How to learn a new keyboard layout
Switching layouts takes about ninety days to reach your old speed, and it is not a speed upgrade: the evidence does not support one. It is a comfort change, and the hard parts are predictable. Week two is worse than week one because you temporarily get slower at both layouts, and most courses abandon you around day twenty with all the letters learned and no idea what to fix. Both are survivable if you know they are coming.
How long does it take to learn a new keyboard layout?
Expect about three months to get back to your old speed. Learner accounts converge closely here: roughly 8 words per minute on day one, about 53 by day thirty, and parity with your previous QWERTY speed near day ninety of consistent daily practice. The first week is the steepest and the most discouraging, because you are typing at the speed you did as a child while knowing exactly how fast you used to be. Nothing about that curve means you are doing it wrong.
Will a new keyboard layout make you type faster?
Almost certainly not, and you should switch for other reasons. The classic study on this, The Fable of the Keys (Liebowitz & Margolis, 1990), found no dependable speed advantage for Dvorak over QWERTY, and no reliable evidence has established one for Colemak either. What alternative layouts do measurably change is where your fingers travel: they put the most common English letters under your strongest fingers and on the home row, which reduces total finger movement and, for many people, strain. Switch for comfort. If someone promises you a speed windfall, they are selling something.
Do you need to change your operating system's keyboard settings?
No, and you probably should not at first. Browser-based trainers, including this one, can emulate a layout by reading which physical key you pressed and resolving it to the character your target layout assigns, so you can practise Colemak for twenty minutes and go straight back to working in QWERTY. This is not unique to any one tool; several trainers support it. Practising in emulation first lets you separate learning the layout from betting your working day on it, which matters most in the first fortnight when your accuracy is genuinely bad.
Why does week two feel worse than week one?
Because you briefly get worse at both layouts, and this is the single most common reason people quit. Learners report it consistently: somewhere around the end of the second week your new layout is still slow, and your old one has started to degrade too, because your hands now hold two competing maps for the same keys. Skill research calls this negative transfer. It is a normal stage rather than evidence that you have ruined your typing, and it resolves as the new layout consolidates. Knowing it is coming is most of the defence against it.
Should you keep using QWERTY while learning a new layout?
Yes if you need to work, but expect it to slow the switch down. Every hour spent in your old layout keeps that muscle memory alive, which is exactly what makes the week-two dip drag on, so people who switch cold turkey generally reach parity sooner and suffer more doing it. The practical middle path most learners settle on is to keep your job in QWERTY, practise the new layout in dedicated sessions, and only move real work across once the new layout clears roughly 30 words per minute. Staying permanently bi-layout is a legitimate end state, and plenty of people do it deliberately, but treat it as a separate skill rather than a stage.
What should you practise first when learning a layout?
The home row, two new keys at a time, with accuracy held high before speed. Start from the eight home-row positions your fingers rest on, add a pair of keys only once the current set is comfortable, and keep your accuracy above roughly 97% throughout, because in the CHI 2018 keystroke dataset (Dhakal et al., 136 million keystrokes from 168,000 typists) typists who prioritised accuracy improved their speed about 19% faster than those who chased speed first. Errors trained at speed are expensive to remove later: each uncorrected mistake costs an estimated three to five extra keystrokes to fix (Soukoreff & MacKenzie, CHI 2003).
How much should you practise each day?
Twenty to thirty minutes daily beats a long weekend session. The classic distributed-practice study on typing training (Baddeley & Longman, 1978) found that one hour of practice per day produced better skill gains per hour invested than two- to four-hour blocks, and the same pattern shows up in how layout switchers describe their progress. Short and daily also keeps the week-two dip shorter, because consolidation happens between sessions rather than during them.
When do most people give up on a new layout?
At two predictable points, and a third that is really a tooling failure. Learner accounts cluster the first around days three to ten, when the novelty has worn off and you are still unusably slow, and the second at the end of week two during the dip described above. The third arrives around day twenty, when most guided courses simply run out of lessons: you know all the letters, you are still slow, and nothing tells you what to practise next. That last one is not a motivation problem. It is the point where you need per-key diagnostics rather than another sequence of lessons.
Which keyboard layout should you choose?
Pick one with an active community and stop optimising. Colemak leaves Z, X, C and V where QWERTY has them, so your existing shortcuts survive, which makes it the gentlest switch; Colemak-DH is Colemak's most-requested ergonomic fix, moving D and H out of the home row's stretch columns, and is worth considering if you are also eyeing a split or columnar keyboard; Dvorak is the oldest and the best supported at the operating-system level, shipping with Windows, macOS and Linux. Workman and Graphite optimise further on paper but have smaller communities and thinner tooling around them. The gap between the good options is much smaller than the gap between picking one and spending another month deciding.
Put it into practice
Polytaipe is built around the two failure points above. It emulates 12 layouts in the browser so you never have to change your system settings, walks you outward from the home row two keys at a time, and once the lessons run out it keeps going: your weakest keys and letter-pairs become the next drill instead of leaving you to guess.
Start learning a layout →Keep reading: every supported layout with its honest tradeoffs and learning curve, a head-to-head on Colemak vs QWERTY, or how to type faster on the layout you already have.