How to launch an Arabic-first product
Most products built outside the GCC bolt Arabic on at the end. They look broken — punctuation flips, numerals collide, RTL leaks, and Arabic UX feels like a translated wireframe. To win in Saudi Arabia your product must be Arabic-first from the data model up: native RTL, locale-aware numerals, Hijri/Gregorian dates, Saudi payments, and copy written by Arabic speakers, not Google Translate.
An afterthought-Arabic product loses 60–80% of GCC users in the first session. They never come back.
An Arabic-first app that works properly in the Saudi market.
How to do it
- 01
Build Arabic from the data layer
Every translatable field is a JSON object {en, ar}. No null fallbacks. No mixed-language strings.
- 02
Native RTL, not CSS hacks
Use logical CSS properties (start/end), Tailwind RTL plugin, and dir='rtl' on <html>. Test every flow in both directions.
- 03
Localize numerals, dates, currency
Arabic-Indic vs Western digits per locale, Hijri optional, SAR formatting (٢٥٠٫٠٠ ر.س).
- 04
Pay with Saudi rails
Mada, STC Pay, Tabby, Tamara, HyperPay, Moyasar — Stripe is for international cards only.
- 05
Hire an Arabic copywriter
Real Arabic UX copy — formal but warm, Gulf-aware. Translation tools produce sterile, off-brand text.
- 06
Launch with Arabic-first marketing
Arabic ad creatives, Arabic landing page as default, WhatsApp support in Gulf dialect.
What we use
We build the app ready for the Saudi market from day one.
Start with these
Common questions
Can I launch English-only and add Arabic later?
Yes, if that fits your audience. If Arabic is planned, account for right-to-left layouts, content and testing from the start so the later language work has a clear scope.
Do you write the Arabic copy?
Yes — Gulf-native Arabic copywriting is included in our standard scope.
Hijri or Gregorian dates?
Configurable per user. Most B2B products default to Gregorian, most consumer products offer both.
What about Modern Standard Arabic vs Gulf dialect?
UI copy in clean MSA. Marketing in a warm Gulf register. AI solutions, when they exist, are tested on both.