Words
What counts as a good typing speed
Forty words a minute is average. What that is enough for is the more useful question.
The honest answer is that it depends what for, and the ranges below are more useful than a single figure.
The ranges
| Speed | What it looks like |
|---|---|
| Under 25 wpm | Hunt and peck, or a phone keyboard |
| 30–40 wpm | Average for an adult who types daily but never learned touch typing |
| 40–55 wpm | Competent touch typing; most people who type for work |
| 55–75 wpm | Comfortably fast. Typing stops being the bottleneck on writing |
| 75–100 wpm | Fast, and almost always someone who types professionally |
| Over 100 wpm | Rare, and always with deliberate practice behind it |
The threshold worth caring about is around 55 to 60. Below it, your hands limit how fast you can get words down. Above it, your thinking does — and typing faster stops making you more productive at writing, because composing was never the slow part.
Why test scores flatter you
A typing test number is usually higher than your real working speed, for three reasons.
Test text is easier than real text. Many tests use a shuffled list of the 200 commonest English words, with no capitals, no punctuation and no unfamiliar vocabulary. Scores on those run 10 to 20 per cent above scores on real prose. That is why our test uses complete sentences: the number is lower and it is closer to what you would get writing an email.
You are copying, not composing. Transcription and composition are different tasks. Nobody writes at their transcription speed, because thinking takes time your fingers spend idle.
A short test rewards a sprint. A one-minute burst is faster than a sustainable rate. Your consistency score is the better guide to what you can hold for an hour.
What employers actually ask for
Less than people fear, and usually with accuracy attached.
General administrative roles typically ask for 40 to 50 wpm. Data entry roles often want 60 or more, and usually specify accuracy above 95 or 98 per cent. Transcription and captioning are the genuinely demanding end, where 80 to 100 wpm with very high accuracy is the working requirement — and captioning frequently uses stenotype machines rather than a standard keyboard, which is a different skill entirely.
If a listing gives a number, take a test on real prose rather than a word list before deciding you fall short. The word-list figure you remember is probably not the one they mean.
Accuracy is the number that matters
An error costs far more than the keystroke that caused it: noticing it, reaching for backspace, deleting, retyping. Comfortably a second, which at 60 words per minute is a whole word.
So 70 wpm at 90 per cent accuracy is slower in practice than 60 wpm at 99 per cent. On our test only correct characters count towards the score, which makes the same point arithmetically — errors do not merely fail to help, they subtract.
The practical instruction is uncomfortable: slow down until your accuracy is consistently above 97 per cent, then let the speed come back. It usually returns within a fortnight and settles higher than before.
What consistency tells you
Consistency measures how steady your pace was, sampled once a second, and it is the number most tests do not report.
A decent average with low consistency almost never means general slowness. It means specific things are stalling you — usually letter pairs needing the same finger twice, reaches to the number row, or punctuation requiring a shift. That is a fixable, targeted problem rather than a general one, and you cannot see it in a WPM figure.
The number that predicts real output
None of the figures above predicts how quickly you finish a piece of writing, and it is worth saying why.
Writing is composition, and composition is bounded by deciding what to say. For anyone past about 55 words per minute, the fingers are idle a good deal of the time regardless — waiting for the next sentence to arrive rather than for the hands to catch up.
What does predict output is how little you stop. Typing that flows in long stretches produces more text per hour than faster typing interrupted every few words by a correction, a reread or a rephrase. That is why the consistency figure is more interesting than the headline speed for anyone whose job involves writing rather than transcribing.
If you want to write faster, the intervention is almost never typing practice. It is drafting without editing, and editing as a separate pass.
Why phone typing is a separate number
Thumb typing tops out far lower than a keyboard for nearly everyone, and the limiting factor is different.
On a physical keyboard your fingers know where the keys are without looking, because they can feel them. On glass there is no tactile confirmation, so you either watch the keyboard — which costs you the text — or accept a higher error rate and let autocorrect clean up. That is why autocorrect matters enormously on a phone and hardly at all on a keyboard.
Typical thumb-typing rates land around 25 to 40 words per minute for someone practised, against 40 to 70 on a keyboard for the same person. Swipe input suits some people substantially better and some not at all, and the only way to know is a fortnight of trying.
None of this shows up on a desktop typing test, and comparing the two numbers is not meaningful.
Testing yourself fairly
Four things make a typing test comparable with your own previous results, which is the only comparison worth making.
Same keyboard. A laptop’s short travel and a mechanical board’s long travel produce genuinely different numbers for the same person, often ten words per minute apart.
Same kind of text. A word-list score and a prose score are not the same measurement.
Long enough to be a rate. A fifteen-second test measures a sprint. A minute or more measures something you could sustain.
Not immediately after another test. Fatigue in the small muscles of the hand is real and shows up as falling accuracy within a few consecutive runs.
Where the plateau usually is
Most people improve quickly for a few weeks and then stop, generally somewhere between 45 and 60 words per minute, and generally not because they have reached a limit.
Three habits account for most plateaus, and all three are invisible from the inside because they have become automatic. Reaching with the wrong finger, which is comfortable and requires repositioning the hand afterwards. Pausing before punctuation, particularly anything needing a shift. And reading word by word rather than three or four words ahead, which caps your rate at your reading speed for single words.
Breaking any of them means typing deliberately slower for about a week while a new pattern forms. That is genuinely unpleasant, and it is the only route past the plateau — which is why most people stay on it.
Does speed matter?
For most jobs, less than people assume. Once you are past about 55 wpm, writing speed is governed by thinking, and the returns from further practice go into a task that is no longer the constraint.
Where it does matter is at the bottom of the range. Going from 25 to 45 wpm genuinely changes what work feels like, because at 25 the mechanics of typing occupy attention that should be going on what you are saying. That is the improvement worth the fortnight of deliberate accuracy work, and it is achievable for almost anyone.
Above that, the honest position is that typing faster is a pleasant thing to be able to do rather than a productivity intervention — which is fine, and worth saying plainly rather than dressing up.