Should I translate app UI screens for every market or keep them in English?

Translate them by default. Keep app UI screens in English only when your app's brand is already recognizable in that target market. In documented localization runs, translating and regenerating 5 app UI screens per market added no separate cost tier — it sat inside a ~$145, ~2-hour-per-ad workflow that held 100% product and text consistency.

Treat UI screens as part of the localization, not an optional extra: a full localization means new characters, new locations, a translated voiceover, AND translated in-app UI screens — not just a dubbed audio track. A winning ad in the wrong language reaches roughly 50% of your potential customers ineffectively, and English screens inside an otherwise localized ad break the native read at exactly the moment you're showing the product. Independent app-marketing research points the same direction — localized screenshots and UI materially improve install conversion versus English-only creative (AppDrift, Adjust).

The one legitimate exception: keeping UI screens in English is viable when the app brand is already recognizable in the target market. It also works as a cheap first pass if you're only testing demand in a market you haven't committed to — localize the voiceover and casting first, then translate screens once the market proves out. For any market you're committed to, translate.

How to execute it: the invideo agent — an agentic video tool with the current image and video models built in — can translate on-screen UI text and regenerate the app screens, then animate them in the same sequence order as the original ad, so the edit structure, music bed, and pacing stay untouched while only cast, location, voiceover, and UI text change. Lock assets in tiers — characters, then locations, then clips, then UI screens, then voiceover — so nothing gets regenerated twice. In one documented run, GPT-Image-2 handled the reference sheets while the screens were rebuilt in-sequence; the on-screen copy was also culturally adapted, not literally translated — the Japanese version ran "初回購入は最大45%オフ" (up to 45% off first purchase) as its CTA.

What it costs: documented localizations ran $70–$145 per ad depending on complexity — one run came to 570 credits (~$145) per localized ad including an 85% clip rejection rate; another produced 6 fully localized ads across Japan and Spain for $425 total ($70 per ad, ~2.5 hours each). The first localization takes about 2 hours; each additional market drops to roughly 1 hour, and teams have shipped 6 localizations in a single day once the workflow is locked. Translating the 5 UI screens sits inside those numbers — there is no separate cost tier for it, which removes the economic argument for keeping English screens. If you're scaling wide, name your character reference files after their target languages (Spanish.jpg, French.png) and the invideo agent assigns each to the correct market ad — one team batched 9 markets in a single pass this way.

Watch some of these to see what works for you:

Most localizations skip this. We didn't. Agent translated the UI text and regenerated all 5 app screens, then animate them in the same sequence as the original.

— invideo's creative team