Who Buys Fonts?
Fonts are durable, highly-reusable, compact, & high-quality software products which do not ‘bitrot’. Nevertheless, hundreds or thousands of new ones come out every year despite enormous duplication; why? I speculate that designer boredom seems to be the answer: they crave novelty.
Fonts are a rare highlight in software design—stable, with well-defined uses, highly-compatible software stacks, and long-lived. Unsurprisingly, there is now a large back-catalogue of tens or hundreds of thousands of digital fonts out there, many nigh-indistinguishable from the next in both form and function.
Why, then do they all cost so much, and who is paying for them all, and even going around commissioning more fonts?
The casualness of the highly marked-up prices & the language around commissioned fonts strongly points to designers spending client money, largely for the sake of novelty & boredom, functioning as a cross-subsidy from large corporations to the art of typography. The surplus of fonts then benefits everyone else—as long as they can sort through all the choices!
Reading up on design & web typography over the past few years for Gwern.net work, I’ve been struck by how many fonts there are, and how many new ones keep coming out. Who buys all these fonts‽
Fonts accumulate because for the most part, font technology has been quite stable and backwards-compatible so fonts do not bitrot: your TTF from 199134ya or your METAFONT file from 198441ya will probably still work; Matthew Butterick notes “it’s interesting to think that the fonts I ship to customers today could be installed and used equally well on a Windows 95 machine.” (They may not have all the features introduced since by OpenType, much less the latest fad of “variable fonts”, but they will work.) The software tools for developing fonts are often free, and digital fonts typically require no large capital investment—only time. Typography blogs could easily post a new font every day being launched commercially, and there are plenty of ‘foundries’ which despite being one-man or 2-man shops still have cranked out dozens of fonts, in addition to the endless flow of design/typography students graduating with font class projects they polish up for release, hobbyists, commissioned fonts, and so on. Someone who collects fonts can easily accumulate an archive of ~100,000 fonts1, and this has been true since fairly early on in computing history when large disc collections would distribute ‘free’ fonts (with names like “1001 Fonts”). Despite this, day in day out, new fonts continue to be released, complete with splashy landing pages and breathless descriptions reminiscent of oenophile review puffery. (I have not yet run across a new launch lauding the “oak-like notes” of a new font, but if I did—it’d probably be a “historical revival” of a woodcut font—I would not be surprised.)
Sometimes these launches are amply justified by a specific use: 8-bit computer or coding fonts, anti-forgery fonts like FE-Schrift, Matthew Carter’s Bell Centennial designed for small numbers printed on low-quality paper (or Frere-Jones’s equivalent Retina font for the WSJ) is a good use-case as it requires idiosyncratic features like ink traps, new monitors/standards enabling subpixel rendering, or the unusual design goal of “uniwidth” fonts to simplify UI design. Making an open source version of a font is a perfectly good reason, as most fonts are proprietary and specific purposes underserved by open source fonts. Countless languages are underserved and have hardly any fonts whatsoever, so it makes sense if a tech giant like Google sponsors the creation of a comprehensive multilingual font family like the Noto fonts. The staggering difficulty of creating fonts for Chinese and Japanese, which might require drawing tens of thousands of different versions to cover merely common characters (Source Han Sans has 65,535 glyphs taking up 600MB), likewise explains why there is a permanent scarcity of adequate fonts for them and the release of a new one is an event. Others are experiments or intended for an effect: a font like Clavichord is certainly striking enough, and although you would use it rarely, if you need it, it’s there (perhaps for haunted house decorations), and I’ve seen many interesting fonts be announced (Louche, Hangulatin…). Occasionally, a font is commissioned by some behemoth not because there is any need, but simply to be different—while this leads to farcical situations like “Goldman Sans” (uninteresting aside from the numbers being uniwidth across all bold/italics/roman variants) originally having a copyright license barring the user from criticizing Goldman Sachs, and bespoke fonts are a bit of a fad, the cost is also a rounding error on marketing/branding budgets, potentially useful in intellectual property ways, and removes the need to pay large-scale licensing fees for an existing font, so why not?
But those sorts of considerations explain only a small fraction of the endless drumbeat of new fonts. For most new fonts, I’m left baffled: did the world need yet another Helvetica-like sans font (get Helvetica—NOW!), or another serif font indistinguishable from Garamond? Apparently someone thought it did, and many other people thought they needed to buy it, at handsome prices like ~$123.98$1002021 but sometimes going into the $1,239.78$1,0002021 range with the occasional ‘call us’ price indicating that if you have to ask, you may be one of the large corporate customers who will pay literally millions of dollars for font licenses. (The sheer number of fonts constantly coming out means that even typography nerds playing the game of “guess that font” must often throw up their hands.) What was so wrong with the previous near-identical font? One might think there must be something terribly wrong with font design if after hundreds of Helvetica-likes that they are all still so flawed that another one must be made; as designer Massimo Vignelli snarked2, “if all people doing desktop publishing were doctors we would all be dead!”
So who is buying these other fonts? It is not hobbyists or small-time users, who blanch when they see the price, and how traffic-based licensing makes a font for their blog cost +$12,397.8$10,0002021/year; they will use an alternative (or just pirate it). They may want to buy lots of fonts—gotta catch’em all!—but they can’t afford that many at most new font price-points; hobbies have their limits, and font hobbyists don’t tend to be that well-heeled. It can’t be people who need a unique effect, because most of these fonts do nothing unique worth noting; while we needn’t go as far as the famous designer who boasted of using only 5 fonts his entire career, the universe of available fonts already vastly exceeds what will be necessary for almost everything most designers do (and is so vast that they will be familiar with only a tiny fraction of it anyway). The buying can’t be just to be different, because when even typographers struggle to see any difference, laymen certainly will not (not even subconsciously); and they could easily differentiate themselves from the majority of their competitors simply by not using a Helvetica-like sans font or bland wordmark logos! Nor can we imagine pressure from the CEO or middle-management on the designers to buy these alternative fonts, because they have such strong feelings about whether the ‘g’ should be open or closed, and are deeply knowledgeable about the proper use of tabular vs old-style numerals and the importance of true smallcaps⸮ The academic programs churning out new designers & typographers are unlikely to be major drivers of demand either, as the tuition goes to other purposes and students typically get free or discounted software for coursework—they may be expected to buy at full price during their careers in the usual “drug dealer pricing” strategy, but that is passing the buck. Big publishing houses and periodicals have established font preferences already, for many good reasons, so while they may occasionally buy a new font (and instances like the Obama presidential campaign adopting the Gotham font be a windfall), font foundries are unlikely to be launching all these fonts in the hopes of Random House buying a license.
A good example of the puzzling accumulation of fonts is the “Microsoft Design Team” announcement in April 2021 that they commissioned 5 new sans-serif fonts, to replace Microsoft’s previously-commissioned Calibri sans font used since 200718ya. What was wrong with Calibri, you might ask? Nothing, really—Microsoft owned it, and it is inoffensive-looking & reasonably well-regarded, fully-featured with the niceties like smallcaps, was designed for computer screens & continues to look fine on them, and no readers were begging for a new sans font. The new fonts are thoroughly bland-looking, indistinguishable from countless other sans fonts; they do not introduce any notable multilingual support, buzzy Variable Fonts optimizations, recherché typographic features, FLOSS licensing, experimental testing (or just drawing on said research), or anything. In fact, the announcement notes that the 5 new fonts will be A/B tested in the future—implying they were not tested before. (“Ready, fire, aim!”) But the descriptions of why, exactly, Calibri is suddenly unacceptable and must be replaced with fancy new bespoke fonts (and they don’t know which one, even) is revealing in the extent of egocentric phrases referring back to the design team (as opposed to, y’know, the readers or MS customers):
We’ve commissioned 5 new custom fonts—which should be the next default font?..Calibri has been the default font for all things Microsoft since 200718ya, when it stepped in to replace Times New Roman across Microsoft Office. It has served us all well, but we believe it’s time to evolve…we’ve interviewed the designers of each to help bring their nuances and unique personality to life…Erin & Wei: After years of Calibri—known for soft corners and narrow proportions—we were craving something very round, wide, and crisp, and the geometric genre felt like the right direction…We’ll be evaluating these 5 directions over the next few months, but we won’t be doing that alone.
The final choice was “Bierstadt”, renamed to “Aptos”:
For 15 years, our beloved Calibri was Microsoft’s default font and crown keeper of office communications, but as you know, our relationship has come to a natural end. We changed. The technology we use every day has changed. And so, our search of the perfect font for higher resolution screens began.
…The typeface was created by Steve Matteson, one of the world’s leading type designers. His previous work includes the development of the original Windows TrueType core fonts and the creation of Segoe. Steve renamed the typeface he designed from Bierstadt to Aptos after his favorite unincorporated town in Santa Cruz, California, whose widely ranging landscape and climate epitomizes the font’s versatility. The fog, beaches, redwood trees, and mountains of Aptos summed up everything that he loved about California. Getting away from digital and evoking the outdoors was akin to getting back to pencil and paper. Drawing letters by hand would play a pivotal role in Steve’s creative process. He designed the font with a slight humanist touch. He wanted Aptos to have the universal appeal of the late NPR newscaster Carl Kasell and the astute tone of The Late Show host Stephen Colbert.3
That is, the real reason is that the designers were bored and they could, and so they decided to spend millions of dollars churning the font used or read by billions of people to a new one which looks eerily like a squarer Calibri and which is indistinguishable from MS’s previous Segoe, Google Roboto, Apple San Francisco, or Helvetica.4 (Can anyone look at Reddit’s custom font and say with a straight face how unique-looking and wonderfully characteristic this new sans font is?)
We could also mention 2021 “Chirp” font, a slight teak on the nigh-indistinguishable Grilli Type’s GT America/Franklin Gothic (which “strikes the balance between messy and sharp to amplify the fun and irreverence of a Tweet, but can also carry the weight of seriousness when needed”—if you say so); the Twitter designer in charge sounded like he had fun, though, even if it came at such minor expenses as shipping broken new fonts rather than old working fonts or allowing trivial impersonation (eg. the “@linux” account) due to confusable homoglyphs like ‘l’/‘I’, an issue of which Twitter is well-aware…5 (Also illustrating the problem with change: every change—no matter how apparently minor or well-tested!—risks breaking things, and externalizing costs onto other people, especially users, thereby creating ‘bitcreep’.)
This clarifies the pricing issue as well: font pricing confirms the true customer is designers on expense accounts, with the price calculated to fall under purchasing limits, and that the buying is effectively a fringe benefit analogous to going to conferences or contributing to open source software or writing software in unusual programming languages or publishing research: everyone knows that it is a mix of personal consumption and professional utility. And the silent partner of designers are the typographers catering to the demand for novelty (but not utility), particular Monotype.6
patio11 notes something interesting about software & software-as-a-service licensing: as one raises the price from $0 to, say, $24.8$202021, sales decrease even if revenue increases; but that once one reaches a level like $123.98$1002021, sales are essentially constant up to a level around $1,239.78$1,0002021, where sales plummet and it may require active salesmanship. This is because consumers are highly price-sensitive, but corporate sales are not price-sensitive (after all, it’s not the employee’s money he’s spending), up to a ceiling at which point corporations start to require additional paperwork & managerial approval. Hence, if one has a software product (such as a font…?) which is mostly corporate users and which costs $123.98$1002021, one could double or triple the price with minimal effect on sales but doubling or tripling revenue, as long as one doesn’t go past the internal ceilings.
What these designers are buying is what a foundry sample book is selling: possibilities—sure, the new font could make the new design a big success but it probably won’t, and is there for the designer to play with and keep up with new developments and their peers, instead of getting stale using the tried-and-trues.7 The amounts are too small relative to everything else to clamp down on, doing so would be bad for morale, and it is difficult for managers or outsiders to decide what is a frivolous font purchase (after all, if they did, they’d be designers themselves). And since fonts require so little capital to develop and may only need to support one creator, who will have a whole portfolio of fonts in addition to more conventional freelance work (or who may be creating fonts at a loss for the love of typography), the sales do not need to be that large in any absolute sense either.
From this perspective, we should not be annoyed by the flood of new fonts (except as it drowns out good new fonts), and regard it as a somewhat convoluted corporate subsidy for the arts.
External Links
“Typophile Small Talk Generator v1.2”; ThisFontDoesNotExist; “Redesign Your Logo”, Lemon Demon
“OpenAI has undergone its first ever rebrand, giving fresh life to ChatGPT interactions: A new typeface, word mark, symbol and palette underpin all the ways in which OpenAI’s technology interacts with the real world”; “Walmart Introduces Updated Look and Feel: A Testament to Heritage and Innovation—The retailer’s refreshed look will embrace both evolution and legacy to better serve customers.”
Butterick again:
↩︎Fonts made 20 or more years ago are still usable on today’s machines. In fact, they make up the bulk of the Monotype and Adobe font libraries. Many remain on today’s bestseller lists. So in fonts, unlike other categories of software, type designers have a peculiar problem: our new work has to compete against decades of accumulated competitors…Furthermore, though in the past I’ve complained about the persistence of digital fonts, it veers toward one of my least favorite arguments: the idea that a creative person has a right to make a living from their work. Sorry, but no one does. The market—though artificial and imperfect—sets the rules. Other fonts exist. Hundreds of thousands, in fact. As type designers, we can either deal with that fact, or find something else to do with our time.
…First, since we’re being candid, most professional graphic and web designers don’t care much about fonts at all. (For instance: Oxford University Press, one of the most respected book publishers, nevertheless sets nearly all its books in Minion, a font that every Adobe customer has gotten free for 20 years.) Second, those that do care about fonts are looking for a low-maintenance relationship, not a lab-monkey experience.
Fuller quote:
↩︎In the new computer age, the proliferation of typefaces and type manipulations represents a new level of visual pollution threatening our culture. Out of thousands of typefaces, all we need are a few basic ones, and trash the rest.
As I said, at the time, if all people doing desktop publishing were doctors we would all be dead! Typefaces experienced an incredible explosion. The computer allowed anybody to design new typefaces and that became one of the biggest visual pollution of all times.
…I still believe that most typefaces are designed for commercial reasons, just to make money or for identity purposes. In reality the number of good typefaces is rather limited and most of the new ones are elaborations on pre-existing faces. Personally, I can get along well with a half a dozen, to which I can add another half a dozen, but probably no more.
Presumably, he carefully kerned the font to avoid having an oaky aftertaste while adding in a bit of a cheeky tartness to evoke the terroir of California.↩︎
I’m reminded of the Apple iPhone commercial which consists of some iPhone swoops against some human hands on a blank white background, and which 100% of viewers will assume to be cheap CGI; in reality, it was an entire empty white room with a large pre-programmed robot arm doing a single continuous long take using a dozen actors holding iPhones. The advertising creatives must have had fun, even if it made no difference in the end.
Similarly, video rental store clerks tended to not play the most popular movies that their customers most want to preview, but instead, what they found entertaining like cult classics or foreign movies—good for educating would-be auteurs like Quentin Tarantino, perhaps, but not for the customers.
Patrick McKenzie notes, of the prevalence of advertisements which are not aimed at ordinary consumers despite appearing to be: “For example, the captive audience ad to airline passengers at the start of every flight is very obviously targeted at retaining cabin crew, and after you see that you can’t unsee it.” That is, if you watch airline safety videos, you notice that they are too good: they are too slick, too exotic & clever, and spend too much time glamorizing the flight attendants (who are often real ones, like “Deltalina”, and not just off-the-shelf actors), at the expense of the fundamental safety information—not that it would matter if the videos were so cursory they failed at that too, given how absurdly safe flying is these days.
This makes sense if you think about it: just as retail stores want to play music that never repeats, lest their workers go insane & massacre the customers after the 10,000th play of “Jingle Bell Rock”, the flight attendants must suffer through countless replays of the same safety video—and it’s a stressful enough job as it is. Given the difficulty of retention and how few things airlines can do to make attendant working conditions better (hemmed in as they are by the iron triangle of FAA regulations, consumer preferences, and the laws of physics), making unnecessarily entertaining flight safety videos is a no-brainer.↩︎
For example, when mocking up a new ‘verified’ system for Elon Musk:
↩︎To illustrate the differences, they created mock-ups of two tweets. One, from the legacy verified @JoeBiden account, advised people to vote. The other, from the Blue-subscribed @JoeB1den account, which replaced the “i” with a “1” [because both look like ‘l’ in “Chirp”], tweeted that it was “starting nuclear war with Russia.” Besides the slight change in spelling, the two accounts could be distinguished by their colored badges.
Adobe is a familiar name & maker of fonts—you are looking at an Adobe font right now: Source Serif Pro. But one doesn’t spend long reading about digital fonts before starting to wonder about this company named Monotype. What is Monotype’s deal? The topic of Monotype’s role in the contemporary font ecosystem could fill an entire article on its own (and does).
Hardly a month goes by without them announcing an acquisition like Colophon, or a talented designer announcing they’ve sold their fonts to Monotype (and their website quickly lapsing into oblivion, bulk-redirected to the Monotype website homepage). When we are amused by Toshi Omagari’s “Comic Code” font or intrigued by his Arcade Typography history, where has he been working a dayjob? Monotype. When a website for picking fonts is bought, invariably it’s by Monotype and will now prominently advertise their web font subscriptions. When Jonathan Hoefler defrauded & stole Hoefler & Co. from his co-founder Tobias Frere-Jones, who did he sell it to a few years after settling the lawsuit? None other than Monotype. When we look at who is pushing variable fonts or announcing “Helvetica Now™” or “Futura Now™” (somehow), it’s Monotype. And when we ask who owns Monotype, it turns out, oddly enough, to be… a private equity firm. Why would a private equity firm be interested in owning Monotype, which specializes in an industry (fonts) where there are increasingly many free alternatives, which deals in an intangible digital product with unusually little copyright protection, and whose most proficient practitioners are often grateful to eek out any living at all, and which is not exactly known for being lucrative or supporting the sort of fast 7%+ annual returns that private equity firms require for their funds? How are they making money, much less, apparently (~$200m annual revenue as of 2023), so much money that they can buy everything in sight?
It’s hard to say since designers, normally so voluble, don’t seem to talk too much about Monotype itself, as opposed to its fonts or history. (This is understandable since it seems hard to find a designer who has not worked for or sold to Monotype, thinks they might, has been taught by someone who has, received a grant for a book, used Monotype historical materials, etc.) So I’ll speculate irresponsibly despite being only a writer who dabbles in web design & typography.
The answer, I gather, is that the Monotype business strategy is a standard private-equity play to monopolize an industry and then quietly raise prices to extract maximal profit while continuing to block or buyout any potential competition. Monotype has a large IP base of fonts which, once created, are pure profit, and so generates considerable cashflow; this allows them to buy up any small ‘foundry’ or rival designer (eg. Hoefler & Co., which had amassed a decent number of fonts of its own), thereby increasing its font collection but also eliminating future competitors; as it now owns all the marquee fonts, it simply becomes the default for designers, and can offer package deals. The designers do not pay for it and are price-insensitive, so Monotype can tailor its pricing to cloud/SaaS-like offerings which price-discriminate & bleed corporate clients to the pain point—just enough millions a year in licensing fees that it’s not worth the corporations commissioning a boutique knockoff font or redoing everything in a cheaper/FLOSS font. (Corporations are creates of very small brains, and do not have the spare capacity to optimize font spending when there are so many larger priorities to tackle. A CEO’s first response to any issue with fonts is to throw some money at it, as cheaper than trying to spend scarce management capacity delving into it and overriding the experts.) It can further create lockin by pseudo-novelty and churn: all those ‘new fonts’ or ‘upgrades’ or features like variable fonts no one asked for but which create excuses & novelty for bored designers. They get to buy Helvetica™ all over again, but this time, as a Variable Font™!
The dynamic is similar to commercial academic publishing, where giants like Elsevier aggregate as many journals as possible into package deals to extort university libraries into multi-million-dollar annual subscription fees; the libraries are forced to do so because if they attempt to bring any market discipline to the monopoly in question by unsubscribing, their faculty will riot. (The faculty, of course, do not care in the least about the library’s budget, and serve the role of ‘designer’ here.) Because the profit margins in the academic publishing industry are among the highest of all industries, even though (or rather, because) they do very little, they have large budgets to buy out existing journals. (Apparently the offers to journal editors can be astonishingly lucrative, and go well beyond mere junkets to exotic locale with ‘conference’ pretexts.) Should a journal hold out, they can simply create a rival to replace it; and they are notorious for creating new journals just to pad out the packages and make it more likely that there will be some journal a library ‘must’ subscribe to.
Whereas the commercial academic publishing monopolies are blatantly parasitic & harmful, I’m less sure about the Monotype monopolizing. It’s hard to tell a story of it being so bad given that there are so many cheaper alternatives not to mention rival corporations like Adobe (nor can Monotype destroy them), and it’s unclear that their competitor-busting really reduces any benefits from new fonts or font innovation. (Assuming there is any to be lost in the first place.) After all, from another perspective, Monotype’s propensity to buy out smalltimers can be seen as a generous subsidy: what’s a small-timer’s retirement plan? A Monotype acquisition. (Similar to startup dynamics, where the prospect of an exit by being acquired is a better motivation than the increasingly distant & unlikely dream of an IPO.) Who does it hurt? Large corporations, one supposes, but that’s not a very sympathetic demographic of victims, nor is it important on any kind of macroeconomic or societal scale if some Fortune 500 chooses to overpay a few million dollars for Helvetica Now!™.
So I mostly regard the Monotype situation with bemusement every time another acquisition is announced. Although it does frustrate me when they break my hyperlinks with their incompetent URL redirects.↩︎
Novelty may help explain another thing I noticed about fonts and designers in general: nobody talks as much about the value of preserving old (pre-digital) history while showing as much contempt for recent (digital) history as they do.
Reading through design blogs like BLDBLG or I Love Typography, the linkrot is extraordinary. Even the Internet Archive helps little, as so many disappeared before it could snapshot them or they blocked it or used brittle web technologies. To the extent any of it survives, it is owed to third parties like programmers or archivists.
Meanwhile, it seems like there is a Kickstarter weekly to ‘revive’ this pre-digital font or do some letterpress imitations or print a book about old Monotype lead fonts. (And naturally, these books, which never turn a profit and will never be reprinted after the first tiny print run, are unavailable anywhere online, and exist only as increasingly rare & exorbitantly expensive used books.)↩︎