Multilingual questionnaires, survey programming files, multi-country trackers, open-ended coding, discussion guides, transcripts and reports, delivered around your fieldwork dates.
When you are briefing countries, locking sample and booking fieldwork dates, the linguistic work has to fit the timetable, not the other way around. Talkbridge works with market research agencies, fieldwork partners and insight teams who need multilingual materials ready for programming, launch and reporting, without a long induction into how surveys are built.
This page is about that operational work: the files, the waves and the deadlines. For the methodological side of instrument adaptation, cultural equivalence and scale design, see our market research translation service.
A typical brief arrives as a mix of questionnaire text, a platform export, last wave's files and a fieldwork date that will not move. Sometimes the English master is still being signed off while country start dates are already in the diary. We can start from a near-final script and take a controlled update when the master is locked, so programming is not left waiting for a single large delivery.
We translate and localise the materials that sit around that script:
A survey is not an ordinary document. Labels, tooltips, error messages, quota fails, loop text and routing notes all have to stay in step with the script. We translate the overlay or translation file you export, keep question IDs and placeholders intact, and return a file shaped for re-import where the export allows it. If a string is too long for a mobile grid, or a scale anchor looks off in a given language, we flag it rather than forcing a literal line that will break the layout or the measure.
We do not need access to your platform to start. Send the export, the language list and any existing glossary or previous-wave files, and we work from those. If you already have a preferred variant (for example European Spanish or Latin American Spanish), say so at briefing so examples, currencies and idioms fit the sample you will field.
Trackers only work if the wording stays stable. Unless you are redesigning a question, language should not drift from one wave to the next. Small unrecorded edits accumulate, and then the time series is harder to trust. We keep terminology in a translation memory and a project glossary so later waves reuse approved wording, including scale anchors, brand mentions and product lists. When a question does change, we can show you what was edited so the analysis team is not left guessing.
The same discipline helps when one study is fielded in many countries at once. Shared items stay aligned across the language versions, while market-specific blocks, such as brands, income bands or legal wording, can be localised without rewriting the rest of the questionnaire.
Fieldwork dates are rarely flexible. Interviewer shifts, panel invites and hall tests are booked first, and the translations have to be in the script before those dates, not after. Typical throughput is around 2,000 words per translator per working day. When a launch date is fixed, we can assign more linguists in parallel and still keep a second-linguist revision under ISO 17100 in the critical path. Tell us the countries, the volume and the programming cutoff as early as you can; that is usually enough to plan capacity.
After close, the language work often continues. Open-ended answers need a consistent codeframe across markets, so a code means the same thing in every country. We can translate responses, apply your existing codeframe, or help you build one, and return a structured file that your analysis team can use. Discussion guides have to sound like a moderator, not like a translated questionnaire. Transcripts and verbatims should stay faithful to what was said, including the hesitations that matter for analysis. Reports and chart decks are translated so bilingual stakeholders can read the same story.
Assignments go to linguists from a network of over 5,000 professionals, with at least five years of professional experience. Delivery follows ISO 17100 translation and revision, not a single unchecked pass. Back-translation can be added when your methodology requires it, and is scoped in the quote.
To price a project we need the source files or a word count, the language list, the fieldwork or programming date, and whether you want extras such as back-translation or open-end coding. Talkbridge LTD was founded in 2010 and is registered in Tortola, British Virgin Islands (company no. 1664226). We do not charge VAT. For a language-specific example, see our English to Spanish survey translation page, or email [email protected].
Yes. Send an editable or structured overlay or translation file from platforms such as UNICOM Intelligence (Dimensions), Confirmit (Forsta), Decipher or Qualtrics. We return content shaped for re-import where the file format allows it.
Yes. We use translation memory and a project glossary so approved wording is reused, unless you ask us to revise a question.
Yes. We can translate and code open ends against your codeframe, or help you build one, and return a structured file for analysis.
Yes, on request, when your methodology requires it. It is scoped in the quote rather than added by default.
About 2,000 words per translator per working day. We can move faster with more linguists when the launch date is fixed, still with revision under ISO 17100.
ISO 17100 processes, a network of over 5,000 translators, a minimum of five years' professional experience, founded in 2010, Tortola, BVI, no VAT. Contact: [email protected].
Request a quotation for your next multi-country study. Email [email protected].