Reading Rhythm: Why Line Height Matters More Than Font Choice

Font choice gets disproportionate attention in design reviews because it's the most visible decision on the page - everyone can look at two typefaces side by side and have an opinion. Line height rarely gets debated with the same energy, and often ships as whatever the framework or design tool defaulted to, unreviewed and unquestioned. That's backwards. Line height does more to determine whether a page is actually comfortable to read than the typeface sitting inside it. A mediocre font set with generous, well-tuned line height will out-read a beautiful font set too tight, every time.

The reason line height gets skipped over is that it's hard to see in isolation. A single line of text doesn't have a line height problem - the gap only becomes visible, and only becomes a problem, once there are multiple lines the eye has to travel between. That makes it a structural property of a paragraph rather than a stylistic property of a letterform, and structural properties are exactly the kind of thing that get missed in a mockup showing one heading and one line of body copy, because the mockup was never built to reveal them.

Too-tight line height makes the eye lose its place jumping between lines. When lines sit close together, the descenders of one row visually crowd the ascenders of the next, and the return sweep - the eye's jump back to the start of the next line - has too little vertical separation to land reliably. Readers compensate by slowing down, or by re-reading the line they just finished because they're not sure they landed on the right one. None of this registers consciously. It just feels, vaguely, like the text is tiring to get through, without anyone being able to say why.

Too-loose line height creates the opposite problem, and it's less talked about because it looks generous rather than cramped. Once the gap between lines grows past a certain point, each line starts reading as its own unit instead of a continuation of the one before it. Paragraphs stop feeling like connected prose and start feeling like a list of disjointed sentences, loosely related but not actually flowing into each other. A paragraph meant to build an argument across four sentences instead reads as four separate observations, because the visual rhythm that ties them together has been stretched past the point where it does that job.

A dependable starting point is 1.4 to 1.6 times the font size for body text, adjusted down slightly for larger display type, which needs less breathing room per line to stay legible. The reason display type tolerates a tighter ratio is simple: headlines are usually one or two lines, so there's no return sweep to protect, and the larger letterforms already have more inherent separation at the same ratio. Apply a body-text ratio to a 48px headline and the lines will look loose and disconnected for no reason - the fix there is a tighter number, not the same one scaled up.

Line length changes the right answer too, and it's worth naming because it's the variable most people forget. A narrow column of text - a sidebar, a comment thread, a mobile screen - can get away with a slightly tighter line height because the eye doesn't have far to travel before the return sweep. A wide column of long-form text needs more room between lines precisely because the eye is traveling further and has more chance of losing its place. Treating line height as a single fixed number across every column width in a layout is a common way this ends up wrong in exactly the places it matters most.

The obvious pushback is that none of this is something most readers consciously notice, so why spend review time on it. That's true and beside the point. Readers don't notice line height any more than they notice the suspension in a car, but they feel the difference between a smooth ride and a rough one within the first few seconds, and they act on that feeling - they finish the article or they don't, they trust the interface or they drift. The things that go unnoticed at a conscious level are frequently the things doing the most work on comfort and completion rate.

There's also a real edge case worth naming: not every block of text wants generous line height. Dense UI copy - table cells, form labels, compact list items - is often read in short bursts rather than continuously, and a spacious line height there just eats vertical space without buying any comfort, since there's rarely more than one or two lines to travel between. The 1.4–1.6 guidance is a body-text default, not a universal law. Applying it uniformly to every text style in an interface, including the tightly packed ones, produces layouts that feel bloated rather than readable.

In practice, this means line height should be a paired decision with font size, not a single global variable set once and forgotten. A well-built type system defines line height per text style - body, caption, heading, label - rather than inheriting one root value everywhere, because each of those styles has a different job and a different tolerance for tightness. When a new component gets added to a design system, the question isn't just what size its text should be. It's what ratio makes that specific text, at that specific size and column width, actually comfortable to read in context.

What this looks like concretely: take a long-form article set at 18px in a column roughly 65 characters wide. At a 1.2 ratio, the lines feel stacked and the eye has to work to track where it left off - three paragraphs in, most readers feel a low-grade fatigue they can't name. Nudge that same text to a 1.5 ratio, nothing else changed, and the paragraph suddenly reads as one continuous thought instead of a stack of lines. The font hasn't changed. The size hasn't changed. The only thing that moved is the space between the rows, and it's the difference between text people finish and text people abandon.

Whatever the exact number, the deciding test isn't how it looks in a static mockup - it's reading three real paragraphs of it, at actual size, on an actual device, ideally on the screen and in the lighting conditions people will actually use. A line height that looks fine as a rectangle of grey placeholder text in a design file can fall apart the moment it's carrying real sentences with real punctuation and real line breaks, because placeholder text is engineered to look even and tidy in a way real writing never is.

The typeface debate will keep happening in every design review, because it's the fun argument, the one with visible options and strong opinions. Line height deserves at least as much scrutiny, quietly, in the background, because it's usually the thing actually deciding whether the words get read at all. Pick almost any reasonable typeface and pair it with a carefully tuned line height, and the page will out-read a striking typeface paired with whatever value shipped by default. Rhythm, not personality, is what keeps a reader moving down the page.