Request an AI translation quote
Tell us what you need, in which languages, and roughly how much. We quote AI translation and localization, SEO and AI-search localization, scanned-document rebuilds, subtitles and captions, quality review of AI or MT output, AI assistant testing, and glossaries and style guides. We build every quote from your content and name the service level for each content type that has one.
Tell us about your content
How we build a quote
We build each quote from your content. These factors shape it:
- Volume and repetition. We analyze your content against your translation memory (TM) and break it down into new words, fuzzy matches, exact matches, and repetitions. The quote shows that breakdown. Repeated and previously translated text needs less work than new text.
- Content type and risk. A support macro and a product safety notice need different depth.
- Service level. AI + Check, AI + Edit, or AI + Edit + Review, chosen for each content type. Regulated and high-risk content runs at AI + Edit + Review or without AI, and the quote names the subject area it needs.
- Your glossary and TM. If you have no glossary or style guide yet, we create them with you before production starts, and the quote lists that work.
- Languages. Which languages you need, and in which directions.
- Timing. Delivery windows and release cadence. Same-day delivery is possible for suitable volumes. We confirm timing in the quote.
- Search localization. SEO and AI-search localization is quoted per page and per keyword set, by the number of pages and markets, the Arabic variants, whether the AI answer snapshot is included, and the service level.
- Scanned files. A scanned file has no word count until it is read. The rebuild is quoted per page, by page type and how readable each page is, and translation per word of the checked text.
- Video and audio. Video has no word count until it is transcribed, so subtitles and captions are quoted per minute, by what each minute needs: transcription, the language or dialect spoken, the target languages, the type of captions, the speech, the audio quality, and the service level.
- AI assistant testing. Testing has no word count, so a quote for assistant testing is built on the languages and dialects, the intents to test, how many prompts and retest rounds you need, and the service level.
- File engineering. Your formats (such as DOCX, XLSX, CSV, XML, JSON, XLIFF, HTML, or PO), tags, placeholders, and length limits. We work from your exports or inside your translation management system (TMS), including Phrase, XTM, Smartling, Crowdin, and Lokalise.
Each quote states the service level for each content type, the basis for the volume, what you receive with each delivery, and an acceptance level, if you want one.
What happens next
We’ll reply by email to confirm next steps. We may ask for sample files, your glossary, or access to your TMS.
If you haven’t worked with us yet, you can start with a free check or a sample. Your quote confirms whether the work is a one-off project or part of an ongoing program. The usual path:
- Free check, if you already use AI or machine translation (MT). Findings on up to 300 source words of AI or MT output you already have. One free check per company.
- Sample on your file. If the fit is right, we work on a defined part of your own content at the service level we propose, with that level’s deliverables.
- Pilot. Real work in your files or your TMS, under real working conditions, with the deliverables of your service level.
- Monthly contract, for an ongoing program. The service level for each content type, a volume band, delivery windows for each content type, and the same team on your account. A monthly summary reports on quality and on deliveries against those windows.
A sample tests a defined part of your content. A pilot tests delivery under real working conditions. Before either starts, we agree on the scope, deliverables, timing, and any charge.