// CASE 01 · AIRBNB × TRANSLATED
Redesigning translation trust, one listing at a time.
20% of Airbnb hosts wrote their listings mixing languages — confusing for guests, impossible to auto-translate through Translated, Airbnb's new machine-translation partner. We redesigned the localization experience so hosts could trust the machine.
fig 01 · The redesigned localization surface, host side.
of hosts wrote mixed-language listings
hosts tested, two rounds plus shadowing
deployed globally in the Winter Release
guest–host chat auto-translated in real time
01
Executive summary
One host in five mixed English, Spanish and German in a single description. Guests were confused; the machine-translation pipeline simply gave up. We redesigned the localization experience through usability testing, expert reviews and design strategy.
02
Four phases, one question
➔ Desk research & hypothesis framing
➔ Big Data-driven participant recruitment via internal Airbnb queries
➔ Expert review of the listing creation journey
➔ Usability testing with 16 hosts in 2 rounds + shadowing sessions
Tools & tactics
Mixed-methods UX research
Remote moderated testing
Heuristic analysis
Microcopy and flow redesign recommendations
03
Issues discovered
➔ Localization features lacked onboarding and discoverability.
➔ No alerts were provided for incorrect language input.
➔ There was no inline translation guidance or language reminders.
➔ Currency conversion wass a direct 1:1 ratio (e.g., $100 = €100).
➔ There was no preview of the final output, leading to a lack of trust in machine translation (MT).
Five design principles, one per issue
Put users in control
Guide & reassure
Explain & build trust
Prevent errors before they happen
Show and tell
04
Results
Shipped in 2021. Still compounding in 2025. A significant drop in mixed-language listings, higher machine-translation adoption, and hosts more confident in the localization flow. By 2025, guest–host chat is translated in real time on the same foundations.
05
Lessons learned
Localization is not a feature — it’s an experience
When users understand the process, they trust it
Microcopy can shape macro outcomes
Cross-functional collaboration with data science accelerates research