Every few months, a post claiming Dvorak or Colemak will "double your typing speed" resurfaces. The real story is less dramatic and more useful.
Why QWERTY exists in the first place
QWERTY wasn't designed for speed - it was designed in the 1870s partly to reduce mechanical typebar jams on early typewriters, spreading common letter pairs apart. That constraint disappeared with electronic keyboards, but QWERTY stuck because switching costs (retraining an entire population) far outweighed the marginal efficiency gains of a "better" layout. This is the core reason alternative layouts never displaced it, and it matters for your decision too: the layout itself is a smaller factor than the switching cost implies.
Dvorak: the original alternative
Dvorak (1936) puts the most-used English letters on the home row and alternates hands more than QWERTY. The theoretical case is real - less finger travel, more hand alternation, which should reduce fatigue and increase speed ceiling. In practice, controlled studies comparing trained QWERTY typists to trained Dvorak typists show smaller speed differences than layout advocates claim - often single-digit percentage gains, not the 20-30% sometimes cited in forum posts. The strongest evidence for Dvorak is reduced finger travel distance, which may matter more for RSI/fatigue than raw WPM.
Colemak: the practical middle ground
Colemak (2006) was designed specifically to fix Dvorak's biggest adoption problem: it keeps most non-letter shortcuts (Ctrl+Z, Ctrl+C, Ctrl+V) in familiar QWERTY positions since it only remaps 17 keys instead of nearly the whole board. This makes it meaningfully easier to learn if you already type QWERTY, and it's become the most common "serious" alternative layout choice among programmers and power users for that reason - not because it's dramatically faster, but because the retraining cost is lower.
What switching layouts actually costs you
This is the part layout advocates undersell: expect weeks to a couple months of significantly reduced typing speed while retraining, and expect QWERTY muscle memory to partially resurface under stress or fatigue for a long time afterward (many switchers report reflexively hitting QWERTY patterns when tired, even years later). If your job or daily life depends on typing speed, that transition window has a real cost.
When switching makes sense
-
You're experiencing genuine hand/wrist strain and have ruled out other causes (posture, keyboard hardware, break frequency) - Dvorak's reduced finger travel may help here more than a WPM gain would.
-
You're starting from a low baseline anyway (learning to type properly for the first time) - there's an argument for learning Colemak/Dvorak from scratch rather than QWERTY, since you're paying the learning cost either way.
-
You genuinely enjoy the optimization problem for its own sake - this is a legitimate reason, just be honest that it's the reason.
When it doesn't
-
You're already a fast, comfortable QWERTY typist with no pain or fatigue issues. The realistic ceiling gain doesn't justify weeks of reduced productivity.
-
You need to type on shared/other people's computers regularly - QWERTY is universal, alternative layouts aren't, and switching back and forth constantly undermines the muscle memory you're trying to build.
The honest verdict
Layout choice is a real but modest factor in typing performance - smaller than people expect, and much smaller than fundamentals like chunking, finger independence, and error recovery (topics worth more of your practice time regardless of which layout you're on). If you're chasing WPM specifically, fixing technique on the layout you already know will almost always beat switching layouts and starting over.
If you do switch, do it deliberately
Commit to a hard cutover rather than splitting time between layouts - alternating between QWERTY and a new layout day to day tends to slow both down and extends the retraining period far longer than a clean switch would. Pick a low-stakes window (not mid-deadline, not before a stretch where fast, reliable typing actually matters) and expect the first two to three weeks to be genuinely frustrating before it starts to feel normal again.