Getting Started with Typeface Design

the process

Getting Started with Typeface Design

In the years I’ve served type designers as their editor, I’ve noticed just how many of the challenges confronted by first-time designers are identical to the guiding principles that professional designers work vigilantly to observe. Here are some good habits to develop.

1. Know your idea.

Make sure you can articulate the brief for your project, so that you’ll have something against which to measure your success. You can’t objectively critique your own work unless you’re clear about what you’re trying to do, and developing some discipline here will pay handsome dividends should you start creating typefaces for others. (You’ll be able to shut down a nuisance conversation about whether the lowercase G is ‘nice’ by instead focussing on whether the brief for the project is sufficiently complete, and whether this letter meets the goals of that brief.) Even if it’s one you allow to evolve over time, make sure your typeface has a tight ten-second elevator pitch, so that you’ll always be able to confirm that you’re headed in the right direction.

2. Focus on systems, not letters.

In typeface design, there’s nothing inherently valuable about a really pretty letter: the goal is to have letters that give rise to handsome and enticing paragraphs. Train yourself not to look at letterforms, but at contexts: how one letter visually relates to the others, how it’s distinguished from its closest relatives, and how it cozies up to every possible neighbor. Always be looking at the larger context: judge letters by looking at words, words by looking at paragraphs, and paragraphs by looking at manuscripts.

3. Prioritize spacing over shapes.

Let the fit dictate the shape, not vice versa. If a letter always seems to be a little famished in that one spot, start by applying some ink, see if it solves the underlying spatial or textural problem, and only then decide what shape it should take. A well-fitted typeface with lumpy drawings is much more successful than a collection of impeccable drawings that don’t fit together. Take it from someone who’s made both!

4. Look!

Spend less time drawing and more time looking. One hundred percent of type design is about applying your judgment, and that kicks in not when you’re making things, but when you’re assessing the things you’ve made. (If nothing else, you’ll be more productive if you spend less time tinkering to make something perfect, and more time submitting first drafts to yourself for comment: sketching fifteen letters is much more satisfying than perfecting just one.) If you’re blocking out four hours to work on your typeface today, plan to spend two of them away from the computer, reviewing proofs at your dining room table. Or your favorite café, where engaging people might approach you with curiosity. Have that elevator pitch ready.

5. Proof with purpose.

It’s undeniably rewarding to see your embryonic font come alive on your résumé, or on a poster or a t-shirt you’re designing, but these aren’t diagnostic tools designed to help you improve your work. Definitely make time for unstructured play, because it’s motivating and healthy and fun, but also make sure that you’re rigorously testing your fonts using concise, organized font proofs. A proof is designed to help you focus on specific issues and diagnose specific problems: here’s one that I created, which you’re welcome to use.

6. Ignore the math.

Trust what you see, not what you measure. It’s hard for the rational mind to accept, but if it looks wrong, it is wrong, math be damned. Letters are processed by the human visual cortex, which through eons of evolutionary training has learned not so much to see, but to infer from clues what it thinks it’s seeing. It can’t be reasoned with, which is why we experience optical illusions, so it falls to the typeface designer to outfox the eye and give the brain what it needs. Instead of making letterforms that are correct, make ones that look correct, regardless of how the numbers shake out. You’d be amazed how many serious, conservative sans serifs have a perfectly balanced letter U in which one stroke is 220 units wide, and the other is 217.

7. Start small.

A lot of first-time designers (raising hand) have lofty ambitions about deep character sets or big families. It’s not just that these projects will exhaust your time and energy, it’s that they almost guarantee that you’ll paint yourself into a corner, owing to the many early decisions you won’t have known that you needed to make. Among the things that are likely to cause you to wholly renovate your foundation are adding an italic, a bold, an unexpected lowercase, small caps, an optical size axis, multiple figure sets, or additional script systems. You’ll learn more by making three small typefaces than one big one, so start with something manageable.

8. Beware the retread.

Even conscientious designers, who know enough not to nick someone’s data or copy their drawings, can find themselves so enamored of a published work that they strive to find some way to participate in it. This often takes the form of rewinding the development of someone else’s font and performing it again, justifying this as ‘a personal take’ on ‘an idea that’s just out there.’

I once got cornered at a party by a guy who was working on a screenplay about prehistoric creatures cloned from recovered dna, but it wasn’t at all like Jurassic Park, he explained, because he took great pains to ensure that it was different, and he was eager to itemize each and every point of difference to demonstrate that his project was totally original and non-infringing. Don’t be this guy. He’s a creep.

9. Throw things away.

Be prepared to throw almost everything away. Not just the beautiful drawings that somehow aren’t fitting in, but also the ideas that just aren’t going anywhere, no matter how attached to them you may become. You may yet find that your least successful output can be recycled or reinterpreted later, so keep a reject file. Two of the most significant typefaces I made in my thirties were nourished by cannibalizing the surviving parts of two failed experiments from my early twenties. Delight in becoming your own most ruthless editor, knowing that it will help you produce only your very best work.

10. Get some distance.

Of all the creative things I’ve done, type design is the one that’s most benefitted from getting perspective, so make it part of your process to put things down and walk away. Maybe for fifteen minutes, maybe for twenty years. Some of my favorite typefaces (including the one you’re reading) are the ones that have spent most of their lives unfinished in a drawer — and interestingly, it’s often these that are the easiest to finish. Sometimes it just takes fresh eyes to see what the impediment was; other times, the key to setting things straight will be the skill you haven’t yet developed. Either way, once you find the problem, you’ll know what to do. If you’re at an impasse, have faith that the version of yourself even one minute older will be wiser and better prepared. —JH

Popular follow-up questions:

I personally use RoboFont, but Glyphs and FontLab are also popular. Any program that lets you draw and evaluate letters in the context of one another (instead of by themselves) will probably do. There’ll likely come a time when you want to interact with your font data programmatically, building tools and jigs to automate repetitive tasks, so I’d encourage you to choose an editor that has a robust scripting language.
There are lectures, workshops, conferences, certificate programs, and advanced degrees. The Cooper Union in New York, and the Letterform Archive in San Francisco, both have good programs of workshops. Cooper Union also offers a certificate program in type design, and runs the annual Typographics conference that’s often stimulating and always pleasantly varied. If you’re looking to study typeface design, there are now a number of very good graduate programs. The two with which I’m the most familiar are the MATD program (Master of Art in Typeface Design) offered by the University of Reading in the UK, and the TypeMedia program at the Royal Academy of The Hague in the Netherlands. Both are excellent, and very different, so if you’re considering attending, seek out a former student or someone on the faculty.
I don’t have any formal training as a designer. I was interested in typography from childhood, but when it came time to consider college, I found that there weren’t any programs in typeface design — few colleges had even a single class in lettering. Because some irresistible professional opportunities presented themselves at the moment I was making my decision, I elected not to go to college.
The follow-up question I’m often asked is whether I recommend learning this way. It was a good fit for me personally, because I lived in a stimulating city with a vibrant design community, I was entrepreneurial, I do best when I’m learning at my own pace, and I enjoyed all of the creative opportunities that came with self-employment. But there’s something to be said for learning from experts, cultivating a supportive peer group, and perhaps most of all, being able to dedicate the time to focus on a project that’s free from the demands of commerce. Had there been a Master of Art in Typeface Design program back in 1988, I might have followed a very different path.
While tastes change, and some of the typefaces I was most excited about early in my career are quite different from the ones that inspire me today, I do think there are some qualities common to all great typefaces, regardless of their style.
For me, the greatest hallmark of excellence is clarity of purpose (see #1, above.) I like a typeface to unambiguously signal what it’s about, and for this to be expressed in both the large gestures and the small details. Even when a great typeface stumbles, it’s a testament to the vision at work: you can spot the flaws that don’t go with the rest of the design, because it’s clear what the rest of the design is trying to do. Even the most unconventional typefaces employ some sort of logical system (see #2), which is intuitively recognized and joyfully resonates in the mind of the reader.
As a corollary, I find it very hard to get excited about typefaces that don’t exhibit a clear purpose. If a typeface isn’t motivated by an idea, it will be guided only by the mannerisms of its designer, and very few designers — none, at the start of their careers — have so compelling a personal style to make an unusually-drawn alphabet inherently interesting. My least favorite typefaces are the ones whose intention is just ‘an alphabet, but by me;’ the phototype catalogs of the seventies are littered with such corpses.
Mechanically, a typeface needs to be well-fitted (#3, #4), which is helped by following a methodical workflow (#5). You can always tell when a designer has labored over the character drawings but not given enough thought to how the letters will come together to form words, a common hazard for first-time designers. Another trap is being doctrinaire, and becoming overly attached to ideas that just aren’t working out (#6, #9, #10, above.) A great typeface is rarely a first draft.
Something small, as I mentioned in #7, above. Try to resist being too ambitious about your character set, or your family, or exploring esoteric concepts that you’ll be better able to handle once you have a few reps. If you’re contemplating a monospace, or something that mixes script systems, or a profoundly unorthodox family structure, you might be getting ahead of yourself — and you might squander the idea of a lifetime at the moment when your skills are at their least developed.
I also recommend exploring more of your idea in sketches before you even launch the software. Every few years, a student has the exact same idea to build an alphabet on a hexagonal grid, riffing on those alphanumeric LED displays, but with angles measuring 60° instead of 90°. They always arrive with a cool S, E and G superimposed over a segmented hexagon, and are thrilled that the Q also works, but they defer confronting letters like I, T, M, K, D, V, and W, assuming they’ll figure these out later, which they never do. If there’s really a solution to these letters, you’ll find it with a pen and paper, not a font editing program. Be certain that your idea can be realized before you commit to it.
Digitizing an existing typeface, as opposed to designing a new one, can be an illuminating exercise, and is a classic way to get comfortable designing type. You’ll learn a lot by being forced to look closely at letterforms, and having to successfully render things that may be ambiguous, and you’ll discover how fitting a typeface is even more of an art than drawing letterforms. Just remember that this is a valuable exercise for learning, and does not need to end with a marketable product. I strongly encourage students who trace someone else’s work, and plan to distribute it, to ask if this is truly ‘their own’ by any meaningful definition. (See #8, above.)
Type clinics are something I enjoy doing on social media, so feel free reach out to me at @JonathanHoefler if you’re interested in doing a crit. I can’t make any promises, but if I can find the the time and have any constructive thoughts that might make for an interesting public conversation, I’ll get in touch.