How WPM and accuracy are measured
Typing speed is conventionally reported in words per minute, where a "word" is standardised as five characters including spaces rather than an actual word. This convention exists because real words vary in length — counting them directly would make a test full of short words score higher than an identical performance on longer ones.
So the raw calculation is:
Gross WPM = (total characters typed ÷ 5) ÷ minutes elapsed
Gross WPM ignores mistakes, which makes it close to useless on its own — you can inflate it arbitrarily by typing nonsense quickly. The figure worth attending to is net WPM, which subtracts errors:
Net WPM = Gross WPM − (uncorrected errors ÷ minutes elapsed)
Accuracy is the percentage of keystrokes that were correct. The two numbers have to be read together: 90 WPM at 85% accuracy is worse in practice than 65 WPM at 99%, because every error costs you the time to notice it, delete it and retype — typically several seconds, which is far more than the keystroke saved by going faster.
What counts as a decent speed
| Net WPM | Level | What it feels like |
|---|---|---|
| Under 25 | Hunt and peck | Looking at the keyboard, one or two fingers |
| 25–40 | Developing | Some touch typing, frequent glances down |
| 40–55 | Average adult | Comfortable for everyday work |
| 55–75 | Proficient | Typing no longer limits your thinking |
| 75–95 | Fast | Typical of people who write for a living |
| 95+ | Very fast | Sustained accuracy at this speed is uncommon |
The threshold that actually matters is around 40 to 50 WPM at high accuracy. Below that, typing is a bottleneck on thinking — you lose the end of a sentence while producing the beginning of it. Above it, further gains are convenience rather than capability. If you are a student writing essays, getting from 30 to 50 is genuinely worth the practice; getting from 70 to 90 mostly is not.
Accuracy first, always
If you are practising deliberately, hold accuracy above 97% and let speed follow. Speed practice at low accuracy trains the errors, because you are rehearsing incorrect finger patterns until they become automatic. Slowing down until you are accurate, then gradually increasing pace, is slower for a week and much faster after a month.
Practising in a way that improves things
Most people plateau because they practise what they can already do. A few things that move the number:
- Keep your eyes off the keyboard. This is the single largest determinant of ceiling speed. Looking down caps you regardless of practice volume, because it prevents the finger-position memory that touch typing depends on. It will feel worse for several days.
- Use home-row anchoring. Index fingers on F and J, fingers returning to the home row between keystrokes. Most keyboards have tactile ridges on those two keys for this purpose.
- Identify your specific weak keys. Errors are not evenly distributed. For most people they concentrate on the same handful — commonly the numbers row, punctuation, and letters reached by the weaker fingers. Drilling those specifically beats more general practice.
- Practise real text, not just word lists. Punctuation, capitalisation and shift-key transitions are where speed collapses in actual writing, and pure word drills never exercise them.
- Short, frequent sessions. Ten minutes daily beats an hour weekly. Motor learning consolidates with sleep, so distributed practice is more effective for the same total time — the same principle that applies to academic revision.
- Don't chase a personal best every session. Testing is not training. Alternate: mostly practise at a comfortable accurate pace, and test occasionally to measure.
Why this matters for exams and coursework
An increasing number of examinations are typed rather than handwritten, and several examination authorities now offer or require on-screen assessment for some subjects. Where that applies, typing speed becomes a direct constraint on how much you can produce in a timed paper — a student typing at 30 WPM in a 90-minute essay paper is at a real disadvantage against one typing at 60.
It is worth checking what applies to you. Our exam schedule pages link to the responsible authority for each country, and their guidance will state whether papers are handwritten, typed, or a mix.
Common questions
Is my typing recorded or sent anywhere?
No. The test runs entirely in your browser. Your keystrokes are processed on the page to calculate speed and accuracy, and nothing is transmitted to us. Results are not stored on any server.
Do corrected mistakes count against me?
Corrections cost you time, which lowers your WPM, but net WPM penalises uncorrected errors specifically. This reflects real writing: a mistake you fixed is a slower sentence, while a mistake you left is a defect in the output.
Why is my score lower here than on another test?
Tests differ in text difficulty, whether they include punctuation and capitals, duration, and how strictly they penalise errors. A test on common short words with no punctuation will produce a noticeably higher number than one on real prose. Compare your scores on the same test over time rather than across different tests.
Does keyboard layout matter?
Familiarity matters far more than layout. Alternative layouts such as Dvorak and Colemak have theoretical advantages, but the evidence that they raise ceiling speed for an already-proficient QWERTY typist is weak, and switching costs weeks of reduced productivity. Physical keyboard quality and key travel affect comfort more than speed.
Can I test on a phone or tablet?
The test will run, but the numbers are not comparable to a physical keyboard, and the WPM conventions above were developed for one. Thumb typing tops out much lower and the accuracy penalty behaves differently.