--- title: "Keep client glossaries in one file so AI translations stay consistent across projects" description: "Store each client's glossary, style rules, past translation decisions, and review feedback in files so Claude applies them to every new project." canonical: "https://scalewithsearch.com/articles/ai-for-translators" date: "2026-01-28" modified: "2026-09-25" --- ## Site navigation - [Scale With Search](https://scalewithsearch.com/) - Real estate - Real estate - [Real estate](https://scalewithsearch.com/for/real-estate) - Work - Start here - [Send your brief](https://scalewithsearch.com/work#send-your-brief) - [Prepare your six-question brief](https://scalewithsearch.com/work#prepare-your-six-question-brief) - Build - [Site build, content library with SEO, signal desk](https://scalewithsearch.com/work) - For your business - Trades and home services - [Auto body and collision shops](https://scalewithsearch.com/for/auto-body-and-collision-shops) - [Foundation and home repair contractors](https://scalewithsearch.com/for/foundation-and-home-repair) - [Garage door and fencing contractors](https://scalewithsearch.com/for/garage-door-and-fencing-contractors) - [HVAC contractors](https://scalewithsearch.com/for/hvac-contractors) - [Janitorial and commercial cleaning companies](https://scalewithsearch.com/for/janitorial-and-commercial-cleaning) - [Locksmiths](https://scalewithsearch.com/for/locksmiths) - [Moving companies](https://scalewithsearch.com/for/moving-companies) - [Pest control companies](https://scalewithsearch.com/for/pest-control-companies) - [Plumbing and electrical contractors](https://scalewithsearch.com/for/plumbing-and-electrical-contractors) - [Restoration and water or fire damage companies](https://scalewithsearch.com/for/restoration-and-water-fire-damage) - [Roofing companies](https://scalewithsearch.com/for/roofing-companies) - [Towing companies](https://scalewithsearch.com/for/towing-companies) - [Tree services and landscaping companies](https://scalewithsearch.com/for/tree-services-and-landscaping) - [Solar installers](https://scalewithsearch.com/for/solar-installation) - [General contractors](https://scalewithsearch.com/for/general-contractors-and-construction) - [Paving, concrete, and flooring contractors](https://scalewithsearch.com/for/paving) - Practices and professional services - [Bookkeeping and tax practices](https://scalewithsearch.com/for/bookkeeping-and-tax-practices) - [Dental practices](https://scalewithsearch.com/for/dental-practices) - [Family and criminal defense law firms](https://scalewithsearch.com/for/family-and-criminal-defense-law-firms) - [Med spas and aesthetics practices](https://scalewithsearch.com/for/med-spas-and-aesthetics) - [Personal injury law firms](https://scalewithsearch.com/for/personal-injury-law-firms) - [Veterinary clinics](https://scalewithsearch.com/for/veterinary-clinics) - [Gyms and fitness studios](https://scalewithsearch.com/for/fitness) - [Therapy and outpatient health practices](https://scalewithsearch.com/for/therapy-and-outpatient-health) - [Medical billing companies](https://scalewithsearch.com/for/medical-billing) - [Insurance agencies](https://scalewithsearch.com/for/insurance-agencies) - [Financial advisors](https://scalewithsearch.com/for/financial-advisors) - [Property management companies](https://scalewithsearch.com/for/property-management) - [Recruiting and staffing agencies](https://scalewithsearch.com/for/recruiting-and-staffing) - [Architects and interior designers](https://scalewithsearch.com/for/architects-and-interior-designers) - [Logistics and supply chain companies](https://scalewithsearch.com/for/logistics-and-supply-chain) - Agencies, MSPs, and manufacturing - [IT and managed service providers](https://scalewithsearch.com/for/it-and-managed-service-providers) - [Machine shops and precision manufacturers](https://scalewithsearch.com/for/machine-shops-and-precision-manufacturing) - [Marketing agencies and freelancers](https://scalewithsearch.com/for/marketing-agencies-and-freelancers) - [SEO agencies and consultants](https://scalewithsearch.com/for/seo-agencies-and-consultants) - [Small manufacturers and fabricators](https://scalewithsearch.com/for/small-manufacturers-and-fabricators) - Restaurants, shops, studios, and nonprofits - [Restaurants and hospitality businesses](https://scalewithsearch.com/for/restaurants-and-hospitality) - [Retail stores and ecommerce sellers](https://scalewithsearch.com/for/retail-and-ecommerce) - [Photographers, event planners, and travel agents](https://scalewithsearch.com/for/photographers) - [Churches and nonprofits](https://scalewithsearch.com/for/churches-and-nonprofits) - [All industries](https://scalewithsearch.com/for/) - Learn - For your office - [Office job guides](https://scalewithsearch.com/guides/) - [Browser calculators](https://scalewithsearch.com/tools/) - Start here - [How it works](https://scalewithsearch.com/how-it-works) - [Free Starter Kit](https://scalewithsearch.com/kit/business-memory-starter-kit.zip) - [Synthetic specimen](https://scalewithsearch.com/specimen/working-session-specimen.zip) - Guides - [The Complete Guide to Business Memory for AI Agents](https://scalewithsearch.com/articles/business-memory-for-ai-agents-guide) - [The Complete Small-Business Guide to AI Agent Governance](https://scalewithsearch.com/articles/ai-agent-governance-guide-small-business) - [The Complete Guide to Leaving Vendor AI Memory](https://scalewithsearch.com/articles/leaving-vendor-ai-memory-guide) - Articles by cluster - [Business memory](https://scalewithsearch.com/articles/business-memory-for-ai-agents-guide) - [Agent governance](https://scalewithsearch.com/articles/ai-agent-governance-guide-small-business) - [Migration and ownership](https://scalewithsearch.com/articles/leaving-vendor-ai-memory-guide) - For machines - [llms.txt](https://scalewithsearch.com/llms.txt) - [llms-full.txt](https://scalewithsearch.com/llms-full.txt) - [Machine view](https://scalewithsearch.com/?view=machine) - Company - Evidence - [Proof](https://scalewithsearch.com/proof) - Company - [About](https://scalewithsearch.com/about) # Keep client glossaries in one file so AI translations stay consistent across projects. You keep client glossaries, hold terminology steady across projects, follow each client's style guide, and check past translations. A fresh ChatGPT chat starts without any of it. A set of plain-text files changes that. Each client's preferred terms, past translation choices, and style rules sit in a file that Claude reads before it works. Your translation memory and CAT tool stay where they are. The files add the decisions and reasons that a segment database does not hold. ## Why translation work needs a record Suppose you work with three clients, translating German into English. Client A wants "software" for *Anwendung*, never "application." Client B wants a formal tone. Client C wants a conversational one. A term in Client A's industry means something different in Client C's field. You ask ChatGPT to translate a technical document for Client A. It writes "application." You correct it. The next week brings another Client A project, and ChatGPT makes the same mistake. You need a phrase you translated for Client B last month. ChatGPT has no access to it. You search old files or translate it again, and you risk an inconsistency. A lot of review time goes to consistency work. A fresh chat does not hold a client's preferences. It cannot check past decisions or apply a pattern you already set. ## How the files work Claude Code reads a file named `CLAUDE.md` from the folder where you start it. Use it for your standing rules and point it to one folder per client: ```text ## Translation practice pair:: German to English (US) rule:: apply the client glossary before any general choice rule:: when a term has different approved renderings across clients, ask which applies clients/ client-a/ glossary.md, style.md, decisions.md, feedback.md client-b/ ... client-c/ ... ``` You start a Client A project. Claude reads Client A's files first and knows the preferred terms, the level of formality, and the earlier choices. You record a term once, and Claude applies it to every later project. You note a hard decision, and Claude cites it when a similar phrase appears. ## Build glossaries that work Give each client a glossary with four kinds of entries: preferred terms, banned terms, industry jargon, and brand or product names that stay untranslated. ```text # Client A glossary Anwendung → software (never "application"; client style guide 2.1) Schnittstelle → interface Mandant → tenant (IT sense; never "client") do_not_translate:: product names, UI labels in quotes banned:: "leverage", "utilize" source:: client termbase export, 2026.08.30 ``` Now you translate a Client B document. Claude applies Client B's glossary to each technical term, so consistency starts in the draft, not in the review. A term can collide across clients. Claude flags it: "Client A uses 'software' for *Anwendung*. In Client C projects, you used 'app.' Which applies here?" You answer once, and the answer goes into the right file. If your termbase lives in a CAT tool, export it in TBX, the TermBase eXchange format (ISO 30042). Many CAT tools read and write TBX, so one glossary can feed both the tool and the files. ## Keep style rules for each client Record each client's style preferences: level of formality, sentence length tendencies, active or passive voice, and how to treat cultural references. You start a Client A translation. Claude uses a formal tone with technical precision. You start a Client C translation. Claude shifts to a conversational register with simpler explanations. You no longer switch mental gears by hand between clients. On a project with many documents, the style stays the same across all of them, because every document draws on the same written rules. ## Record hard decisions with the reason Keep a `decisions.md` file per client for idioms, cultural references, and ambiguous terms. Record the choice and the reason: ```text 2026.07.14: "Das ist nicht mein Bier" → "That's not my concern" reason:: idiom; the literal image confuses a US business reader scope:: Client B, formal register; do not reuse for Client C ``` When a phrase has no literal equivalent, Claude checks the decisions file first. Its suggestion follows your earlier choice. For a document series, record the choices made in Part 1. When you translate Part 2, Claude reads them, and the terminology stays the same without a side-by-side comparison. ## Turn review feedback into rules Keep a `feedback.md` file per client. Record the errors you catch in review: phrases that are correct but wrong in context, and terms that the client flagged. When you review a draft, Claude checks it against those patterns. For example: "This sentence is grammatical, but Client A marked similar phrasing as 'too casual' in the May review." Client feedback then improves every later project. Record what the client prefers, and the next draft starts from it. When a rule is wrong, correct the file, not the chat. The [correction log template](/articles/ai-correction-log-template) shows how to record the old rule, the new rule, and the source. ## Keep project context Track project timelines, recurring clients, and the subject areas you have built expertise in. A new project arrives in a field you have worked in before. Claude checks your earlier projects in that field and suggests consistent terminology. This works even for a different client, where that client's glossary does not decide the term. Some clients send work every quarter. Record their cadence, typical document types, and seasonal terms. You can then plan capacity and prepare before the work arrives. ## Protect client confidentiality Many translation contracts and NDAs restrict where source text may go. The files stay on your computer, but Claude Code sends the text it reads to Anthropic's servers during a session. That includes any source document you ask it to translate. Retention and training terms depend on your plan. Read each client contract before you use an AI tool on that client's text. Some clients allow it, some require a specific approved tool, and some forbid it. Record the answer in the client folder so you check once. Keep each client in a separate folder. A draft for one client should never pick up another client's terms or content. The pattern is in [separating client context before agent work](/articles/separate-client-context-before-agent-work). ## Test the files with a finished job Choose a short document you already translated and delivered for one client. Ask Claude for a fresh translation with only that client's files loaded. Compare the draft with your delivered version, term by term. Every glossary term should match. Every recorded decision should appear where its phrase occurs. Where the draft differs, find out why. Either the file lacks a rule you applied from memory, or the rule is written too loosely to apply. Tighten the entry and run the test again. Repeat the test for each client once. It takes about 20 minutes per client and shows where each glossary is thin. ## Set up the files in an hour 1. Set up Claude Code, which needs a paid Claude plan or an Anthropic API account. 2. Create a practice folder that you can open in Obsidian, which is free to download. 3. Write `CLAUDE.md` with your language pair and standing rules. 4. Create one folder per client with a glossary, style rules, decisions, and feedback. 5. Export existing termbases to TBX and add them to the client folders. 6. Test with one short document per client, and compare the draft with the glossary. Add to the files as you work: new client preferences, choices you want to keep, and terms that caused confusion. Claude reads the relevant files every session. The files do not replace your CAT tool or translation memory. Translation memories move between tools in TMX, the Translation Memory eXchange format. The files hold what those tools lack: the reasons behind choices and the rules that span projects. Keep them in plain text so that any tool can read them. [Plain text as the durable layer](/articles/plain-text-ai-memory) explains why. When the same fix recurs, the file is the place to fix it. [Corrections that become system memory](/articles/corrections-become-system-memory) describes how one fix reaches every later draft. ---- ```text .|########||. .|########||. .|########||. |##||. .||##|. |##||. .||##|. |##||. .||##|. |#|. .|#|. |#|. .|#|. |#|. .|#|. |#| |#| |#| |#| |#| |#| .#| |#. .#| |#. .#| |#. .#. .#| .#. .#| .#. .#| |#. .#| |#. .#| |#. .#| |# ...... #| |# ...... #| |# ...... #| .# ||#########| #| .# ||#########| #| .# ||#########| #| .#. |######||######|. .#. .#. |######||######|. .#. .#. |######||######|. .#. #. .##|###|##|#|######| .# #. .##|###|##|#|######| .# #. .##|###|##|#|######| .# || |##|#||||||||||||#||#| || || |##|#||||||||||||#||#| || || |##|#||||||||||||#||#| || # |#||||||||||||||||||||#| #. # |#||||||||||||||||||||#| #. # |#||||||||||||||||||||#| #. || |#||||||||||||||||||||||#| || || |#||||||||||||||||||||||#| || || |#||||||||||||||||||||||#| || # .#||||||||||||||||||||||||#| # # .#||||||||||||||||||||||||#| # # .#||||||||||||||||||||||||#| # || ....|||#||||||##|#|||#|#||##||||....|. || || ....|||#||||||##|#|||#|#||##||||....|. || || ....|||#||||||##|#|||#|#||##||||....|. || #. . ....|# ....#||| ||| .#||| ....#. .# #. . ....|# ....#||| ||| .#||| ....#. .# #. . ....|# ....#||| ||| .#||| ....#. .# # . ||||#| .#####|| . .#| .#||| ||||# #. # . ||||#| .#####|| . .#| .#||| ||||# #. # . ||||#| .#####|| . .#| .#||| ||||# #. .| ||||| #. |#|||||. || #. |#||. .||||| || .| ||||| #. |#|||||. || #. |#||. .||||| || .| ||||| #. |#|||||. || #. |#||. .||||| || || |.... #. ....|#. |. ...|. ....|| || || |.... #. ....|#. |. ...|. ....|| || || |.... #. ....|#. |. ...|. ....|| || #. .||||||##||||||##||####|||||||#||||||#| .# #. .||||||##||||||##||####|||||||#||||||#| .# #. .||||||##||||||##||####|||||||#||||||#| .# # .||####|###########################|. # # .||####|###########################|. # # .||####|###########################|. # # .#||||||.#.|| # |. ..# #| #|||||#| # # .#||||||.#.|| # |. ..# #| #|||||#| # # .#||||||.#.|| # |. ..# #| #|||||#| # # .#|||||| .. |# ## |#| ...#|#|||#| # # .#|||||| .. |# ## |#| ...#|#|||#| # # .#|||||| .. |# ## |#| ...#|#|||#| # # .#|||#|# .# .#| #| ##|.#..#|||||#| # # .#|||#|# .# .#| #| ##|.#..#|||||#| # # .#|||#|# .# .#| #| ##|.#..#|||||#| # # .#||||||############|######|||||#| # # .#||||||############|######|||||#| # # .#||||||############|######|||||#| # # |...||#||||||#||||#|#||||||#|||||#||| |.| # # |...||#||||||#||||#|#||||||#|||||#||| |.| # # |...||#||||||#||||#|#||||||#|||||#||| |.| # #. .. ||||# .|||##|. |#| .|| || .|||#. #| # .# #. .. ||||# .|||##|. |#| .|| || .|||#. #| # .# #. .. ||||# .|||##|. |#| .|| || .|||#. #| # .# || | ...|| ...##| | #| .|. |. ##### .. .| || || | ...|| ...##| | #| .|. |. ##### .. .| || || | ...|| ...##| | #| .|. |. ##### .. .| || || ||||| || ||||#| . |. |. |#. ||||| |#| || || || ||||| || ||||#| . |. |. |#. ||||| |#| || || || ||||| || ||||#| . |. |. |#. ||||| |#| || || .# |.....#|....|#.||||.|||##.|#|....||.#||.#. #. .# |.....#|....|#.||||.|||##.|#|....||.#||.#. #. .# |.....#|....|#.||||.|||##.|#|....||.#||.#. #. #.|||||||#############################| ||| .# #.|||||||#############################| ||| .# #.|||||||#############################| ||| .# || ##||#||#||||||#||#||#||#||##. || || ##||#||#||||||#||#||#||#||##. || || ##||#||#||||||#||#||#||#||##. || # .#|||||||||#||#|||||||||||#| # # .#|||||||||#||#|||||||||||#| # # .#|||||||||#||#|||||||||||#| # || |#||||||||||||||||||||||#| || || |#||||||||||||||||||||||#| || || |#||||||||||||||||||||||#| || .# |#||||||||||||||||||||#| #. .# |#||||||||||||||||||||#| #. .# |#||||||||||||||||||||#| #. || |#||||||||||||||||||#| || || |#||||||||||||||||||#| || || |#||||||||||||||||||#| || #. .######|#|#########| .# #. .######|#|#########| .# #. .######|#|#########| .# .#. |#####||||#####| .#. .#. |#####||||#####| .#. .#. |#####||||#####| .#. |# ||########|| #. |# ||########|| #. |# ||########|| #. |# ...... #| |# ...... #| |# ...... #| |#. .#| |#. .#| |#. .#| |#. .#. |#. .#. |#. .#. .#| |#. .#| |#. .#| |#. |#| |#| |#| |#| |#| |#| .|#|. .|#|. .|#|. .|#|. .|#|. .|#|. |##||. .||##| |##||. .||##| |##||. .||##| .||########||. .||########||. .||########||. Scale With Search 2026 [scalewithsearch.com](https://scalewithsearch.com) ```