Cross-border matters don’t ask for translation. They ask for coding decisions made by someone with native-level fluency, in records where idiom, register, and code-switching carry the meaning. A reviewer deciding whether a Mandarin email is privileged, or whether a German contract clause is responsive, is doing the same legal work as their English-language counterpart — with the added burden that the wrong reading of a colloquialism can mislabel the document entirely. Machine translation is a useful first pass for triage. It is not a substitute for native review of anything privileged or production-grade.
That distinction sets the entire staffing problem.
The bench math
A workable language bench needs three things to be true of the same person, at once:
- Native or near-native fluency. Not high-school French, not two semesters of Mandarin. Fluency at the level where idiom and tone read correctly the first time.
- Legal context. Review experience, ideally in the same matter type. A native speaker with no privilege-review background will miss privilege.
- Geography that fits the matter. The bar admissions and time-zone coverage the matter actually requires, which a remote-first bench can quietly underweight.
Any one of these is easy to source. Two is harder. All three, across a dozen languages, on the timeline a live matter runs on — that’s the job, and it’s why language depth is the part of a bench most likely to be thin when you test it.
How fluency actually gets measured
“Fluent” on a résumé means nothing on its own. The reliable signal is a tested scale. ALTA and similar testing map to the Interagency Language Roundtable (ILR) framework the U.S. government uses, scoring speaking and writing on a defined range rather than self-report. A provider that scores its reviewers against a real scale can tell you not just who speaks a language but at what level — and can pair that with each reviewer’s English-language review record to set a baseline for overall skill.
Prior experience is the other signal that holds up. A reviewer’s track record on similar foreign-language matters predicts performance on the next one better than any credential, which is why a calibrated bench tracks language reviewers the same way it tracks everyone else: by demonstrated performance on specific work, not by the line on the résumé.
Why rates move
Foreign-language review prices differently than English review, and the drivers are worth understanding before you budget. Rate tracks scarcity: a language with few qualified legal reviewers in the relevant market commands more, and even a widely spoken language can run expensive when demand spikes and the qualified reviewers are already committed elsewhere. Some languages are simply harder to review well, which raises the rate again. And whether the matter requires bar-admitted attorneys or can be handled by credentialed non-attorney reviewers changes the number materially. None of this is avoidable; it’s just the labor market for a specialized skill, and a provider should be able to explain where your matter sits in it.
The review process
A foreign-language review runs the same arc as any managed review, with the language layer threaded through it. Set the goals first — a deposition prep reads differently than a privilege review or a patent matter. Size the team to the corpus and the depth required, splitting the work across review tiers where it makes sense so that not every document needs the most expensive reviewer. Define what each reviewer captures and how the fields are structured before anyone codes. Then deploy a team that clears all three bars — fluency, legal context, geography — for the languages in scope.
The reason to run it through a managed provider is the same reason managed review works generally: no single client can keep a deep, calibrated, multilingual bench on retainer between matters, but a provider running many matters can — and can match the right reviewers to your languages without standing up the operation from scratch each time.
For how that bench gets sourced and deployed on a live timeline, see managed review and document review staffing. For the broader service model, see managed document review.
Tell us which languages your matter needs and we’ll walk the bench depth before you commit.