Built for the region, not translated for it
From today, researchers can publish surveys in Arabic with full right-to-left layout, and participants can switch the entire platform between English and Arabic from the top navigation. This was not a find-and-replace translation - question rendering, validation, and quality checks were all rebuilt to be script-aware.
What that means in practice
- •Text-quality checks understand Arabic. Gibberish detection and minimum-length rules account for Arabic script; participants writing genuine Arabic answers are never penalized by Latin-centric heuristics.
- •The attention-check library is natively bilingual. Checks injected into Arabic surveys read naturally instead of like machine output - which matters, because a confusing attention check punishes honest respondents.
- •Bilingual studies field from one link. Publish in both languages and each respondent takes the survey in theirs, with results merged in one dataset.
- •Everything around the survey is translated too - onboarding, wallet, dashboards, email notifications, withdrawal receipts, and export headers.
Why it matters for data quality
Respondents answer richer and faster in their first language. In our pre-launch pilot, Arabic-language versions of the same consumer survey produced open-text answers 31% longer than English versions answered by the same demographic. If you research the MENA market and field only in English, you are sampling the English-comfortable subset and calling it the population.
“Field in the language people think in, and the data thinks with them.”
- Tayqun product team