Beyond translation: what localisation entails
Why do users keep their phones in a language they're not fluent in?
Context
A global technology company was preparing a localisation investment across regional languages in India and needed to understand a puzzling pattern first: users kept their phones set to English despite limited fluency. On the surface, this looked like a straightforward comprehension problem with an obvious fix: translate the interface, and understanding would follow.
The brief
Understand why users weren't switching their phones to a language they were more comfortable in, and use the findings to inform where localisation investment should go.
The real problem
English wasn't a comprehension choice. It was an identity choice.
We used a projection technique in focus groups, asking participants why someone like them might behave this way, rather than asking directly, to get past the polished, defensive answers a direct question tends to produce. What surfaced was that switching to Hindi or a regional language carried real social risk. It could signal something about education or class that people didn't want exposed. Staying in English, even while understanding very little of it, felt safer than the alternative.
To check whether this held beyond the group setting, we followed up with individual interviews. Group dynamics can shape what people are willing to admit publicly, so one-on-one conversations tested whether the status and judgment themes were genuine individual behaviour or something people performed for the room. They held, and in some cases, participants spoke more openly one-on-one about the shame attached to using a regional language than they had in the group.
There was a second layer underneath the language question entirely. People weren't actually reading most of the English on screen; they'd built their own working knowledge of the interface. Where to tap. What a certain icon meant. What colour signalled "yes" versus "no." They were navigating successfully through visual and positional consistency, almost entirely independent of the words in front of them.
Comprehension and usability turned out to be two different things, and the product's design was quietly relying on the second one to compensate for the first.
My Role
- Designed and led focus group research using projective techniques to surface honest answers to a sensitive question participants wouldn't answer candidly if asked directly
- Followed up focus group findings with individual interviews to validate emerging themes and reduce the risk of groupthink or socially performed answers
- Identified that language choice was driven by status and cultural aspiration, not comprehension
- Surfaced the secondary insight that users had built functional mental models of the interface independent of language literacy
- Reframed the localisation business case from "translate to increase comprehension" to "protect consistency, make language switching optional"
- Presented findings that directly informed where localisation budget was allocated
What changed
Localisation strategy
The assumption going in was that full translation would drive both comprehension and adoption. The research showed this wasn't guaranteed. If people were avoiding local-language options for identity reasons, translation alone wouldn't solve the real barrier, and risked missing the actual problem entirely.
Design principle
Because users were relying on visual and positional consistency to navigate the product, moving or restructuring familiar icons and layouts to accommodate translated text risked breaking the very thing that was actually helping them. Protecting UI consistency became a design priority, not an afterthought.
Investment decision
Rather than a blanket localisation spend, the findings redirected the team toward making language switching optional and low-stakes, avoiding investment that wouldn't have moved the needle, and would have missed the real barrier.
Considerations
Getting honest answers to a sensitive question
Asking people directly why they didn't switch languages risked defensive, socially desirable answers. Projective techniques, framing the question around a hypothetical "someone like you", created enough distance for people to speak honestly about status and judgment without having to admit it about themselves directly.
Validating sensitive findings beyond the group setting
Focus groups surfaced the status and identity themes, but group settings can shape what people are willing to admit publicly. Individual follow-up interviews tested whether these themes held one-on-one, reducing the risk that findings reflected group dynamics rather than genuine individual behaviour.
Comprehension is not the same as usability
The strongest finding wasn't about language at all. It was that people can be functionally fluent in an interface while remaining linguistically illiterate in it. This challenged an assumption embedded in the original brief: that translation was the primary lever for adoption.
Reflection
This project taught me that localisation isn't a translation problem, it's a cultural one. Language carries meaning far beyond comprehension: it signals status, education, and aspiration, and those associations can shape behaviour in ways that run completely counter to what a straightforward "does the user understand this" question would predict. Designing for a market means understanding the deeper cultural and aspirational forces shaping how people present themselves, not just the words they need translated. Get that wrong, and even a well-executed localisation effort can miss the people it's meant to serve, or spend budget solving a problem that was never really about language at all.