I spent an embarrassing amount of time last month adjusting the leading on a chapter opener. Leading, for the uninitiated, is the vertical space between lines of text. I was working on the interior layout of a forthcoming book, and something about the opening page felt wrong. The text was too cramped, or too loose, or somehow both at once. I nudged the leading up by half a point. Better, but not right. Down by a quarter point. Worse. Up by three-quarters of a point. There it was. The page suddenly looked like it wanted to be read.

My partner, who had been watching this over my shoulder, asked me if I honestly thought anyone would notice the difference. And the honest answer is: no, not consciously. Nobody picks up a book and thinks “ah, the leading on this chapter opener is exquisite.” But they feel it. They feel it the same way you feel the difference between a well-lit room and a poorly lit one. You might not be able to identify what’s wrong, but you know something is off. Typography is like that. It’s an invisible art that shapes your reading experience in ways you never think about, and that invisibility is precisely what makes it powerful.

I didn’t always care about typography. When I started in publishing, I thought fonts were fonts. You picked one that looked nice, set it at a reasonable size, and moved on. The idea that typographic choices could affect comprehension, mood, pacing, and emotional response seemed absurd. Then I started paying attention, and I couldn’t stop.

The rabbit hole started with a simple observation: why do some books feel easier to read than others, even when the prose quality is similar? I was comparing two novels, both literary fiction, both well-written, both about the same length. One was a pleasure to hold and read. The other felt like work. When I looked more carefully, the difference was entirely in the design. The pleasant book used Garamond, a typeface designed in the sixteenth century, with generous margins, comfortable leading, and chapter headings that gave the eye a place to rest. The difficult book used a modern sans-serif with tight spacing, narrow margins, and no visual breathing room. The words were fine. The container was wrong.

This sent me down a path of reading about typography that I’m still on years later. I read Robert Bringhurst’s “The Elements of Typographic Style,” which is to typography what Strunk and White is to writing: a slim, opinionated guide that makes you see the world differently. I read about the history of typeface design, from Gutenberg’s blackletter to the digital fonts we use today. I learned about kerning (the space between individual letter pairs), tracking (the overall spacing of a block of text), and the dozens of other micro-decisions that go into making text readable.

Here’s what I’ve come to believe: typography is not decoration. It’s communication. Every typographic choice sends a message, whether the designer intended it or not. A serif font says something different from a sans-serif. A large x-height (the height of lowercase letters) conveys something different from a small one. Bold chapter headings create a different reading rhythm than italic ones. These signals are mostly subliminal, but they’re real, and they affect how readers process and respond to text.

Let me get specific. Serif fonts, the ones with the little feet and strokes at the ends of letters, have been the standard for book text for centuries. There’s a reason for this beyond tradition. The serifs create a visual baseline that guides the eye along the line. They connect letters into words and words into sentences. When you’re reading a long block of text, serifs reduce eye fatigue by providing subtle cues about letter shapes and word boundaries. This isn’t opinion; it’s been studied. For extended reading on paper, serif fonts consistently outperform sans-serif fonts in readability tests.

Screen reading is a different story. On low-resolution screens, serifs can appear fuzzy and actually hinder readability. This is why websites historically favored sans-serif fonts like Arial and Helvetica. But as screen resolution has improved (think Retina displays and modern e-readers), the gap has narrowed. Many e-readers now default to serif fonts, and they look beautiful on high-resolution screens. The old rule that “serif is for print, sans-serif is for screen” is becoming less true with each passing year.

Font size is another area where most people’s instincts are wrong. When I design a book interior, I typically set body text between 10.5 and 12 points, depending on the typeface. That might seem small, but point size is misleading. Different fonts at the same point size can look vastly different in apparent size because of variations in x-height, letter width, and stroke thickness. A 10-point Garamond looks considerably smaller than a 10-point Georgia. Judging by point size alone is like judging shoes by their number without specifying the brand.

Margins might be the most underappreciated element of book design. Most readers don’t notice margins at all, which is exactly the point. Good margins are invisible. Bad margins are distracting. Too narrow, and the text feels cramped, the words crowding against the edge of the page like commuters on a packed train. Too wide, and the text floats in a sea of white space, feeling unanchored. The inner margin (the gutter, where the pages meet at the spine) needs to be wider than the outer margin, because some of that space disappears into the binding. Getting this right requires physical prototyping: printing pages, binding test copies, and holding them the way a reader would hold them.

Line length is something I obsess over, and I think it’s the single most important factor in readability after font choice. The optimal line length for body text is generally considered to be 45 to 75 characters per line, including spaces. Shorter than that and your eye is constantly jumping to the next line, creating a choppy reading experience. Longer than that and your eye loses its place when moving from the end of one line to the beginning of the next. Most well-designed books fall in the 60 to 66 character range, and it’s not a coincidence. That range feels right because it matches how our eyes naturally track across a page.

This has practical implications for book formatting that many self-publishers get wrong. If your page is a standard 6×9 trim size with reasonable margins, and you’re using a standard body font at a standard size, your line length will probably be fine. But if you’re using a larger trim size, or narrower margins, or a condensed font, you can easily end up with lines that are too long for comfortable reading. I’ve seen self-published books with lines running 90 or 100 characters, and they’re genuinely difficult to read even when the writing is good. The fix is usually some combination of wider margins, a larger font size, or a typeface with wider letterforms.

Chapter openings are where typography gets to be a little more expressive. The convention of starting a chapter with a drop cap (a large initial letter that extends down several lines) dates back to medieval manuscripts, where monks would hand-paint elaborate initial letters at the start of each section. Modern drop caps serve the same function: they signal a new beginning, give the eye a landing point, and add visual interest to the page. They’re not strictly necessary, and some books work better without them, but when they’re done well, they add a touch of elegance that readers feel even if they don’t notice.

I also care deeply about the small stuff: orphans and widows (single words or short lines stranded at the top or bottom of a page), hyphenation (which should be minimal and never occur on more than two consecutive lines), and the alignment of running headers and page numbers. These details might seem obsessive, but in aggregate, they’re the difference between a book that feels professionally made and one that feels homemade. Readers can’t usually articulate what’s different, but they respond to it. A well-typeset book feels authoritative. A poorly typeset one feels like something’s wrong.

Cover typography deserves its own discussion. The typeface on a book cover isn’t just conveying the title; it’s conveying the genre, the tone, and the intended audience. Look at the thriller section of a bookstore. You’ll see bold, stark sans-serif fonts, often in white or red against a dark background. Now look at literary fiction. More variation, but a tendency toward elegant serifs, sometimes hand-lettered. Romance? Flowing scripts and warm colors. Science fiction? Clean, geometric fonts that suggest technology and the future. These aren’t accidents. They’re genre signals, visual shorthand that tells a browser what kind of book they’re looking at before they read a single word of the description.

At ScrollWorks, we agonize over cover typography as much as any other element of design. The font we choose for a title has to work at thumbnail size (because most book browsing now happens online) and at full size. It has to be legible, distinctive, and tonally appropriate. It has to pair well with the author’s name, the subtitle (if any), and the cover art. Getting all of these things right simultaneously is harder than it sounds. I’ve spent entire days choosing between two fonts for a cover, printing test versions, pinning them to a wall, and staring at them from across the room to see which one reads better at a distance.

The digital age has been both a blessing and a curse for typography. On the positive side, we now have access to thousands of high-quality typefaces, sophisticated layout software, and the ability to adjust every typographic parameter with precision. On the negative side, that same accessibility has led to a lot of bad typography. When anyone can set type, many people do it badly. Poorly formatted e-books with no attention to leading or margins. Self-published print books with clashing fonts and erratic spacing. Websites with body text set in a decorative display font. The tools are better than ever. The average skill level of the people using them hasn’t kept pace.

E-books present a particular typographic challenge. Unlike print books, e-books are reflowable: the text adapts to the reader’s screen size, preferred font, and chosen font size. This means the designer has less control. You can’t guarantee line length, pagination, or even which font the reader will use. Some e-book readers override the publisher’s font choices entirely. This used to frustrate me, but I’ve come to accept it as a tradeoff. E-books prioritize reader control over designer control, and that’s not a bad thing. The designer’s job shifts from dictating the reading experience to establishing defaults that work well while being flexible enough to adapt.

If you’re a self-publishing author reading this and feeling overwhelmed, here’s my practical advice. First, use a serif font for your body text. Garamond, Caslon, Minion, Sabon, or any of the classic book fonts will serve you well. Second, set your body text between 10 and 12 points, depending on the font. Third, use generous margins, at least 0.75 inches on the outside edges and 1 inch in the gutter. Fourth, set your leading at 120 to 145 percent of your font size (so 12-point text gets 14.4 to 17.4 points of leading). Fifth, keep your line length under 70 characters. Sixth, use no more than two font families in the entire book, one for body text and one for headings. These six rules will get you 90 percent of the way to a professional-looking interior, and they don’t require a design degree to implement.

Typography matters because reading is a physical act, not just a mental one. Your eyes move across the page in patterns that are influenced by the letterforms, the spacing, the margins, and a hundred other design choices. When those choices are made well, reading feels effortless. When they’re made badly, reading becomes work, and the reader might not even know why they’re struggling. The best typography is the typography you never notice, the invisible architecture that holds the reading experience together. I realize that spending an afternoon on leading sounds ridiculous. But every reader who picks up that book and thinks “this is a pleasure to read” is proving that it wasn’t ridiculous at all. It was exactly the kind of care that the work deserved.