Typing Speed Test

How it works

1. Sit the Way You Normally Type. A test taken hunched over a laptop on your knees measures that posture, not your typing. If you want a number you can compare to later attempts, take it at the desk and keyboard you actually use.
2. Pick a Length. Fifteen seconds is a snapshot and swings wildly between attempts. Sixty seconds is the length most standards and job requirements are written against. Thirty is the useful middle, and the default here.
3. Choose Words or Sentences. Words mode draws from a frequency list, so it measures raw finger speed. Sentences mode gives you connected text, where punctuation, rhythm, and reading ahead all start to count.
4. Start Typing — the Clock Starts With You. The timer begins on your first keystroke, not when the text appears, so time spent reading the prompt or finding the home row is never counted against your speed.
5. Let Mistakes Go. You can correct the word you are on, but not a word you already committed. Chasing an error you have passed costs more time than the error itself, both here and in real typing.
6. Run It More Than Once. A single result is noise. Three runs of the same length, on the same keyboard, in the same sitting is a measurement — and the spread between them tells you as much as the best one does.

What WPM Actually Counts (It Is Not Words)

Words per minute has almost nothing to do with words. The standard, which predates computers and comes from typewriter-era classification, defines one "word" as five characters including the space that follows it. That convention exists so that "word" means the same thing whether you are typing "cat" or "international" — otherwise a test made of short words would flatter you and one made of long words would punish you, and no two tests would be comparable.

• Net WPM — correct characters ÷ 5 ÷ minutes. This is the headline figure, and the one every serious test reports.
• Raw WPM — every character you typed ÷ 5 ÷ minutes, mistakes included. Always higher than net; the gap between the two is exactly what your errors cost you.
• Characters per minute — correct characters, no division by five. The same run, expressed in the unit most of the world outside the Latin alphabet uses.

Spaces count as characters, because you type them. A test that silently dropped them would report a number several percent lower than every other test on the internet, for no reason a visitor could see.

Accuracy and Consistency Measure Different Failures

Accuracy is the share of characters you typed that matched the text you were given. It is deliberately positional: if you drop a letter early in a word, everything after it is compared against the wrong position and counted wrong. That is harsher than an alignment-based score would be, and it is the honest way round — a dropped character really does produce a wrong word, and a scoring system that quietly re-synced would forgive the single most common mistake in typing.

Consistency is a different question: not how much you got right, but whether you held one pace. It is computed from the variation in your speed across each second of the run, mapped onto a 0-100 scale. Two people can finish with the same WPM, one by typing steadily and one by alternating bursts with pauses to think — and in sustained real work, the steady typist is the faster of the two.

Words Mode and Sentences Mode Measure Different Things

Words mode shuffles a list of common words in your language and shows them in a random order. Nothing connects one word to the next, so there is nothing to anticipate and nothing to misread — it isolates how fast your fingers move between keys you already know.

Sentences mode gives you whole sentences. Suddenly you can guess the next word from the last three, which usually makes people faster, and you have to handle real word lengths and real letter pairings rather than a shuffled bag. Most people score a little higher in sentences mode, and the difference between your two scores is a fair measure of how much you rely on reading ahead.

• Words mode is the better comparison against other online typing tests, most of which default to it.
• Sentences mode is the better predictor of how fast you write an email.
• Both draw from lists written for this site — no quoted books, no scraped passages, and the same seed always produces the same text, so you can retake an identical run.

Phone Typing Is Graded on Its Own Scale

If you open this on a phone or tablet, the test notices the touch pointer and switches to a separate set of speed bands. This is not a consolation prize: on-screen typing is a different motor task with a different population distribution, and grading it against desktop numbers would tell almost everyone on a phone that they are slow, which is useless information.

• A study of 168,000 volunteers typing on physical keyboards found an average of 52 words per minute.
• A study of 37,000 volunteers typing on mobile devices found an average of 36.2 words per minute, with 2.3% of errors left uncorrected.
• The gap has closed considerably over the last decade, largely because people who grew up with phones use two thumbs rather than one finger.

What Your Keyboard Contributes — and What It Does Not

Keyboards matter less to raw speed than the marketing suggests and more to accuracy than most people notice. The hardware failures that genuinely cost you words per minute are the ones that drop or duplicate keystrokes, and none of them are visible in a typing test — they look like your mistakes.

• Key rollover: a cheap membrane keyboard can miss a keystroke when three or four keys are held at once, which happens constantly at speed. Our keyboard test shows exactly how many simultaneous keys yours registers.
• Chatter: a worn switch that reports one press as two, adding doubled letters you never typed.
• Layout changes: a keyboard with a different Enter or left-Shift shape will cost you accuracy for days after a switch, regardless of its quality.
• Actuation force and travel: real, but a preference rather than a speed factor — the fastest recorded typists work on everything from laptop scissor switches to heavy mechanical boards.

If your accuracy here is far below what it feels like it should be, run the keyboard test before you blame your hands. A key that intermittently fails to register is indistinguishable from a typo in every typing test ever made.

Typing in Scripts That Use an Input Method

In Japanese, Chinese, Korean, and Hindi the characters on screen are not the keys you press. You type a reading and an input method converts it — romaji to kana to kanji, pinyin to hanzi, jamo to composed Hangul syllables. This test measures the text you actually produced, not the keystrokes on the way there, which is why it counts composed characters and waits for a composition to settle before judging anything.

• Japanese and Chinese commit each unit when the converted text reaches its length, because the space bar belongs to the input method there — it converts and selects candidates, and never reaches the page.
• Korean and Hindi type ordinary spaces between words, so they commit on the space like any Latin-script language.
• Character counts are measured in graphemes, not code units, so a Devanagari conjunct or a composed Hangul syllable counts as the one character a reader perceives.

How to Actually Get Faster

The unglamorous truth is that speed is a by-product of accuracy and of not looking down, and almost nobody gets faster by trying to type faster. Every correction costs a backspace, the retype, and the moment of attention that spots the error in the first place — call it four keystrokes for one wrong one.

1. Slow down until your accuracy sits above 97%, then hold that pace until it feels easy. Speed follows on its own.
2. Stop looking at the keyboard, even if it costs you half your speed for a week. There is no version of fast typing that involves finding keys with your eyes.
3. Return your fingers to the home row after every word rather than drifting; drift is what makes the same key land wrong twice in a row.
4. Practise the letter pairs you actually miss. Most people have three or four recurring transpositions, and fixing those beats general practice by a wide margin.
5. Check your typing on the keyboard you use most. Speed transfers between keyboards far less than people expect.

Frequently asked questions