Tinder Jason Picallos Jun 29, 2022

How Tinder Solves Complex Lottie Localizations with Server Driven UI

Article Summary

Tinder was manually creating 40+ localized Lottie animations for every campaign. Their solution? A clever architecture that reduced it to just one.

Tinder's engagement team faced a scaling nightmare: every animated campaign needed separate Lottie files for each locale. Jason Picallos and the team engineered a server-driven solution that leverages dynamic text layers, eliminating manual duplication across their global user base.

Key Takeaways

Critical Insight

Tinder transformed Lottie localization from a manual, per-locale process into a single animation that dynamically adapts to any language through server-driven text overrides.

The team's approach to applying algorithmic thinking (time complexity) to solve a design operations problem offers a framework any mobile team can adopt.

About This Article

Problem

Tinder's engagement team had to customize each Lottie animation for every new campaign locale. After Effects engineers manually updated text layers across multiple regions, which made the process unscalable and unsustainable.

Solution

Jason Picallos' team built dynamic text layers in After Effects with a standardized naming convention. This let Insendio specify which text values should be localized while the backend and mobile clients override animation text at runtime using Lottie SDK's textProvider.

Impact

The solution achieved O(1) time complexity instead of O(n). A single animation now serves infinite localized campaigns without requiring new animation files for each locale or manual duplication work.