Typing Speed Test (WPM), Explained
After reading this you will know exactly how words per minute is counted, why net WPM punishes sloppy speed, and how to read your accuracy, consistency and error stats to train the fastest way.
What the test measures and one quick example
A typing speed test shows you a passage, starts a clock on your first keystroke, and reports how fast you typed. The headline number is words per minute, but a "word" here is not an English word. It is a fixed unit of five characters, spaces included. That convention lets the test compare short words and long words fairly.
Here is the hook. Suppose you type the quick brown fox in 4 seconds. That string is 19 characters. Divide by 5 and you get 3.8 words. You typed them in 4 seconds, which is 4/60 of a minute. So your speed is 3.8 \div (4/60) = 57 WPM. The word "the" and the word "quick" pull the same weight per character, which is the whole point of the five-character rule.
This tool tracks that number live, deducts a penalty for mistakes you leave uncorrected, reports your accuracy, and draws a second-by-second speed curve so you can see whether you type in a steady rhythm or in bursts.
When this test is the right tool, and when it is not
Use a one-minute test to measure sustained speed on ordinary prose. Sixty seconds is long enough to smooth out a lucky opening burst and short enough that you can repeat it often. The 15 and 30 second options are useful for warm-ups or when you only want a rough reading.
Do not read too much into a single run. Typing speed varies from trial to trial by several WPM for reasons that have nothing to do with skill: an unfamiliar word, a sticky key, a moment of distraction. If you want a real estimate, take five runs and look at the range, not the best single score.
This test measures typing on the exact passage shown. It does not measure your speed writing your own thoughts, which is usually slower because you are composing at the same time. Treat the number as a mechanical ceiling, not your everyday drafting rate.
The formulas, and the intuition behind them
Gross WPM ignores errors. It counts every character you typed, including wrong ones, divides by five, and scales to a minute.
Here C is the total number of characters you entered and T is the elapsed time in seconds. The C/5 converts characters to five-character words, and dividing by T/60 scales the count up to one minute.
Net WPM subtracts a full word for each error you leave uncorrected when time runs out.
Now E is the number of uncorrected errors. Each one costs you one whole word (five characters' worth of credit), which is a steep price. Because this tool keeps mistyped characters highlighted until you fix them with backspace, most errors get corrected during the run and do not count as E. The penalty mainly bites when you race ahead and leave wrong letters behind.
Accuracy is a separate figure: the share of keystrokes that were correct the first time.
Here K_\text{total} counts every key you pressed including corrections, and K_\text{correct} counts the ones that matched the target on the first try. A run can have 99% accuracy and still show a net WPM near its gross value, because you fixed the few errors before the clock stopped.
A worked example you can reproduce
One clean minute at a steady pace
Run the 60-second test with the default passage. Suppose when the clock stops you have typed 312 correct characters, made 6 mistakes that you caught and fixed with backspace, and left 2 wrong characters standing at the end.
- Characters entered C = 312, time T = 60 seconds.
- Gross WPM = (312/5) \div (60/60) = 62.4 \div 1 = 62.4, which the tool rounds to
62. - Uncorrected errors E = 2, so net WPM = (62.4 - 2) \div 1 = 60.4, rounded to
60. - Keystrokes total 312 + 6 + 2 = 320. Correct-on-first-try keystrokes = 320 - 8 = 312. Accuracy = 312/320 = 97.5\%.
So this run reads 60 net, 62 gross, 97.5% accuracy. The 2-point gap between gross and net is the cost of the two errors you left in place. Fix those before the buzzer and the two numbers would match.
Reading the consistency score and the speed curve
The tool samples your speed once per second and builds a curve. Consistency compares those samples: a flat curve scores high, a jagged one scores low, even at the same average. One common way to express this is the coefficient of variation, the standard deviation of your per-second speed divided by the mean.
Here \bar{x} is your average per-second speed and \sigma is its standard deviation. A typist who holds 60 WPM every second has \sigma = 0 and scores 100%. A typist who alternates bursts of 90 and stalls of 30 has the same average of 60 but a large \sigma, so the score drops well below 100.
The chart below shows two runs with the same 60 WPM average. The steady one scores near 96%. The bursty one, with the same mean, scores near 70%.
Rhythm is the thing typing courses actually build. If your consistency sits below 80%, slow your fast stretches down until the curve flattens. A steady 55 WPM usually converts to a higher clean speed than a jerky 70.
What your error stats tell you to train
The results screen lists the characters you miss most. This is the single most useful part of the report, because errors cluster. Most people do not miss letters at random. They miss the same three or four keys over and over: often a and s confusion, the reach to p, or the shift for capitals and punctuation.
Fixing a top error is cheap. If t accounts for a third of your mistakes and you drill just that reach for a few minutes, your accuracy can climb a full point without any change to your speed. That single point often matters more than raw WPM for real work, because every uncorrected error either costs you a net-WPM word or costs you the time to backspace and retype.
Explore the speed-accuracy tradeoff
The most instructive thing about this test is not the top speed. It is how errors erase speed. Push the error slider and watch net WPM fall while gross WPM stays put.
Notice the test-length effect. On a 60-second run each uncorrected error removes 1 net WPM. On a 15-second run the same error removes 4, because the deduction is scaled up to a full minute along with everything else. Short tests reward clean typing even more sharply than long ones.
Common mistakes that distort your score
The first mistake is chasing gross WPM and ignoring the net gap. A run that reads 80 gross and 68 net is not an 80 WPM run. The 12-point gap means you left roughly 12 uncorrected errors on a 60-second test, and that error rate would ruin any real document.
The second is racing on a 15-second test and treating the number as your true speed. Short tests favor a strong opening burst that you could not hold for a minute. Compare like with like: 15 against 15, 60 against 60.
The third is testing tired or cold. Your first run of a session is usually your slowest by 5 to 10 WPM. Do a throwaway warm-up run, then measure. Because your best net WPM is saved on this device, one lucky warm-up score can sit at the top for weeks and mislead you.
Related tools on this site
Typing links speed, timing and steady rhythm, so several other tests here probe the same underlying abilities. If your consistency score interests you, the Tempo & Rhythm Tapping Test measures timing jitter directly. For raw motor speed without the keyboard, try the Reaction Time Test. To pair typing drills with number drills on the same keyboard, the Mental Math Trainer runs timed arithmetic with a per-operation breakdown. And if you enjoy training a specific mental skill to automatic speed, the N-Back Working Memory Test and the Doomsday Weekday Trainer use the same drill-and-measure loop.
Frequently asked questions
Why is a word counted as five characters instead of a real word?
Because real words vary in length from 1 to 15 letters, counting them directly would reward passages full of short words. The five-character convention (characters divided by 5) makes every keystroke count equally, so scores compare across different texts. It is the standard used by nearly every typing test.
Why is my net WPM lower than my gross WPM?
Net WPM subtracts one word for every uncorrected error left when time runs out. If you finished with 3 wrong characters standing on a 60-second test, your net score is 3 lower than your gross score. Fix mistakes with backspace before the clock stops and the two numbers converge.
What is a good typing speed?
About 40 WPM is the adult average. A fluent touch typist runs 60 to 70. Above 100 is competitive territory. For everyday work, accuracy above roughly 97% matters more than raw speed, because errors cost you backspacing time on every document.
Does pasting the text work?
No. Pasting is blocked so the number reflects real typing. The timer also starts on your first keystroke, not when the passage appears, so reading time does not count against you.
How do I raise my score fastest?
Train your top two or three error keys and flatten your speed curve. A single point of accuracy and a steadier rhythm usually add more clean net WPM than trying to type faster, because both cut the errors that trigger the per-word penalty.
Related tools
Color Discrimination Test
Spot the one tile with a slightly different color — the difference shrinks every time you get it right.
Contrast Sensitivity Test
How faint a pattern can you still see? A staircase test that estimates your contrast threshold.
Doomsday Weekday Trainer
Learn to name the weekday of any date in your head with the Doomsday algorithm — then drill it.
Hearing Frequency Test
Find the highest pitch you can hear and sketch your hearing profile across octave bands.
Mental Math Trainer
Timed arithmetic drills with streaks, accuracy and a per-operation breakdown — keyboard or keypad.
N-Back Working Memory Test
The dual N-back task from working-memory research: match positions and letters N steps back.