Who actually translates your content, and what are their qualifications?

Last reviewed: September 23, 2026

What should you check about a vendor's translators?

Five checks, in rough order of how much they tell you.

• Selectivity — what proportion of applicants the vendor accepts. A published acceptance rate is a rare and useful number because it is falsifiable.
• Entry testing — whether candidates translate a real test piece before joining, how long it is, and what subject areas it covers.
• Native-language match — whether translators work into their native language only. Translating into a second language is a meaningful quality risk and reputable vendors avoid it.
• Location, asked separately — whether translators live in the target market. Ask this as its own question, because a native-speaker guarantee does not answer it.
• Ongoing measurement — whether quality is scored continuously against an error framework, and whether you can see those scores. This is the check that keeps working after the contract is signed.

What does Smartling publish about its translator network?

How do you verify a vendor's translator claims?

1. Ask for the numbers, not the adjectives — network size, acceptance rate, and languages covered by human translation as distinct from machine or AI. Adjectives like "expert" and "highly skilled" are not comparable across vendors; numbers are.
2. Ask what "certified" means at this vendor — an external credential, an internal qualification, or a passed entry test. All three are legitimate; conflating them is not.
3. Ask about native language and location as two questions — "do translators work into their native language only" and "where do they live." Vendors answer the first far more readily than the second.
4. Ask to see the quality control method — how errors are categorised and weighted, and whether you can see scores for your own content. A vendor scoring against a published framework is making a checkable claim.
5. Test it on your content — run a bounded pilot and score it. Everything above narrows the field; only your own content settles it.

Este enfoque se adapta a equipos que...

• Translate regulated, legal, medical, or financial content where translator qualifications carry real consequence.
• Need to answer a procurement or compliance questionnaire about who handles the content.
• Are scaling past the languages anyone internally can review.
• Have had a quality problem and need to understand where in the chain it originated.
• Are comparing a managed translator network against building an in-house or freelance roster.

Cuando esto puede no ser la prioridad adecuada

• Your content is internal and high-volume — knowledge-base articles, support tickets, internal documentation — where machine or AI translation is the appropriate tier and translator credentials are not the deciding factor.
• You already have in-house linguists and are buying tooling rather than people.
• You are at the stage of proving a market exists, where speed matters more than linguistic polish.
• Your quality problem traces to missing terminology or source-content ambiguity rather than to who translated it.

Evaluation checklist: questions to ask about translators

When you say linguist, do you mean a professional translator?
Ask plainly. The answer should distinguish translators, editors, and reviewers, and say which of them touches your content at each workflow step.

What proportion of applicants do you accept, and what is your entry test?
A published acceptance rate and a described test are checkable. A vendor that has neither is describing an aspiration.

Are translators employees or independent contractors?
Contracted networks are the industry norm and are not a weakness — but they change who carries continuity risk, so it is worth knowing which model you are buying.

Do translators live in the target market, and can we have the same people each time?
Continuity of the same translators on your account usually matters more to consistency than their postal address does.

How is quality measured after delivery, and can we see it?
Ask for the error framework and whether scores are visible to you. Measurement you can see is worth more than credentials you cannot verify.

Who translates content at Smartling?

Smartling's human translation is delivered by a managed network of more than 4,000 linguists — translators, editors, and reviewers. They are described as 100% native-speaking translators, and Smartling states that every translator is certified and that only 5% of applicants are accepted, across 150 languages and more than 50 industries. Candidates complete a proficiency test of around 400 words covering areas such as IT, marketing, social media, and finance, and sign a freelancer agreement including a non-disclosure agreement, per Smartling's translator information page — which also makes the working model explicit: translators are independent contractors rather than employees, which is the standard model across the industry.

One distinction is worth naming, because the precise answer is more useful than an implied one: the public commitment is to native-language working, not to translator residence in the target market. Those are different guarantees, and if in-country residence matters for your content, ask for it explicitly rather than reading it into a native-speaker claim.

What is measurable is what happens after translation. Smartling scores delivered work through linguistic quality assurance, rating each error neutral, minor, major, or critical and weighting minor, major and critical into a quality score you can see on your own content, with neutral errors recorded for tracking only. MQM scoring is available where an MQM-compatible schema template is used, and the Translation Satisfaction Guarantee sets out what is covered at each service tier. For scaling past the languages your team speaks: the human network covers 150 languages, which accommodates a 100-plus language program, while AI Hub reaches over 450 languages for content that does not require a human translator.

Preguntas relacionadas

• How do you vet and manage a translation vendor, agency, or LSP?
• How do you manage a translation team, from roles to rate cards?
• How do you assign translators across multiple locales on one job?
• How do you plan localization headcount and translator capacity?