How Fonts Actually Make Money

This module has nothing for you to type into a terminal, because what you take away from it is not a file but a decision: which route you will use to earn money from fonts, and what you can reasonably expect from that route. Choose wrong here and the remaining eleven modules will diligently walk you into building something nobody can buy. That is why I have opened this module up for free. If you read it and conclude it is not worth your time, I would rather you knew that before you paid.

Nobody knows what Thai type designers earn, me included

There is no publicly available income data for independent type designers, in Thailand or anywhere else. The most frequently cited survey of the type industry is Typographica’s Type Foundries Today, whose data collection finished in 2013. It records foundry names, company structures and locations, but it does not publish a single revenue figure, and as far as I can find there is no equivalent survey for Thailand at all.

Which means anyone telling you “you can make four or five figures a month selling fonts” is either speaking from personal experience or guessing. Both may be true for them. Neither is data, and neither is a promise I can make you.

Two things are verifiable: the prices Thai foundries actually list on their own websites, and the terms marketplaces publish about whether they will accept your font. This whole module deals with those two things only. The projections are yours, not mine.

”Why would anyone pay for my font when Kanit is free?”

Because the free fonts already answer “legible” completely, and answer “owning a voice nobody else has” not at all. That is the only gap money still moves through.

Start by admitting the size of the wall. Google Fonts carries 34 families supporting the Thai character set. Twenty-six of them are Cadson Demak’s work, and all of them are OFL, free for personal and commercial use alike. On top of that sit the 13 national fonts from SIPA, which a cabinet resolution of 7 September 2010 requires government agencies to install. If your client is a government agency, they already have fonts they are required to use, by cabinet resolution.

And if you are thinking of putting a font on f0nt.com in order to sell it, you cannot, under the site’s own license, which says plainly: “ไม่อนุญาตให้นำ ‘ไฟล์ฟอนต์’ ไปขาย เว้นแต่จะเป็นการแนบไฟล์ติดไปกับสื่อ โปรแกรม” — font files may not be sold, except when bundled with media or software. That place is a community for giving fonts away, not a sales channel, and getting listed there requires releasing at least one weight for free.

The answer is that nobody pays to make letters legible. They pay to make their letters look unlike their competitors’. A brand using Kanit shares its type with anybody who can click download on Google Fonts. It is the same reason companies pay for bespoke type while free fonts sit on the machine already.

High-quality free fonts raised the floor of Thai design work. They did not raise the ceiling. They cover body text, UI and documents, which need safety more than they need difference. The work that needs a specific voice — logos, shopfront signage, packaging, brand typefaces — is not being given away, because by definition it cannot be. Give it away and it stops being specific.

The last point is the one that letter-drawing courses tend to skip: a Thai font that is finished and a Thai font that is beautifully drawn are not the same thing. Microsoft states in its documentation on developing fonts for Thai script that the required OpenType features are mark and mkmk in GPOS — the Glyph Positioning table, the part of a font file that holds the rules for moving glyphs relative to one another, which is where the anchor points that pin a tone mark onto its consonant live — and recommends ccmp to decompose Sara Am so that marks land in the right place. That is not a decorative detail. It is the difference between a font that drops the tone mark on top of the upper vowel in the word “ที่” and one that does not. The volume of work genuinely differs too: Pracha Suveeranont records that Thai metal type carried more than 210 characters, where Latin had only 70.

The shortest version of this argument: you are not competing with Kanit. On Kanit’s field you lose. You compete on the field Kanit cannot enter, because it is too free to be special.

Money arrives by four routes, and they are nothing alike

There are four ways to earn from fonts, and they differ in what you have to build, who you have to find, and what the licence lets you do afterwards. People fail because they pick the most expensive route without realising it. I am not going to rank them by payout or by speed, here or anywhere else in this toolkit, because there is no data to rank them with — see the first section. What follows is what each route requires and what it forecloses.

1. Retail licensing

You finish one font family, then sell usage rights to a lot of people. This is the route everybody thinks of first, and it is the one where the whole product has to exist before anybody can buy anything. What it needs is not just a complete glyph set, but originals you hold full copyright in (not something built on top of OFL — more on that shortly), OpenType features that work correctly, a clear EULA, a specimen page, and a way to get paid. Note what that list does not contain: a client. Nobody has committed to anything at the point you start, and nobody is obliged to at the point you finish.

2. Commissions

A client pays you to make the typeface for their brand. You get paid before the market votes. This is the lowest risk of the four, because the money is locked into a contract before the work starts. What it needs is one piece of work to show (which this toolkit will take you through building) and the ability to explain to a client why mark/mkmk matter — people who can explain that in plain language are far rarer than people who draw beautiful letters.

No foundry publishes a commission rate card, so I have no figures to give you and will not invent any. The usable anchor is the retail pricing in the next section: bespoke work that the client alone owns should not come out cheaper than a retail license sold to anyone who wants one.

3. Font as lead magnet

You release a font free under OFL and sell design services. This is not selling fonts. It is using a font as a portfolio piece that distributes itself. f0nt.com forbids selling font files but permits using fonts to earn a living, so listing there fits this route exactly. Every time somebody uses your font on a sign, on a cover, in an ad, your name is inside the file and in the metadata. The thing to be clear-eyed about is that the font is not the product on this route. It is advertising. If you have no service to sell behind it, there is nothing for the advertising to point at, and releasing the font is simply a gift. And remember the one-way door: a font given away here is a font MyFonts will not accept later.

4. Bundling

Thai sellers price a set as one product, not as a stack of separate ones. DB Fonts lists the DB Heavent bundle — 4 families, 56 styles — at ฿5,885 for installation on 1-5 desktops, and Katatrad lists ฿11,800 per weight against ฿35,000 for a complete family. Those are list prices, which is all they are: neither seller publishes what they take, and I am not going to read revenue into a price tag. What the prices do show is a decision you can copy, and it is a decision about how you build, not about how much you will make — several weights off one skeleton, sold as one thing, rather than four unrelated single fonts. That is cheaper to produce and simpler to licence than four separate releases, and Modules 4 and 6 are built around it.

Real Thai market prices, taken from the sellers’ own pages

The publicly listed price range for Thai fonts is very wide, from ฿49 to ฿35,000. Where you choose to stand on that range is a business decision, not a design one.

Seller	Listed price	Scope
BEARTOR	฿49 / ฿99	Personal use / commercial use
Pocket Fonts	฿400 / ฿600 / ฿2,000 / ฿4,000 / ฿6,000	Students and teachers / individuals and freelancers / company ≤5 machines / ≤10 machines / >10 machines
Pocket Fonts (Extended)	฿1,000 / ฿2,000	Extended Pro / Extended Pro Plus+
DB Fonts	฿5,885	DB Heavent bundle, 4 families, 56 styles, desktop 1-5 machines
Katatrad	฿11,800 per weight · ฿35,000 per family	Excluding VAT — one license covers up to 20 machines within a single company

Read that table for more than its numbers. What it tells you is that the entire market sells on scope of use, not on number of files. The same customer, with the same file, pays ฿600 or ฿6,000 depending on how many machines and for what purpose. If you remember one thing from this module, remember that one. It is the difference between selling a font and giving one away with a download fee attached.

Something else the table does not say: DB Fonts takes payment by bank transfer with the customer uploading the slip. There is no automatic card processing. If a Thai foundry that actually sells fonts can work that way, you do not need a perfect payment stack before your first sale.

Whether a marketplace will accept your font is already written down

Every platform publishes its acceptance terms in advance, and those terms have already decided which of your fonts can go where. Read them before you build, not after you finish.

Platform	Share / fees	What to know first
MyFonts (Monotype)	Not published	Thailand is not on the excluded-country list · will not accept fonts given away free elsewhere · prices elsewhere must not be lower than here · submissions must be in English · display fonts need at least 186 Latin glyphs · will not accept a Sans or Serif family submitted as a single weight
Fontspring	50% (plus a 35% bonus for fonts released within 30 days)	Will not accept personal-use-only licenses · must offer an @font-face option
Creative Market	50% of net revenue	Portfolio review required first · US$20 minimum payout
Gumroad	10% + US$0.50 (direct sales) · 30% (buyer discovers you through the marketplace)	Pays into Thai bank accounts in baht · ฿600 minimum payout
Etsy	6.5% transaction fee + 6% Thai payment processing + US$0.30	Thai sellers need a Payoneer Payment Account
f0nt.com	Not a sales channel	Font files may not be sold · at least one weight must be released free

Three rows in that table wreck most people’s plans before they even start.

First, giving it away is a one-way door. Once you have released a font on f0nt.com or Google Fonts, you cannot later decide to sell it on MyFonts, because they do not accept fonts that can be downloaded free elsewhere. Decide before you release the file, not afterwards.

Second, the 186-Latin-glyph requirement means a Thai-only font cannot be sold on MyFonts. You have to design the Latin alongside it. This is why Module 6 works on making Latin that suits the Thai, rather than pasting in a borrowed skeleton.

Third, OFL closes the door on solo sales permanently. SIL Open Font License 1.1 clause 1 says “Neither the Font Software nor any of its individual components, in Original or Modified Versions, may be sold by itself.”, and clause 5 requires that modified versions be distributed entirely under the same license. Which means a font you build on top of an OFL font can never be sold on its own, no matter how much you changed.

I can say this without flinching, because it happened to me. ThaiGraph Condensed, produced with this exact pipeline, works, has 4 weights and 790 glyphs, and covers all 87 Thai codepoints — and it cannot be sold on its own, because it was built on top of an OFL font. Module 2 exists so that you do not discover this after you have finished.

How the money reaches you, and when the state starts paying attention

You can begin selling as an individual, and you must register for VAT once annual revenue passes ฿1.8 million. The current VAT rate is 7%, held in place by royal decrees renewed period by period rather than being a permanent rate.

On collecting money, ETDA’s e-Commerce survey records mobile/internet banking at 68.12% of payment channels, with cash on delivery at 7.92%. Thai customers pay through banking apps, not by card. But Stripe restricts PromptPay to legal entities registered in Thailand, so an individual freelancer cannot offer it through Stripe whatever storefront sits on top. What Gumroad’s checkout offers a Thai buyer I could not establish from Gumroad’s own published documentation, so I am not going to assert it either way — open its help centre and check before you build a plan on the answer. Module 9 says the same thing in the same words, for the same reason. If your target customers are Thai, the channel that fits their behaviour best may well be transfer-and-send-the-slip, exactly as DB Fonts does it. If they are overseas, Gumroad pays into a Thai bank account in baht, with a ฿600 minimum.

Who this toolkit is not for

I would rather you did not buy it than buy it and ask for a refund. So read these four first.

If you will not touch a command line, this is not for you. Eight of the twelve modules have commands to type into a terminal — Modules 0, 3, 4, 5, 6, 7, 8 and Appendix A. The four that do not are this one, Module 2 on licensing, Module 9 on listing your font for sale, and Appendix B, the glossary; those are the decision modules. I take you through them one line at a time on both Windows and macOS, assuming you have never opened a terminal before, but I cannot spare you them, because compiling OpenType features is the part no mouse-driven program can do for you. If you want to stay entirely inside a graphics application, look at Glyphs 4 at €319 or Glyphs Mini 2 at €49 and learn from somebody else. I mean that.

If you expect a twenty-thousand-glyph text font over a single weekend, this is not for you. Module 0 does put a genuinely working font in your hands in about 45 minutes at the keyboard — budget ninety the first time, because installing Python and getting past its usual detours is real work too, and Module 0 says so plainly. But a font you can sell is weeks of work. The 210 characters Pracha recorded are only the starting point, before you count mark, mkmk and alternate glyphs.

If you want video, this is not for you. This is a set of documents to read and to copy commands from. There is not one video clip in it. If you learn better from video, the market has options — the Font Making on Mac course on SkillLane, ฿3,990, 33 videos, 4 hours 46 minutes — but that course states itself that “คอร์สนี้จะสอนผู้เรียนสร้างฟอนต์ลายมือภาษาอังกฤษ”, it teaches making English handwriting fonts, not Thai ones.

If you want me to guarantee income, this is not for you. Go back to the first section: there is no income data to cite. The only thing I can guarantee is that you will hold font files that pass QA. What the market will pay, nobody in this country can honestly tell you.

What you should take away from this module

By the end of this module you should be able to answer these four questions in writing. Put them in one file and keep it. The remaining modules refer back to these answers.

1. Which of the four is your main route — retail, commissions, lead magnet, or bundling
2. Will your first font be OFL or commercial — if OFL, you give up solo sales permanently; if commercial, you have to design originals rather than build on somebody else’s work
3. Are your customers Thai or international — this decides both your payment channel and whether you need 186 Latin glyphs
4. What will you never give away free — because giving away is a one-way door

Answer all four and you have everything this module set out to give you. Free, no conditions.

The rest — building the actual files, compiling mark and mkmk so tone marks do not collide with vowels, running QA, packaging and listing — is in the other 11 modules of Make Thai Fonts That Sell, written to the same honesty standard as this page throughout. The same toolkit exists in Thai as ทำฟอนต์ไทยให้ขายได้; one purchase unlocks both languages, so buy whichever edition you would rather read.

If you are already a buyer and reading this inside the toolkit, you have all eleven. Next: Module 2 — Licensing: What You Can Sell and What You Cannot, which turns the OFL warning above into a decision map you can apply to any font file in under a minute.