How Typing Scores Work

Official scoring formulas

A worked example, end to end

Formulas are easier to trust once you have seen one run through. Take a two-minute test where you typed 480 printable characters and got 12 of them wrong and left them uncorrected, so 468 were right. Every figure on your result comes from those three numbers. Leaving the errors uncorrected keeps the example simple: correct one and the accuracy numerator stops matching the net WPM numerator, for the reason set out further down.

Notice how little separates 48 from 46.8. Twelve wrong characters in two minutes cost barely more than one word per minute, which is why accuracy rarely feels expensive while you are typing. It is the accuracy figure that moves visibly — 97.5% is already below the 98% many employers set as a floor, on a run that felt clean.

Typing speed chart: where do you stand?

Typing speed requirements by job career

Gross WPM vs. Net WPM: what is the difference?

Gross (Raw) WPM calculates the sheer mechanical speed of your fingers. It counts every keystroke typed in five-character word units, regardless of whether words were typed correctly or misspelled.

Net WPM is the universal standard used by hiring managers, government civil service exams, and typing certifications. It measures usable, error-free work by dividing only correctly typed characters by elapsed time.

Why a “word” is five characters

Every WPM figure on every typing site divides by five, and the five is a convention rather than a measurement. It dates from typewriter-era testing, where averaging English text including its spaces produced roughly five characters per word, and it stuck because it makes results comparable when the material is not.

It does mean the unit is not an English word. Typing hippopotamus scores as 2.4 words; typing a cat sat scores as 1.8 for three real ones. That is the point: without a fixed unit, a test on short words would flatter you against a test on long ones, and no two results could be compared. The space bar counts as a character too, which is why it is a keystroke worth practising.

Why a mistake you corrected still counts

This surprises people, so it is worth stating the reasoning rather than just the rule. Your accuracy figure counts every incorrect key you pressed, including ones you immediately backspaced away.

If corrections erased mistakes from the record, the fastest strategy on any test would be to type carelessly and repair afterwards, and the score would measure nothing but your willingness to do that. Accuracy and net WPM answer two different questions: accuracy asks how often your fingers hit the right key, and net WPM asks how much correct text you produced. A run can be strong on one and weak on the other, and the gap between them is more informative than either figure alone.

A large gap — good net WPM, poor accuracy — means you are making and fixing a great many errors. That works on a typing test and fails in data entry, where an error you do not notice is not a slower keystroke but a wrong record.

How KPH (Keystrokes Per Hour) is calculated

In banking, finance, accounting, and warehouse inventory jobs, employers measure typing on the numeric 10-key keypad using KPH (Keystrokes Per Hour) instead of WPM. TypingNova calculates KPH by multiplying your correct characters per minute by 60.

A standard benchmark of 8,000 KPH corresponds to approximately 27 words per minute on numbers, while 10,000 to 12,000+ KPH represents an experienced 10-key operator. The KPH typing test has a converter for either direction and sets out which figure each kind of role asks for.

Why your score moves between runs

A single result varies by several WPM in either direction, and none of that variation means your typing changed. Treating one run as your speed is the commonest way people mislead themselves about progress.

• Warm-up. A first attempt from cold typically lands several WPM under your third.
• The prompt. Familiar words and common letter pairs type faster than unusual ones, and the draw differs each run.
• The keyboard. Key travel and spacing matter more than people expect. Comparing a laptop run against an external keyboard compares two things.
• Length. A one-minute score measures your opening burst and sits above a five-minute one almost every time.

The useful figure is the median of about five runs at one length on one keyboard, tracked over weeks. The progress tracker keeps that history on your own device so the trend is visible rather than remembered.

Comparing a score from another typing site

Two sites can report different WPM for identical typing, because the arithmetic above is standard and the counting rules are not. Before you compare, check three things:

1. Whether the headline number is net or gross. Some sites lead with gross, which is always the higher figure. This site leads with net.
2. Whether corrected errors are forgiven. Sites that ignore backspaced mistakes report higher accuracy than sites that do not. This one counts them.
3. Whether you can skip ahead. Tests that let you move past a word you cannot type produce speeds that no strict test will reproduce.

None of those choices is wrong, and there is no governing body that settles them. It simply means a number is only meaningful with its method attached, which is why the formulas on this page are published rather than described.

What these numbers do not tell you

A typing test measures copying, and copying is not the job. It is worth being clear about the limits of anything measured here:

• It is not your writing speed. Composing while you type is far slower than transcribing, because the bottleneck moves from your hands to your decisions.
• It does not transfer between materials. Prose speed predicts very little about structured records, and nothing at all about the numeric keypad.
• It is not proctored. Nobody watched you type, and nothing here verifies who was at the keyboard.
• It is a practice result. An employer testing you will use their own assessment, and a certificate from this site records a practice score rather than replacing that.

What it does measure, it measures the same way every time, which is what makes the trend worth watching even where the absolute number is not worth quoting.

When timing starts & test rules

The test timer starts when your first accepted character is entered. Changing mode or duration resets the test. Timed modes finish automatically when time expires; No Time mode finishes when you select Finish Test after meeting the 30-second and 150-character threshold.