You pick a typeface for your brand. It looks great in your English homepage. Eighteen months in, your business has Russian customers, or you expand to the Middle East, or you launch in Japan. Now you discover your typeface doesn't support Cyrillic, or Arabic, or Han characters. Now your brand looks one way in English and a totally different way everywhere else.
This problem is preventable. The fix is to consider script coverage when you make the typography decision, not when expansion forces it. Here's the practical guide.
The script families that matter
Roughly speaking, the major non-Latin scripts you might need:
- Cyrillic (Russian, Ukrainian, Bulgarian, Serbian, others). Covers about 250 million speakers
- Greek, 13 million speakers, but important for academic and scientific markets
- Arabic (Arabic, Urdu, Persian, others). Covers 400+ million speakers. Reads right-to-left, which is a layout consideration as well as a typography one.
- Han characters (Chinese, Japanese kanji, Korean hanja partially). Covers 1.5 billion+ speakers depending on overlap
- Kana (Japanese hiragana and katakana). Paired with kanji for Japanese
- Hangul (Korean), 75 million speakers
- Devanagari (Hindi, Marathi, Sanskrit, others), 600 million+ speakers
- Thai, Hebrew, Tamil, others. Smaller markets but important if you target them specifically
Most brands don't need all of these. Decide which markets your business might realistically operate in over the next 3-5 years. Then choose typography that supports those scripts.
How typeface families handle multiple scripts
Three approaches typeface families take:
Approach 1: Unified design across scripts. A single typeface family that includes coordinated versions for multiple scripts. The Cyrillic version was designed to feel like the Latin version. The Arabic version was designed to harmonize. Examples: Noto (Google's massive family covering nearly every script), IBM Plex (covers many scripts), Inter (Latin, Greek, Cyrillic), Roboto.
Why it works: brand consistency across markets. Your French and Russian sites can use the same family and feel like one brand.
Approach 2: Coordinated pairings. Two typeface families designed to work together, one Latin and one for another script. Examples: some brands pair a Latin typeface with a specifically chosen Arabic or Chinese typeface that visually harmonizes (different weight characteristics, but similar feel).
Why it works when unified doesn't: some scripts have such different visual logic that "unified" feels forced. Coordinated pairings let each script feel native while still looking like the same brand at a category level.
Approach 3: Different typefaces per market. Each market gets its own typography choice. The connection is at the broader brand level (colors, layout, voice), not at the typeface level.
Why it works in specific cases: when your audience in each market expects category-appropriate typography that differs significantly. Sometimes a Japanese audience expects a different feel than a Western audience would want from the same brand.
The practical checklist when picking a typeface
If you're picking brand typography now and want it to support international expansion, check these:
- Script coverage. Does the typeface support every script you might need in the next 3-5 years? Check the font's character set documentation.
- Quality across scripts. Some typefaces have excellent Latin and mediocre Cyrillic, or great Latin and weak Arabic. Test the typeface in the scripts you need before committing.
- Weight availability. Does the typeface have the weights you need in every script? Some families have 9 Latin weights and only 3 Cyrillic weights, which can break your design system.
- Hinting and rendering. Some typefaces look great in print but render poorly on screen, or vice versa. Especially important for Han characters where rendering complexity is high.
- Licensing. Make sure your license covers commercial use in every market you'll operate in.
Specific typefaces worth knowing
If you want recommendations to start from:
Noto family (Google). Designed specifically to cover every script. Free, open-source, well-engineered. The downside: visual personality is intentionally neutral (so it works across scripts), which can feel less distinctive.
IBM Plex (IBM, open-source). Covers Latin, Cyrillic, Greek, Arabic, Hebrew, Devanagari, Han, Hangul, Thai. Has more personality than Noto but still feels institutional.
Inter (Rasmus Andersson, open-source). Latin, Greek, Cyrillic. Very common in SaaS for a reason. High quality and free.
Adobe's Source families (Source Sans, Source Serif, Source Han). Source Han specifically covers CJK (Chinese, Japanese, Korean) well; pair with Source Sans for Latin.
Commercial families: Klim Type Foundry, Process Type Foundry, and other commercial foundries often have excellent multi-script families if you have budget for paid type.
RTL: the layout consideration that comes with Arabic/Hebrew
Beyond typography, supporting right-to-left languages (Arabic, Hebrew, Farsi, Urdu) means your site layout itself has to support RTL. Specifically:
- Navigation and reading order flow right-to-left
- Icons and images that have directional meaning (arrows, progress indicators) need RTL versions
- Typography needs proper Arabic kashida and ligature support
- Your CSS framework needs to handle RTL properly (most modern frameworks support `dir="rtl"`)
If you're using a modern framework (Tailwind, Material UI, Chakra), RTL support is mostly a matter of setting the direction attribute correctly. The work is more about content production than technical setup.
The decision tree
If you're making the typeface decision now:
If you're 100% sure you'll only ever operate in Latin markets: pick freely. Inter, IBM Plex, Plus Jakarta Sans, Cormorant Garamond. Any of these work without worrying about script coverage.
If you might expand to Western Europe and Russia within 3 years: pick a typeface with Latin + Greek + Cyrillic. Inter and IBM Plex both qualify.
If you might expand to the Middle East within 3 years: pick a typeface with Arabic. IBM Plex and Noto both qualify. Consider how Arabic and Latin will be paired in your designs.
If you might expand to East Asia within 3 years: pick a typeface with CJK support. Noto family is the safest choice. IBM Plex has Han Sans/Serif.
If your business model is potentially global and you can't predict where: pick from the Noto family. It covers nearly every script and gives you maximum optionality.
The retrofit problem
If you're reading this and realizing your existing typeface doesn't support the scripts you now need, you have three options:
- Switch typefaces entirely. Most disruptive. Best if you were going to refresh anyway.
- Add coordinated typefaces for the new scripts. Less disruptive. Requires careful pairing.
- Use different typefaces per market. Easiest but produces less brand cohesion across markets.
None of these are free. The cheapest path is making the right typography decision the first time, with international expansion in mind. The difference between a unified global brand and a fragmented one often comes down to a decision made in week one.
Your brand kit, ready in 10 minutes.
Five quick taps. Free preview before you pay.
Start building free →