WPM, CPM, Accuracy and Consistency: Typing Scores Explained

Every typing test hands you a set of numbers and leaves you to guess how they were produced, and if an employer is going to score you on one it is worth knowing them — dispatcher assessments report net WPM and nothing else. Here are the exact formulas: net WPM is correct characters divided by five, divided by minutes. Raw WPM is the same arithmetic applied to every character you typed. CPM is correct characters per minute. Accuracy is correct keystrokes as a share of total keystrokes. Consistency is the variation in your per-second speed, inverted onto a 0 to 100 scale. Where a job converts these numbers into a response-time target, the arithmetic gets unusually concrete. And where the number is produced in front of an assessor, expect it to understate you.

Those five definitions are the whole answer, and the rest of this page is what they imply — why CPM carries no information beyond a change of units, why your accuracy can sit at 96 percent on a passage that looks flawless, and why the gap between two of these numbers tells you more than any of them alone. Once you can see the arithmetic, the score stops being a verdict and starts being a diagnosis. The same care is needed when comparing WPM to metrics from outside typing entirely, such as gaming actions per minute.

Try the formulas on your own numbers

The calculator below runs the four speed and accuracy formulas live. It is loaded with the worked example used throughout this page — 340 characters typed in sixty seconds, 300 of them correct, from 352 keystrokes of which 14 were wrong. Change any figure and every score updates.

Consistency is the one metric missing from it, and the reason is worth stating because it explains what the number actually is. The four figures above can each be derived from a single total — how many characters, how many keys, how long. Consistency cannot. It is computed from the whole series of per-second speeds recorded across a run, so there is no summary figure you could type into a box to produce it. Two typists who finish on identical totals can have completely different consistency scores, and that difference is precisely the information the metric carries.

The arithmetic itself is straightforward once you have the series. Take the standard deviation of your per-second speeds, divide it by their mean, and you have the coefficient of variation — a measure of how spread out your pace was relative to how fast you were going. Invert that onto a 0 to 100 scale and you have consistency. A typist averaging 60 WPM with readings clustered between 58 and 62 scores around 98; the same 60 WPM average swinging between 20 and 95 scores under 50. The section below works through it with a real series.

What does WPM mean, and how is it calculated?

WPM means words per minute, and the first thing to understand is that a word is not a word. In every standard typing test, one word is defined as five characters including spaces. Rather than counting actual words, which vary from two letters to fifteen, the test counts characters and divides by five. That convention is also why a stenographer's 300 words per minute is not the same measurement as yours — why 300 WPM is not really typing sets out what is being counted instead.

So the calculation is: count the characters that matched the target text, divide by five to convert them into words, then divide by the minutes elapsed. Typing 300 correct characters in one minute gives 60 words per minute — 300 divided by 5 is 60 words, divided by one minute is 60 WPM. Over a two-minute run, 300 correct characters would give 30 WPM instead.

One detail matters more than it looks. A character that never matched the target is worth nothing, but it does not subtract either. That matters when a new method temporarily raises your error rate, as it does when learning to type with one hand. It simply never contributes to the total. That distinction separates net WPM from the adjusted scoring many employers use, where each uncorrected error is deducted as a whole word — a stricter method covered in typing speed for jobs.

Why one word is five characters

The five-character word is inherited from typewriter-era testing, and it survives because it solves a real problem. If tests counted actual words, a passage of short common words would produce a far higher score than a passage of long technical ones for exactly the same amount of physical typing. The five-character convention removes vocabulary from the measurement.

The trade-off is that WPM measures keystrokes rather than meaning. A test full of three-letter words and a test full of ten-letter words should score you roughly the same, which is what makes scores comparable between sites and between sessions. It is also why a WPM figure is only meaningful alongside the length of the test it came from — a point the section on score variation returns to.

Net WPM vs raw WPM: the most useful gap on the screen

Net and raw WPM use the same formula and differ only in what they count. Net counts characters that matched; raw counts everything you typed. Because raw includes your mistakes, it is always the higher of the two, and it is the number that matches your subjective sense of how fast your hands were moving.

That gap is the single most diagnostic figure on a results screen, because it separates two problems that look identical from the outside. A narrow gap means your hands are moving at close to their honest speed and the constraint is genuine hand speed. A wide gap means your hands are already fast enough and errors are consuming the difference — which is much easier to fix. The habits that produce a wide gap are covered in common typing mistakes.

What is CPM, and is it different from WPM?

CPM means characters per minute, and it is often assumed to be a separate, more granular measurement. It is not. CPM is the count of correct characters divided by the minutes elapsed — the same correct-character total behind net WPM, without the division by five.

That makes CPM exactly five times your net WPM. A 60 WPM run is a 300 CPM run by definition. Knowing this is worth something practical: if a test reports both figures, the second one carries no information the first did not already give you, and there is no point tracking both over time.

One caveat, because it looks like a contradiction on screen. That five-to-one relationship holds on the underlying values, but each figure is rounded to a whole number independently — so a run of 302 correct characters in a minute reports 60 WPM and 302 CPM, not 300. The measurement is still identical; only the rounding differs. Most of the time the two numbers on your results screen will not be exact multiples, and nothing is wrong when they are not.

Keystrokes per minute is the same idea again under a third name. Data entry roles more often quote keystrokes per hour, which is simply KPM multiplied by sixty — divide KPH by 60 and then by 5 to reach WPM, so 10,000 KPH is roughly 33 WPM.

How typing accuracy is calculated

Accuracy is correct keystrokes as a percentage of every key you pressed. Press 300 keys, get 12 of them wrong, and your accuracy is 96 percent, because 288 of 300 were correct. The formula is simple; what surprises people is the thing being measured.

Accuracy is recorded on keystrokes as they happen, not on the passage you end up with. Every key you press is counted once, and any key that did not match what was expected at that moment is counted as an error permanently. The count is made in real time and never revised.

Why your accuracy is not 100 percent on a perfect passage

This follows directly from the previous section and it is the most common source of confusion about typing scores. Backspacing over a mistake repairs the text but does not repair the record. The backspace keystroke itself is not scored as an error, but it cannot un-count the error that prompted it. So you can finish a test with a visibly flawless passage and still score 96 percent.

That design is deliberate rather than an oversight. Accuracy is meant to measure how cleanly you type, not how diligently you proofread. A typist who hits the right key first time is doing something genuinely different from one who types twice as many keys and corrects the difference, and a metric that scored both at 100 percent would be hiding the distinction that matters most for improvement.

It also explains a pattern worth recognizing: if your accuracy is low while the finished text looks clean, you are not a careless typist so much as a fast one working above the pace your hands have actually learned. Slowing down by ten words per minute usually recovers most of it. Accuracy is also the more consequential of the two metrics for most real work, for reasons that go beyond the score.

What is typing consistency, and how is it scored?

Consistency is the least standardized metric in typing, so here is precisely how it works. Your speed is sampled once per second for the whole test, producing a series of per-second WPM values. The chart below is what such a series looks like for a typist averaging 60 WPM.

From that series we take the standard deviation and divide it by the mean. That ratio is called the coefficient of variation, and it expresses how spread out your speed was relative to how fast you were going. Consistency is that value inverted and turned into a percentage: 100 minus the variation, clamped to a 0 to 100 range.

Dividing by the mean is the part that matters, because it makes the score fair across ability levels. Swinging by ten WPM around an average of forty is a much bigger wobble than the same ten WPM around an average of a hundred, and the coefficient of variation reflects that where a raw standard deviation would not. A metronomic typist approaches 100; one who fires off bursts between pauses can score below 60 even with a strong WPM.

The practical value of consistency is that it moves before speed does. Because it measures automaticity rather than peak output, it climbs while your WPM is still flat — which makes it the earliest honest evidence that practice is working, and the reason a great many people quit at exactly the point where they were about to improve.

The four numbers side by side

Each metric answers a different question, and each is silent on the others. Reading them as four views of one run rather than four grades is what turns a results screen into something actionable.

Reading the numbers as one diagnosis

The combinations are more informative than any single figure, and there are only a handful worth memorizing. Each points to a different constraint and a different fix.

The pattern worth internalizing is that speed is an output, not an input. Accuracy and consistency are the two variables you can actually control on the next test, and WPM is what happens when both improve. The sessions and drills that move them are in how to practice for a higher WPM.

Why the same typist scores differently between tests

None of these formulas contain a term for how difficult the passage was, which means identical arithmetic can produce quite different numbers for the same typist on the same day. Test length is the largest factor: a thirty-second run can be carried by a lucky passage and a burst of familiar words, and it ends before concentration slips or an unfamiliar sequence appears.

Text difficulty is next. Common-word prose scores well above technical vocabulary or heavily punctuated text, because the five-character convention normalizes word length but not familiarity. Then warm-up, keyboard, and fatigue each move the figure by a few WPM. For that reason a single score is a snapshot rather than a measurement — where your number really sits is set out in average typing speed, what makes two sites disagree about the same typing covers the scoring differences, and why you type faster some days covers the biological ones.

What these numbers do not measure

It is worth being clear about the limits of the whole set, because a typing score is narrower than it appears. All five metrics describe transcription: copying visible text at a keyboard, with no thinking required about what to write.

• Composition speed. Typing your own sentences is bounded by how fast you think, and most people compose far below their transcription speed — the same distinction decides whether typed or handwritten notes retain more.
• Numbers and symbols. Alphanumeric data is typed considerably slower than prose by almost everyone, which is why data entry roles test it separately.
• Code. Bracket and punctuation density is far higher than prose, and the weak fingers carry most of it — how far the symbol density goes measures it at roughly one character in six.
• Sustained work. A three-minute run says little about how you type in hour four of a working day.
• Editing. Real writing involves moving around a document, selecting and rewriting — none of which a linear passage test touches.

How other typing tests score things differently

The formulas above are the standard ones, but implementations differ in ways that make cross-site comparison unreliable. If your score moves sharply when you switch tests, the method probably changed rather than your typing — why your WPM is different on every typing test site works through each of those differences and how the major sites compare.

• Some tests report gross WPM rather than net, counting every character typed. That figure is always higher and is not comparable to a net score.
• Employer assessments commonly deduct each uncorrected error as a full word, which is stricter than simply not counting it.
• Some tests lock backspace entirely, so an early mistake stays in the text and affects the rest of the run.
• Accuracy is sometimes computed against the finished passage rather than on keystrokes, which produces a much more forgiving number.
• Consistency has no agreed definition at all. Where one site samples per second and inverts the coefficient of variation, another may simply report your slowest and fastest seconds.

Where to go from here

The formulas are only useful applied to your own numbers. Take a timed test at sixty seconds or longer, then read all five figures rather than the headline one — the raw-to-net gap and the consistency score are where the actionable information sits.

From there the path depends on what the numbers say. A wide raw-to-net gap points to common typing mistakes; a low score with visible glances at the keyboard points to how to learn touch typing; clean numbers that are simply short point to drills and a practice routine. The typing tutorial explains where each figure appears on the results screen, and the leaderboard shows what the same score looks like against everyone ranked at that duration.