Typing Tests in Hiring, Explained

In most cases the figure is inherited rather than derived. It appeared in a job description years ago, has been copied forward, and nobody currently hiring can say which test it was calibrated against.

Where it does mean something specific, it usually traces to one of two places: a certification standard the employer accepts, or the throughput of whoever last held the role. The second is more common than the first and is worth thinking about, because it means the bar is a description of a person rather than a measurement standard.

The practical implication is that the number is best treated as a rough band. An advert asking for 60 is unlikely to reject a candidate at 58 and unlikely to be satisfied by one at 35. Where the requirement is genuinely load-bearing — transcription, court reporting, some data-entry roles — the employer will normally say so and will test rather than ask.

Four are common, and they measure different things.

A plain timed transcription test is the most frequent: a passage of ordinary prose, one to five minutes, net speed reported. This is closest to what people mean by a typing test and closest to what the tests on this site measure.

A ten-key or numeric test is standard wherever the work is numbers rather than text, and it is a genuinely separate skill worked by one hand on a keypad with its own home position. Prose speed predicts it poorly, which is why the numeric keypad drill exists apart from the number-row work.

An error-capped test sets a minimum accuracy alongside the speed target and disqualifies a run that exceeds the error allowance regardless of pace. These are common in roles where mistakes are expensive, and they reward a completely different strategy from a speed-only test.

And a data-entry simulation puts you in a form with fields, tabbing between them, often with a mix of text and numbers. Raw typing is only part of it; navigation without a mouse is the rest, which is covered in typing for data entry.

Three things systematically pull the assessment figure below the number people quote for themselves.

The first is that self-quoted figures usually come from a short test, often a best-of-several, on familiar text. Assessment figures come from a single attempt, on unfamiliar text, once. A best-of-five on a fifteen-second test is a substantially different quantity from one attempt at three minutes.

The second is the error convention. If the assessment counts every incorrect keystroke rather than only what remained wrong, a typist who corrects constantly will score well below their usual figure. The definitions and how far apart they can be are set out in what a typing accuracy percentage means.

The third is content. Assessment passages often include proper nouns, numbers and punctuation at rates ordinary practice text does not, and almost everyone is slower on those. Comparing your prose figure against a punctuation-heavy test will show you the size of your own gap before someone else measures it.

This is the format that most rewards changing your approach, because the optimal strategy under an error cap is not the optimal strategy under a speed target.

Under a speed target you push and correct. Under an error cap that strategy can fail outright: exceeding the allowance invalidates the run no matter how fast it was. The correct approach is to identify the slowest pace that clears the speed requirement and type at that, treating the speed target as a floor rather than a goal.

That is a genuinely different discipline and it can be practised. The zero-error accuracy drill trains a clean run as the only success condition, and the one life game makes the consequence of a single error immediate, which is a fair simulation of how an error cap actually feels.

The common failure is practising for speed and then meeting an error cap on the day, at which point the habit of pushing is exactly wrong and there is no time to change it.

Two honest paths, and it is worth knowing which one you are on.

If the gap is small — under about ten words a minute — it is usually closeable within weeks of structured practice, and the fastest route is almost always accuracy rather than pace, for the reasons set out in accuracy versus speed. Reducing errors raises the net figure without requiring your fingers to move faster at all.

If the gap is large, or if you are still looking at the keyboard, the honest timeline is months rather than weeks and the work is different: it is the practice path rather than test repetition. Repeating tests is a measurement activity and it does not train anything.

In either case, practise on the format you will face. If the assessment is three minutes of prose, a three-minute test is worth more than twenty short ones, because pacing over that window is itself part of what is being measured.

Most employers will answer these if asked, and the answers change how you should prepare.

How long is the test, and is it one attempt or best of several? Is the reported figure gross or net? Does correcting a mistake remove the error from the count, or does the keystroke still count against you? Is there a minimum accuracy, and what happens if it is missed? Is the content ordinary prose, or does it include numbers, names and punctuation?

Asking is not presumptuous — it is the same clarification you would seek about any other assessment, and an employer who cannot answer has told you something useful about how much the number really matters to them.

The broader case that adverts ought to specify these things in the first place, rather than leaving candidates to guess, is made in the case against WPM in job adverts.