Multilingual UX Research Interviews Through Interpreters
Run multilingual UX interviews with interpreters while preserving participant meaning, improving consent and creating defensible cross-market findings.
When an interpreter is the right research setup
Use an interpreter when a participant can best describe their experience in their preferred language and a qualified local-language moderator is not available. This makes the session accessible without asking the participant to do research in a second language.
Interpretation concerns spoken communication, during a session or afterwards. Translation concerns written material, such as consent forms, screeners, task instructions and reports. Plan for both where needed. 18F’s guidance distinguishes the two while recognising that cultural context affects each.
An interpreter enables communication; they do not make two markets directly comparable. Product expectations, institutional context and local vocabulary can change what a response means. Treat that context as evidence.
Interpreting also adds a third person to data collection. Their wording choices, clarification requests and presence affect the interview interaction and the resulting account. The Institute for Employment Research describes interpreter influence as part of the interview process, so document it as part of your method. Institute for Employment Research
For the wider choice of methods, see our guide to UX research methods.
Plan and brief the interpreter before recruitment begins
Start with the decision the research must inform. Record the target language, regional dialect, participant preference and the role the researcher needs to play. If cultural exploration, quick probing or direct rapport are central to the work, a suitably qualified local-language moderator may be the stronger setup.
Use an interpreter whose language, dialect and relevant subject knowledge fit the session. Do not assume that a friend, family member or unbriefed bilingual colleague can act as a neutral substitute. Their relationship to the participant or product may shape the conversation.
Prepare participant-facing material in the relevant language, including consent, recording explanations, screening information, incentives and task instructions. Check that consent is meaningful, rather than relying on a literal rendering. Decide how you will capture evidence if the participant declines recording.
The Department for Education advises teams to plan language services, consent, timing, question testing and analysis in advance. Department for Education guidance Send the interpreter a briefing pack before recruitment begins:
- The study purpose and decision to be informed
- Participant context, language preferences and dialect considerations
- The prototype, service flow or task materials
- The discussion guide and approved product terminology
- Sensitive subjects and wording to avoid
- Consent, recording and confidentiality arrangements
- Session logistics, escalation contact and clarification protocol
Agree boundaries before the session. The researcher moderates. The participant is the person being interviewed. The interpreter relays meaning and flags ambiguity, culturally specific terms or non-equivalence. They should not silently resolve uncertainty or answer for the participant.
Allow time for preparation, arrival, breaks and contingency cover. Consecutive interpretation adds turns to an interview, so pilot the guide with the interpreter before setting the session length. A well-structured discussion guide for user interviews gives the interpreter context without making them a co-moderator.
Moderate the translated interview without losing the participant
Speak to the participant, not the interpreter. Face or look at the participant, use their name and wait for their response. The Department for Education specifically advises researchers to face the participant in interpreter-supported interviews. Department for Education guidance
Open by introducing each person’s role in the participant’s preferred language. Confirm confidentiality, recording arrangements, the reason for pauses and how either person can request clarification. Ask whether the participant is comfortable proceeding.
Use short, direct questions. Ask one thing at a time. Avoid idioms, double questions and examples that assume a familiar payment system, product or household arrangement. For example, replace:
“What did you think about the checkout and delivery choices?”
with:
“Please tell me about choosing delivery.”
“What happened next when you paid?”
For depth interviews, use consecutive interpretation: one person completes an idea, the interpreter relays it, then the next person responds. Do not interrupt an interpreted response to recover time. A pause may reflect thinking, vocabulary checking or a sensitive topic; do not treat it as evidence of misunderstanding.
Invite the interpreter to flag an unclear reference, dialect issue or phrase without a clean equivalent. Ask the participant a plain-language follow-up instead of guessing from an English rendering.
During usability tasks, separate observation from interpreted content. Record actions, hesitations, navigation paths and recovery attempts in one field. Record the interpreted account in another. This keeps behaviour and explanation distinct during analysis.
The wider moderation discipline still applies. Read how to conduct user interviews alongside this approach.
Analyse multilingual findings as interpreted evidence
Keep source-language recordings or notes where consent and data-handling rules permit. Label an English rendering as “interpreted” or “translated”, rather than presenting it as a verbatim quote. The participant’s words, the interpreter’s rendering and the researcher’s code are separate layers of evidence.
Use this three-column evidence record for material excerpts:
| Source-language evidence | English rendering | Researcher interpretation |
|---|---|---|
| Timestamp or approved source-language note | Interpreter or translator rendering | Code, finding and limitation |
Add a translation note where a term has no direct equivalent, carries hedging or emotional emphasis, or refers to a local practice. A smooth English sentence may conceal uncertainty that matters to a product decision.
Review ambiguous or decision-critical excerpts with the interpreter or another qualified reviewer. Record whether that review changed the wording, intended meaning or research code. This creates an auditable route from a participant response to the claim shared with stakeholders.
Analyse each language or market in context before comparing patterns. Report consistent behavioural evidence where the research setup supports it. Qualify claims that depend on local terminology, norms or unresolved translation choices. Interpreter mediation affects the interview interaction, so it belongs in the method description and limitations. Institute for Employment Research
A hypothetical example: a source-language phrase is rendered as “I did not trust it”. The interpreter explains that it can also express uncertainty about a process rather than suspicion of a company. Preserve that note, check the relevant task behaviour and avoid reporting a stronger conclusion than the evidence supports.
Report the language setup, interpreter role, translation workflow and material limitations with the findings. This gives stakeholders the context needed to turn research insights into impact.
Use the Interpreter Interview Control Sheet
Use this sheet to prepare one multilingual UX research interview through an interpreter and retain a traceable route from source-language evidence to the reported finding. It makes translation decisions visible; it is not a scoring tool for languages or participants.
Filled illustrative example: Portuguese-language checkout interview
This fictional example is a planning illustration, not research evidence.
| Field | Entry |
|---|---|
| Study decision | Decide whether checkout instructions need revision before release |
| Participant language and dialect | Portuguese; confirm regional preference during recruitment |
| Roles | Researcher moderates; participant completes tasks; interpreter relays meaning and flags ambiguity |
| Interpreter credentials and scope | Professional Portuguese interpreter; briefed on digital-payment vocabulary; no product advice |
| Consent wording | “You may pause, decline a question or stop recording at any time.” Confirmed in Portuguese |
| Recording and fallback | Screen and audio recording with consent; timestamped researcher notes if declined |
| Glossary | Checkout, delivery address, payment method, confirmation |
| Question-risk checks | Avoid assumptions about payment methods; use one direct question at a time |
| Pause protocol | Interpreter says “clarification needed”; researcher pauses and asks a plain-language follow-up |
| Source evidence and confidence | Keep timestamp, interpreted rendering, translation note and confidence flag together |
| Breaks and escalation | Break after task set; named research lead handles consent or safeguarding concerns |
Blank worksheet
| Field | Your session record |
|---|---|
| Study decision | |
| Participant language, dialect and preference | |
| Interpreter credentials, language scope and subject knowledge | |
| Researcher, participant and interpreter roles | |
| Participant-facing consent and recording wording | |
| Recording method and non-recording fallback | |
| Approved glossary and product terms | |
| Tasks, question-risk checks and task vocabulary | |
| Expected cultural or contextual risks | |
| Sensitive topics and prohibited assumptions | |
| Pause and clarification protocol | |
| Breaks, contingency cover and escalation contact |
Live-session evidence log
| Question or task | Source-language timestamp or note | Interpreted rendering | Translation note or clarification | Observable behaviour | Analyst code | Confidence flag |
|---|---|---|---|---|---|---|
| High / medium / review |
Close with a short interpreter debrief. Mark high-stakes excerpts, request review where needed and attach interpretation notes to the evidence repository. This is part of research operations, not a final-stage administrative task.
Common failure modes and a final pre-session check
Literal translation can erase intent. An unbriefed bilingual colleague can introduce unrecorded assumptions. Speaking only to the interpreter can reduce participant rapport. Treating translated notes as verbatim evidence can overstate what was said. Comparing markets before contextual analysis can turn a language difference into a misleading product conclusion.
Use this go/no-go check before the session:
- Is the participant’s preferred language and dialect confirmed?
- Are consent and task materials accessible in that language?
- Has the interpreter received the briefing pack and glossary?
- Are roles and the clarification protocol agreed?
- Are extra time, a break and contingency cover planned?
- Is the capture, review and source-evidence workflow ready?
Choose a local-language moderator where exploratory depth, cultural probing or rapid follow-up matter more than centralised moderation. Choose an interpreter where the research lead needs to moderate and the language fit, roles and evidence trail have been planned.
Read usability testing versus user interviews before deciding how much task observation and interview depth the study needs.
Frequently asked questions
Should I use an interpreter or a bilingual UX researcher?
Prefer a suitably qualified local-language moderator when deep cultural probing and direct rapport are central. Use an interpreter when the research lead must moderate and the role, language fit and evidence workflow are planned.
How much longer do interpreted UX interviews take?
There is no reliable universal multiplier. Interpretation adds turns, clarification and breaks. Pilot the guide with the interpreter, then schedule from that rehearsal.
Can I quote translated interview responses in my UX report?
Yes. Label the wording as a translation, retain source evidence where consent permits and review decision-critical excerpts. Document ambiguity rather than presenting a translation as verbatim.
What should I send an interpreter before a user interview?
Send study and participant context, the guide, prototype or tasks, glossary, consent and recording plan, logistics, confidentiality expectations and the agreed clarification protocol.
About Glasgow Research — Glasgow Research helps B2B SaaS teams turn customer and market research into product decisions. Work with us.
Author
About Vadim Glazkov
Vadim Glazkov is the founder of Glasgow Research and a product research expert working with founders and B2B SaaS teams on customer interviews, JTBD, market validation, and decision-ready research.