pruned habits. A 200-page book takes 4 hours instead of 6. Across a year of reading, that's a 30-book lift to a 50-book lift, which compounds for the rest of your life.

What you can't do is go from 240 to 1000 WPM in five minutes. Anyone selling that is selling certainty, not capability.

The number isn't the point

Knowing you read at 247 WPM doesn't make you read faster. The number is the baseline you train against. Without it, "I think I'm getting faster" is just a feeling — and we already established that feelings about reading speed are systematically wrong.

If you've never measured, do that this week. If you have a number, look at it honestly. If you've been training with no comprehension check, you've been training the wrong thing. We get into that next in reading comprehension and speed.

Brysbaert's average is interesting trivia. Yours is the only one that matters.