How fast do you actually type?
Typing test
The passage appears here โ it is stored inside this page, so the test works with the network switched off.
Typing speed
0WPM
One word means five characters including the space โ that is the standard every typing test uses, not real words, so a long word costs you nothing.
- Accuracyโ
- Charactersโ
- Mistakesโ
- Time leftโ
One gear check a month
MonthlyOne thing a month worth measuring before it fails properly โ a mic that reset itself, a stick that drifts, a switch sending two clicks. Two sentences and the test.
Try also
How is typing speed measured?
The typing test reports words per minute the standard way: every five characters you type counts as one word, spaces included, divided by the minutes spent typing them. Type “the quick brown fox” correctly and that is 20 characters, so four words, not five. Counting real words instead would make a passage of short words look faster than the same effort spent on long ones.
Accuracy is scored separately, character by character against the target text, and the two numbers only mean something read together. A burst at 80 WPM and 88% accuracy is slower in practice than a steady 65 WPM at 99%, because every wrong character costs a backspace and a retype.
The clock starts on your first keystroke rather than when the page loads, so reading the passage through before you begin costs you nothing at all.
Words per minute is a character count wearing a friendlier name, so it converts cleanly into the figures that actually describe the work. The right-hand column is the honest part: the bands are wide and the boundaries are soft.
| WPM | Characters per minute | Characters per second | Typical of |
|---|---|---|---|
| 20 | 100 | 1.7 | hunt and peck, eyes on the keys |
| 40 | 200 | 3.3 | most people who type every day |
| 60 | 300 | 5.0 | touch typing without looking down |
| 75 | 375 | 6.3 | transcription and data entry work |
| 100 | 500 | 8.3 | roughly the top few per cent |
| 120 | 600 | 10.0 | competitive typing |
One more figure worth keeping: at 60 WPM a single keystroke takes 200 milliseconds, so a backspace and a retype cost 0.4 seconds. That is the entire price of fixing an error as you go.
If a letter is not appearing at all, the keyboard tester will show whether the key is reaching the browser before you blame your hands. For the other half of the desk, the mouse tester checks buttons and the wheel, and the polling rate test measures how often your mouse reports back.
Frequently asked questions
Why is my score here different from the test I took last week?
Four things move it, and none of them is your typing. Whether errors are subtracted from the total, whether the clock starts at page load or at the first keystroke, how much punctuation and capitalisation the passage contains, and how long the run is. A 15-second run flatters almost everyone, because nobody loses rhythm in 15 seconds. Compare scores from the same test and the same duration or do not compare them.
What counts as a word if I type numbers or symbols?
Five characters, exactly as with letters. A ten-digit phone number is two words, and a password full of symbols is scored the same way. Symbol-heavy text does score lower in practice, because every capital and most symbols need a modifier held down at the same time, and that is two keys where prose needs one.
Is 60 words per minute good?
It is comfortably above average. Most people who use a computer daily land near 40 WPM, 60 is where touch typing without looking down usually settles, and 75 is the floor for transcription and data entry work. Past 100 you are in the top few per cent. Accuracy matters more than any of it: 60 WPM clean beats 80 WPM that needs a proofreader.
Should I correct mistakes or keep going?
Correct them here, because accuracy is measured against the finished text. A backspace and a retype cost two keystrokes, which is under half a second at 60 WPM, while an uncorrected error stays in the accuracy figure for the whole run. In real work the same maths applies with interest, since a typo found later costs a hunt as well as a fix.
Does this work on a phone keyboard?
It runs, but the number is not comparable to a desktop one. Autocorrect and swipe input insert whole words in a single event, so the character count arrives faster than any thumb could produce it, and predictive suggestions do part of the typing for you. Treat a phone score as a measure of your phone keyboard, not of your typing.
Why does my keyboard layout change my result?
Because punctuation moves and some characters take two presses. On AZERTY the full stop needs a shift, on QWERTZ the Y and Z are swapped, and layouts with dead keys produce an accented character only after a second keypress. The letters cost the same everywhere; the passage’s punctuation is what makes one layout slower than another.
How the test works
Worked example
With characters typed in 12 seconds 43, characters wrong 2, this page works out 41 WPM net. A word is five characters including the space, so the count is the same whatever you type. Net speed drops the wrong characters; gross keeps them, which is why gross is the bigger number everywhere.
- Net speed41 WPM
- Gross speed43 WPM
- Accuracy95.3 %
Sources:
Where typing speed actually comes from
Rhythm beats bursts. A steady stream at 60 WPM outruns a run of sprints and stalls averaging the same figure, because every stall breaks the sequence your hands had memorised and the next few words are typed from scratch.
The slow part is almost never the letters. Numbers, symbols and capitals need a modifier held while another finger moves, and most people who type prose at 70 WPM drop below 40 on an address or a serial number.
Speed gains come from the keys you look at. If your eyes leave the screen for the number row or for punctuation, that is where the time is going โ not in the letters you already touch-type.