Typing practice for students: what the speed actually buys you

Typing practice for students is usually sold on a promise nobody has checked: that faster typing produces better work. A narrower version of that is true and far more useful. Your typing speed costs you something whenever somebody else sets the pace, and it costs you close to nothing when you are alone with an essay and a deadline.

If you want the short answer: aim for 50 to 60 words per minute at 95% accuracy or better. That is a working target rather than a standard, and it is the point where the keyboard stops interfering with your schoolwork. You do not need to be an elite typist.

The longer answer is that the question is not whether to practise. It is which of the three situations below is the one actually taking marks off you, because that changes how many weeks this is worth and what you should drill.

Notes, where you are racing a speaker#

A lecturer is not waiting for you. Steve Tauroza and Desmond Allison measured speech across four situations, academic lectures among them, and proposed 125 to 160 words per minute as the range for ordinary speech, replacing an older estimate taken from radio announcers. Call it 140 for somebody talking through a slide.

Read the right-hand column carefully, because it is a ratio of rates and not a measure of understanding. At 40 words per minute your hands run at under a third of the speaker's pace; a regular touch typist at 70 runs at half. Nobody in that room is transcribing the lecture, and no typing speed would let them. Chasing one is the wrong goal.

What speed buys you instead is the gap between hearing a sentence and having finished writing it down. That gap is where you decide whether the sentence mattered. At 30 words per minute you are more likely to spend the hour catching up on the last point while the next one goes past. At 55 there is room to finish a note and still be listening.

Essays, where speed matters least and a floor still exists#

Composing is slower than copying by a wide margin, and your hands are not the reason. Clare-Marie Karat and colleagues, in a 1999 study of average computer users presented at the CHI conference, measured 32.5 words per minute transcribing against 19.0 composing. You stop to think, and the thinking is most of the time.

Which reads like an argument that typing speed cannot matter for essays, and that is where most articles leave it. It misses what happens below a threshold. Vincent Connelly, Deborah Gee and Elinor Walsh measured handwriting and keyboarding in 300 primary school children, and found handwriting was faster than keyboarding at every age they looked at, with only a small minority of ten and eleven year olds typing faster than they wrote. When those children then wrote compositions both ways, the typed ones were judged worse.

The explanation they offer is that nobody had taught these children to type, so hunting for keys came out of the effort available for writing. Their conclusion was that explicit touch-typing instruction is what makes a word processor worth having in the first place.

So the shape of it is a floor rather than a slope. Type slower than you write and the keyboard takes something off your work. Get past that point and further speed adds very little to an essay, and starts paying you back somewhere else instead.

Typed exams, and how much of this is already real#

This part is oversold, so here is what the regulator actually proposed. Ofqual consulted between December 2025 and March 2026 on regulating on-screen assessment in GCSEs, AS and A levels. Pen and paper stays the main mode. Boards would be allowed up to two on-screen specifications each, and subjects with more than 100,000 entries a year are excluded at this stage, which rules out most of the large GCSEs.

So if you are sitting GCSEs, AS or A levels in England this year or next, expect paper to stay the main format. Other countries and other boards are each on their own timetable, and the honest answer anywhere is to check yours rather than to believe a headline about exams going digital.

University is where the transition gets much harder to ignore. Essays, take-home papers, seminar notes and much of the coursework are typed, and nobody there is going to teach you how. It is the shift worth being ready for.

The number worth aiming for#

Three thresholds, in the order they matter:

• Fast enough to beat your own handwriting. Below that line the keyboard may be taking something off your work rather than giving you time back, which is the Connelly finding above. There is no universal number for this and you should not trust one: time yourself writing a paragraph by hand, then type the same paragraph, and use your own two figures.
• 50 to 60 wpm, at 95% accuracy or better. A practical target for schoolwork: enough that you are writing your own notes rather than spending the lecture catching up on the last point.
• Beyond 70 wpm. Real, and worth having if you write code or work with data, but it will not improve an essay.

Accuracy comes first at every one of those, and not as a matter of discipline. An error costs you far more than the character it landed on: you have to notice it, stop, backspace, retype and find your place again, and the noticing is the expensive part because it breaks the rhythm carrying the rest of the sentence. A student who types quickly and inaccurately is buying more to undo. We set out why that trade goes the way it does separately. Fix the error rate before you push the pace.

You will need your own number before any of that means anything. Take a timed test and read the accuracy figure first, then put the speed against the ranges on our speed-by-age reference, which carries the warning that the spread inside any one age band is wider than the gap between bands.

Notes are selection, not transcription#

You may have been told to handwrite your notes. The research behind that advice is thinner than the headlines were.

Pam Mueller and Daniel Oppenheimer reported in 2014 that longhand note-takers outperformed laptop users, and suggested the laptop group did worse because they transcribed verbatim instead of processing what they heard. Kayla Morehead, John Dunlosky and Katherine Rawson ran it again in 2019 with extra conditions, and performance did not consistently differ between any of their groups, including one that took no notes at all. Their own summary was that calling either method superior looks premature. Heather Urry and a large team published a direct replication in 2021 in Psychological Science, with meta-analyses across similar studies, and found small effects favouring longhand that fell short of significance.

What survives all of that is the mechanism rather than the tool. Copying a speaker word for word is a poor way to learn, whether you do it with a pen or a keyboard. The pen worked in the original study partly because it was slow enough to force a choice about what to keep.

Which means the speed is only worth having if you spend it on deciding rather than on typing more. Write the point, not the sentence. If your notes read like a transcript, the fix is not a slower keyboard.

The keys school work needs and a typing test never drills#

Most typing practice is lowercase prose, and most academic writing is not. A reference is brackets, quotation marks, colons, initials with full stops, and a page range with a digit at each end. A chemistry answer is numbers and subscript-adjacent punctuation. A statistics module is decimal points and percent signs by the hundred.

You can spot this in yourself in one paragraph. Type a few lines of ordinary prose, then type a citation with a year, a volume number and page range. If the second one takes three times as long per character, the problem is the number row, not your speed, and no amount of prose practice will touch it.

That is what structured practice is for, and the code drills are the heaviest version of the same thing: brackets and operators on their own until the reach stops being a decision.

Four weeks, then upkeep#

Fifteen minutes, four days a week. Keep the sessions short enough that they survive a week with two essays due in it, because a routine you abandon in February teaches nobody anything.

The order matters more than the schedule does. A wrong finger assignment drilled for three weeks is much harder to remove than to have avoided, which is why the beginner course starts where it does and refuses to move on early.

During exam term, stop practising and type your revision notes instead. The keyboard work is done by then; what you need is for it to stay done.

For a teacher handing this to a class#

The whole catalogue is 185 lessons across 5 courses, free, with no per-seat licence and no purchase order, and it works offline once loaded, which matters more on a school network than anything else on this page. The lessons run in order if you want a curriculum, and a class can be dropped in at any point in it.

Two things worth doing on day one. Set an accuracy floor before you mention speed, because a class told to go fast will rehearse errors for six weeks and you will spend the rest of term undoing it. And run the first measurement privately: a public speed ranking in week one tells thirty students that this is a thing some of them are bad at.

For a younger group, or a sibling watching over a shoulder, typing for children covers what changes below about eleven, where ten minutes is a full session and the speed figure means very little.

Two questions students actually ask#

Is it too late if I am already at university? No, and the arithmetic favours you. Four weeks at fifteen minutes, four days a week, is four hours, which is less than one essay costs you, and you will type every essay after it faster than the one you skipped the practice for.

Will this raise my grades? Not on its own, and anybody promising that is selling something. What it does is stop the keyboard taking anything off you: no notes lost to being behind, no reference that takes a minute, no paragraph that came out shorter because writing it was work. Whether the writing underneath is good is still your problem.