Desk height and monitor position: the numbers nobody measured on you
We made the argument for workbenches already: surface height should follow the task and the body rather than a convention. The same reasoning applies one room over, and it reaches an uncomfortable conclusion — the standard desk is too tall for most people who sit at one.
If you are building a desk rather than buying one, the single advantage is that you are not obliged to reproduce a number that was never measured on you. Worth using.
Where 29 inches came from
The 29 to 30 in. desk is a mid-century office standard, set for a posture that involved a mechanical typewriter — hands high, wrists cocked, forearms angled up to reach a keyboard sitting on top of the desk surface with a carriage above it.
Nobody works that way now. A modern keyboard is about an inch thick and sits flat. The posture it wants is forearms parallel to the floor, wrists neutral, and that is a different height from the one the convention encodes.
Seated elbow height is the measurement
Sit in the chair you will actually use. Feet flat on the floor, thighs roughly parallel to it, back against the chair, upper arms hanging straight down and forearms level. Measure from the floor to the underside of your elbow.
seat height, feet flat — about 17 to 18 in.
elbow above the seat — about 9 to 10 in.
seated elbow height — about 26 to 28 in.
standard desk — 29 to 30 in.
difference — the desk sits 1 to 4 in. too high
Those are typical ranges, not your numbers — measure rather than assume. But the direction holds for most people of average height and all of it holds more strongly for anyone shorter.
What the gap produces is a wrist held in extension, bent upward, for every hour at the desk. That is the posture that generates the ache people attribute to the keyboard or the mouse. The keyboard is rarely the problem; the surface it sits on is.
Note the symmetry with the bench. The standard 34 in. workbench sits above the planing height for an average-height person, and the standard 29 in. desk sits above the keyboard height for the same person. Both conventions were set high, and both were set for a task people no longer perform the way they did. If any of this is new, the longer version is in bench height: the standard number is wrong for most people.
The order is chair, then desk, then screen
This matters more than any individual number, and it is where most setups go wrong — people fix the desk first and then adjust everything else around a height that was chosen by a catalogue.
- Chair first. Set it so your feet are flat and your thighs are level. This height is determined by your leg length and nothing else. If the chair cannot go low enough, that is the problem to solve first.
- Desk to the elbow. Now measure, and build to that number. If you are stuck with an existing desk that is too tall, a keyboard tray below the surface is the retrofit — it is solving exactly this.
- Screen last. Monitor position is independent of the other two and is adjusted by the stand, not by the desk.
Doing it in the other order means compensating for each mistake with the next decision, which is how people end up with a footrest propping them up to reach a desk that is too high.
The screen
Two numbers, and they are less contentious than desk height because the anatomy is simpler.
The relaxed gaze falls slightly below horizontal, so a screen centred on eye
level is marginally too high.
Distance — roughly arm's length. About 24 to 30 in. for a 24 in.
panel, 28 to 34 in. for a 27 in. panel. Larger screens want more distance,
not less.
Eye height, seated, 5 ft 10 in. person — about 45 to 47 in. from floor
Run the arithmetic for a common case: a 28 in. desk, a monitor whose stand puts the centre of the panel 14 in. above the surface, gives a screen centre at 42 in. against a seated eye height near 46 in. The screen is four inches low, which is the position almost everyone sits in, and it is why neck complaints are more common than wrist complaints despite the wrist being the thing people worry about.
A riser fixes it and costs nothing to make. The correct height is whatever brings the top edge level with your eyes, which is a measurement you take once and then build to.
Standing changes the numbers, not the method
If the desk is adjustable, the standing position is derived the same way: elbow height, measured standing, shoes on. It is typically 40 to 45 in. for an average-height adult, and it is not simply the sitting number plus a constant — the ratio differs between people with different leg-to-torso proportions, which is exactly why a preset height on an adjustable desk is a starting point rather than an answer.
The screen has to come up with you, which is the thing most people discover after buying the frame. A monitor arm handles it; a fixed riser does not.
Build to the measurement, then live with it
Same advice as the bench: make the legs an inch over the calculated height and work there for a couple of weeks before trimming. Removing an inch is twenty minutes. Adding one is a repair.
And add the top thickness to the calculation before cutting anything — a desk built to 28 in. of leg with a 1 in. top is a 29 in. desk, which is the number you were trying to avoid.
Which raises the other question about the top, and it is not cosmetic: how thick it has to be before a monitor sitting in the middle of an unsupported span starts to produce a visible dip. That has an answer with arithmetic behind it, and it is the subject of the next note.