Best keyboard for typing

No keyboard will make you fast, but the wrong one can make you sore. What to look for, and what to ignore.

Test your current keyboard

Somewhere right now, someone is choosing between a key switch that registers a press after 2 millimetres under 45 centinewtons of force and one that does it after 2.2 millimetres under 50, and wondering which will make them faster. Neither will. The research on keyboards and typing speed is thin, and what there is says that within the normal range of keyboards, speed hardly changes.

That doesn't mean the keyboard is unimportant. If you type for hours a day, the right one is more comfortable, kinder to your hands and quieter for the people around you. Here is what really differs between keyboards, what the studies found, and how to test one yourself before you trust anybody's review, including this one.

What makes you fast is mostly you

The 2018 study of 168,960 typists that this site's comparisons are based on found something revealing. The fastest tenth of typists held each key down for about 104 milliseconds; the slowest tenth for about 129. Almost the same. The real difference was the gap between one key and the next: about 120 milliseconds for the fast group against more than 480 for the slow. Speed comes from fingers that already know where they're going next, and no switch can supply that. How to type 100 WPM covers the habits that do.

The one thing researchers have varied carefully is key spacing. In a 2013 study, typists used keyboards with keys 17, 18 or 19 millimetres apart (19 is the usual spacing on desktop keyboards). Their speed, errors and comfort ratings didn't differ significantly between the three. Only a cramped 16 mm layout was clearly worse. If even a 2 mm squeeze in key spacing makes no measurable difference, the colour of a switch is unlikely to.

What does matter for speed is familiarity. Any new keyboard feels slow for a few days while your fingers recalibrate: keys sit at slightly different heights, need different pressure and stop at different depths. That's why the first test on a new keyboard is never a fair one.

The three kinds of key

Rubber dome (membrane) keyboards

Most office keyboards and cheap desktop keyboards work this way. Under each key is a small rubber dome; press it and it collapses, pushing a contact onto a printed membrane underneath. They're quiet, cheap and usually spill-resistant. The feel varies a lot from one model to the next, and many register the press only near the bottom of the stroke, so the habit of hammering every key all the way down comes easily.

Scissor switches

Laptops and slim desktop keyboards use a scissor-shaped hinge under each key that keeps it level wherever you press it. Travel is short: Apple's 2019 16-inch MacBook Pro, for example, has 1 mm of key travel. Short travel suits a light touch and takes some getting used to if you come from a chunkier keyboard. The flat, closely spaced keys can also be awkward with long nails.

Mechanical switches

Each key has its own switch with a metal spring and contacts. Most designs register partway down, so you can learn to type without bottoming out. They come in three broad types, and Cherry's MX2A range is a handy example because its figures are published:

Figures from Cherry's product pages. A centinewton (cN) is a hundredth of a newton: 45 cN is about the weight of 46 grams resting on the key, which is why shops often call these "45 gram" switches. Other makers' switches range from lighter to much heavier.

Does a mechanical keyboard make you faster?

There's surprisingly little research comparing mechanical and rubber dome keyboards directly, and nothing that shows one makes people type faster. The better-studied question is how hard keys are to press, and there the answer is clear enough.

In a 1997 study, ten experienced typists used keyboards whose keys needed 0.34, 0.47 or 1.02 newtons to register. Between the two lighter keyboards, the force their fingertips applied and the activity in their finger muscles didn't change. On the stiff keyboard, fingertip force went up by about 40% and muscle activity by about 20%. The authors recommended keys of 0.47 N or lighter. That's 47 cN, about the same as the switches in the table above, so most ordinary keyboards are already in the comfortable range, and a stiff one makes your hands work harder for nothing.

So a mechanical keyboard's real advantages are consistency and feel. Every key behaves the same, the press registers at a predictable point, and a tactile bump tells you a key has gone in without slamming it to the bottom. Some people find that makes a light touch easier to learn. Others type just as well on a laptop. If you like the feel and the price, it's a fine choice. If you're buying one to gain 10 WPM, keep your money and do ten minutes of practice a day instead.

Size and layout

Keyboards are usually described by how much of a full-size board they keep:

• Full-size (104 keys in the US layout, 105 in the international one): everything, including the number pad. Essential if you enter figures for a living; see 10-key typing.
• Tenkeyless (87 or 88 keys): no number pad, so the mouse can sit closer to your body, which saves a long reach to the right hundreds of times a day.
• 75% and 65%: the function row, the navigation keys or both are squeezed in or moved to a second layer, reached with an Fn key.
• 60%: no arrow keys, function row or navigation keys at all without the Fn layer.

Small keyboards are lovely for travel and small desks, and annoying if you edit a lot of text. If you use Home, End and the arrow keys to move around a document, and you should (typing tips and tricks explains why), keep them on dedicated keys.

Check the layout too. US keyboards (ANSI) have a wide left Shift and a long, single-row Enter; UK and most European keyboards (ISO) have a tall Enter and an extra key beside the left Shift. Buy the one that matches the layout set on your computer, or the symbols printed on the keys won't match what appears on screen. If you're curious about remapping the letters themselves, that's a different question: see QWERTY vs Dvorak vs Colemak.

Ergonomic keyboards

A standard keyboard makes your hands meet in the middle, which bends the wrists outwards, and lies flat or tilts towards you, which bends them up. Ergonomic designs try to undo that in three ways: splitting the keyboard so each hand lines up with its forearm, raising the middle like a tent (often called gabling or tenting) so your palms turn partly inwards, and tilting the front edge up (negative slope) so your wrists stay straight.

The posture research supports all three. A 2007 study compared six keyboards and found the most outward wrist bending on a conventional laptop and the most turned-down forearms on a conventional straight keyboard; a fixed, moderately split keyboard with a moderate tent came closest to a neutral position overall. A 1995 study gave typists a keyboard tray that sloped away from them: they settled on a slope of about 12 degrees, and their wrists went from bending back by about 13 degrees to almost straight.

What nobody has shown is that a keyboard prevents injuries. A 2018 Cochrane review of ergonomic changes for office workers, covering 15 trials and 2,165 workers, found the evidence for preventing arm and neck problems mostly inconclusive; the clearest hint of a benefit was for extra breaks in data entry work. So start with the free fixes: fold the keyboard feet away, set your chair so your forearms slope gently down, and take short breaks (typing posture has the full setup). If you're still uncomfortable, an ergonomic keyboard is a reasonable thing to try, ideally with a generous return policy.

One side effect of a split keyboard is worth knowing about: it ends any habit of crossing over. If you press B with your right hand or Y with your left, the gap in the middle will retrain you, which is good for your technique and irritating for a week.

General advice for comfortable typing, not medical advice. If you have pain, numbness or tingling in your hands, wrists or arms that doesn't go away, see a doctor or physiotherapist.

Rollover, wireless and other spec-sheet words

Rollover. Fast typists often press the next key before letting go of the last one; the fastest tenth in the 2018 study did it on about half of their key presses. Your keyboard has to register both. "6KRO" means a keyboard reports up to six ordinary keys held at once, the limit of the basic USB keyboard format; "NKRO" means any number. For typing, either is plenty: six ordinary keys held down at the same moment is far beyond anything typing produces. Cheap keyboards can miss or invent keys when certain combinations are held together (called ghosting), which bothers gamers far more than typists.

Wired or wireless. For typing it doesn't matter. A small, steady delay shifts every key by the same amount, so your rhythm is untouched. The practical questions are batteries, and whether you want one keyboard to switch between several devices.

Gaming features. Polling rates, adjustable actuation points and lighting effects are aimed at games. They do no harm, and they don't make anyone type faster.

Noise. Clicky switches are loud by design, which is lovely for you and wearing for everyone within ten metres. In a shared office, choose linear or tactile switches, a quieter variant with dampers, or a scissor or rubber dome keyboard.

Choosing for your situation

Test it yourself

Reviews tell you how a keyboard felt to somebody else's hands. Your own numbers are better. Before you switch, take five one-minute tests on your current keyboard and note the median speed and accuracy. Then use the new keyboard for everything for a week, without going back. After the week, take five more tests and compare medians.

If the new keyboard is within a few WPM and feels better, keep it: you'll catch up as your fingers adjust. If your accuracy has dropped, look at the slowest keys and the mistake types on the results page. Hitting the neighbouring key a lot usually means the spacing or key shape is throwing your aim, and that tends to improve with a few more days. If it's still worse after two weeks, it's not the keyboard for you, and that's what return policies are for.

Where to start

Before buying anything, fold the feet away on the keyboard you have, check your chair height and take a one-minute test. Plenty of people find the old keyboard was fine and the setup was the problem. If you do buy, decide on the size first (do you need a number pad, do you need arrow keys?), then the switch type by feel, in a shop or with a cheap switch tester. Light keys, a layout that matches your computer and a keyboard you enjoy using will do more for you than any spec sheet.

Sources

• Dhakal, Feit, Kristensson and Oulasvirta, “Observations on Typing from 136 Million Keystrokes”, CHI 2018.
• Pereira, Lee, Sadeeshkumar, Laroche, Odell and Rempel, “The effect of keyboard key spacing on typing speed, error, usability, and biomechanics: Part 1”, Human Factors 55(3), 2013.
• Rempel and colleagues, “The effect of keyboard keyswitch make force on applied force and finger flexor muscle activity”, Ergonomics 40(8), 1997.
• Rempel, Barr, Brafman and Young, “The effect of six keyboard designs on wrist and forearm postures”, Applied Ergonomics 38(3), 2007.
• Hedge and Powers, “Wrist postures while keyboarding: effects of a negative slope keyboard system and full motion forearm supports”, Ergonomics 38(3), 1995.
• Hoe and colleagues, “Ergonomic interventions for preventing work-related musculoskeletal disorders of the upper limb and neck among office workers”, Cochrane Database of Systematic Reviews, 2018 (summary).
• Cherry, product pages for the MX2A Red, MX2A Brown and MX2A Blue switches.
• USB Implementers Forum, “Device Class Definition for Human Interface Devices (HID)”, version 1.11, Appendix B (the boot keyboard report).
• Apple, “Apple introduces 16-inch MacBook Pro”, press release, November 2019.