Why will you not just tell me how many languages you cover?
Language counts are the easiest claim to publish and the hardest to check. A vendor can count a language because one linguist once passed a test in it, because a partner agency lists it, or because a machine engine supports it. All three are defensible. None of them tells you whether your Karen medical consent form gets translated and reviewed by two qualified people in the week you need it.
The count is a property of a list. What you are buying is a property of a schedule. Those two things are related the way a restaurant's menu is related to what the kitchen can cook at 8pm on a Saturday.
This is not a claim that vendors are lying. It is that the unit is wrong, and everyone — buyer and vendor — has quietly agreed to keep using it because it is easy to put in a table.
Why "we support 300+ languages" and "we can staff your pair" are different sentences
MoniSa publishes 300+ company-wide languages, including rare, indigenous and low-resource language work, drawn from a verified specialist network. Both are accurate. Neither answers a procurement question on its own, and we would rather say so than let a buyer discover it in week three.
The question a scope actually turns on is narrower: for this pair, this content type, this deadline and this review depth, is there a qualified person available now, or does someone need to be recruited and calibrated first? That answer changes by month. A number on a website cannot change by month.
Ask any vendor for the count and you will get one. Ask for the count's basis and the conversation gets interesting.
Confirm availability for each language pair
Send your actual language list — not a representative sample — and ask for one of three answers against each pair. The coverage decision matrix is the blank form for exactly this, and it is worth sending alongside the list:
The first is staffed today, meaning named qualified people are available inside your window. The second is needs a recruitment window, with the window stated in days or weeks. The third is not offered, which is a perfectly good answer and far more useful than a padded yes. For translation-service resource and process requirements, not a current availability guarantee: ISO 17100:2015 — Requirements for translation services (2015).
A vendor who can answer in that shape is describing a real roster. A vendor who cannot, or who returns a blanket yes across forty pairs including three you deliberately inserted as long shots, is describing a list.
Put it in writing before the contract, not after. A coverage claim you cannot check before signing is not coverage.
What a straight answer costs the vendor, and why that is the point
Answering pair by pair is expensive to fake and cheap to honour. It forces the vendor to look at a roster rather than a marketing page, which is the same reason a scope conversation beats a quote request, and it creates a record you can hold them to when a date slips. For translation-project preparation and service processes, not a vendor's delivery promise: ISO 17100:2015 — Requirements for translation services (2015).
It also exposes the thing most buyers find out too late: that the rare pairs on your list — the ones that made you go looking for a specialist supplier in the first place — are usually the ones being covered by a promise rather than a person.
That is why MoniSa reports the recruitment window openly rather than absorbing it silently, and why nobody new goes onto live work until a pilot batch has been reviewed and signed off. A window you know about is a schedule input. A window you are not told about is a missed launch.
Where this sits in the vendor-selection cluster
Follow these in order if you are still building the shortlist rather than checking one supplier.
- Language coverage decision matrix: Run your own list through the three-way position before you send it to anyone.
- Rare language translation services: See how coverage is scoped and reviewed once a pair is confirmed.
- Specialist network: How linguists are verified before they reach live work, which is what a recruitment window actually buys.
- Supplier evidence matrix: The wider evidence set to request when coverage is only one of several open questions.
- Transcreation vs translation for marketing: When marketing content needs transcreation instead of translation, and how to brief, run, and review it so the brand voice survives each market.
- Terminology governance for multilingual content: How to run terminology and glossary governance across multilingual content so product, brand, and compliance terms stay consistent at scale.
Record the project requirements
Use this before a contract is signed, not during delivery.
- The coverage position for every pair on your real list, in the three-way shape above
- The recruitment window in days for any pair not staffed today
- Who reviews the work, and whether the reviewer is independent of the translator
- What happens to a pair that cannot be staffed after the window closes
What a padded answer looks like
Any one of these is worth a follow-up question. Two of them together usually means the coverage line is describing a list rather than a roster.
- A blanket yes across every pair, including any long shots you inserted deliberately
- A language count offered in place of a pair-level answer
- Coverage confirmed verbally but never put in writing before signing
- A reviewer who is the same person as the translator, or unnamed
- A recruitment window that appears only after the contract starts
What to send us if you want this run on your list
Send the list and the constraints and you get the position back in the same shape, before any commercial conversation.
- Your full language list, including the pairs you expect to be difficult
- Content type and volume, with the unit you want quoting in
- The deadline, and whether it is fixed or a target
- The review depth the content requires, and any acceptance criteria you will judge against
