--- title: "Keep client voice files your AI reads so copy matches each brand from the first draft" description: "Record each client's voice, banned phrases, style rules, audience notes, and campaign results so AI drafts copy that fits the brand before you edit." canonical: "https://scalewithsearch.com/articles/ai-for-copywriters" 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 voice files your AI reads so copy matches each brand from the first draft. You write for several clients. Each has a distinct voice, a list of banned phrases, and messages that proved themselves. ChatGPT forgets all of it between sessions. A folder of client files fixes that. Claude Code reads the file for the client you name before it writes. It knows that client's voice, past campaign results, and style rules. You write the voice down once and correct it when the client corrects you. ## The problem: an assistant that forgets your clients Client A hates exclamation marks and prefers arguments built on data. Client B wants conversational copy with emojis and personal stories. Client C needs technical accuracy and no marketing filler. You ask ChatGPT for a Client A email. It writes generic marketing copy with three exclamation marks. You explain Client A's voice. Next week you need another email, and the explanation is gone. You want to reuse a subject line that won last month. ChatGPT has no access to that campaign. You search old documents or test a message that already proved itself. Copywriters lose hours to this. The assistant never learns what converts for each client, so it cannot build on winners or avoid repeat failures. ## How the files work Put a short `CLAUDE.md` at the root of your practice folder. Claude Code reads it when a session starts. Use it as a registry and a rule set. Put each client in a separate folder so that one voice never leaks into another: ```text copy/ CLAUDE.md registry and rules clients/client-a/ voice.md tone, banned words, sentence patterns, positioning style.md formatting and mechanics audience.md pain points, objections, customer language campaigns.md dated results: subject lines, CTAs, angles corrections.md client feedback, turned into rules ``` The rule set can be short: ```text Read only the folder of the client named in the prompt. If no client is named, ask. Do not guess. Apply corrections.md over voice.md when they conflict. Never invent a statistic, testimonial, or result. Mark any claim that needs proof. ``` You open a Client A project, and the assistant knows the voice, the preferred structure, and the last campaign results. ## Brand voice that stays consistent Each `voice.md` holds tone attributes, banned words, preferred sentence structures, and how the client talks about its product and its competitors. ```text ## Client B voice Tone: conversational, warm, first person plural Sentences: short; one idea each; stories welcome Emojis: allowed in social and email subject lines, max one Never: industry jargon; "leverage"; "solution" Product vs competitors: name the difference, never the competitor ``` You ask for Client B web copy. The assistant writes conversationally, with a personal anecdote, because the voice file says so. You skip the "make this more casual" round. You switch to a Client C project. The assistant moves to technical precision backed by data, because Client C's audience expects evidence, not enthusiasm. For a deeper treatment of voice files that do not freeze old claims, see [AI brand memory for content creation](/articles/ai-brand-memory-content-creation). ## Campaign results that build on each other Log each campaign result in `campaigns.md` with a date: subject lines that got opens, calls to action that got clicks, angles that converted. ```text ## 2026-06 spring launch email (Client A) Winner: curiosity subject line; loser: stated-benefit subject line Source: client's email platform report, exported 2026-06-20 Sample: one send, two variants; treat as a lead, not a law Next test: curiosity line with a number in it ``` You plan a new Client A email campaign. The log shows that the best subject line last time used curiosity rather than a stated benefit. The assistant proposes variations that build on that result instead of generic best practice. You write ad copy for Client B. The log shows that ads with "you" language beat ads with "we" language in last quarter's test. The assistant asks whether to apply that pattern. The log entry names the test and its date so that you can judge whether the result still applies. ## Banned phrases and corrections Write down what not to write. Client A hates "game-changing." Client B never uses industry jargon. Client C will not approve a social proof claim it cannot verify. The assistant reads the banned list before each draft, so the phrases appear far less often. Keep the search in your review anyway. The list is context, not a filter. When a client gives feedback, turn it into a rule in `corrections.md` with the date and the source email. The same revision should not happen twice. The [AI correction log template](/articles/ai-correction-log-template) gives a format that keeps the old wording, the new rule, and the reason together. ## Project context over the year Record each client's cadence: seasonal campaigns, launch patterns, and audience segments. Client A launches a product each quarter. The assistant reads past launch messaging, the positioning that worked, and the channels that drove conversions. You do not start from zero. You write Client B's holiday campaign. The assistant reads last year's holiday messaging, conversion rates by channel, and the offers that landed. ## Style rules that apply without a reminder `style.md` holds the mechanics. Does Client A use the Oxford comma? Does Client B capitalize product feature names? Does Client C prefer bulleted lists or paragraphs? The draft follows the client's formatting rules because they are in the file. When you onboard a new client, you write its style rules once, and each later project follows them. ## Audience notes that grow `audience.md` holds what you learn about each client's customers. It records the pain points that land, the objections that repeat, and the words customers use or find confusing. You write a Client C sales page. The file says their audience struggles with implementation complexity, not price. The copy handles implementation first. Client A's audience responds to return-on-investment messages. Client B's audience wants a sense of community. The assistant adjusts the message for each client without a reminder. ## Onboard a new client in an hour A new client arrives with a brand guide, three approved pieces, and a list of complaints about the last writer. Turn that material into files in this order. 1. Copy the client folder template and name it with the client ID. 2. Read the three approved pieces. Write five to eight sentence-level observations into `voice.md`: average sentence length, person, contractions, how the product is named. 3. Move every rule from the brand guide that changes wording into `voice.md`. Move every rule that changes formatting into `style.md`. 4. Turn each complaint about the last writer into a rule in `corrections.md`, dated, with the source. 5. Ask the assistant for a 100-word product description. Compare it with an approved piece. Add the differences you see as rules. The first draft after this hour is not perfect. It is close enough that the client's edits become rules instead of rewrites. ## Test for voice leakage Run three checks after setup. 1. Ask for the same product email for Client A and for Client B. Pass: the two drafts differ in the ways the voice files describe. Fail: they read alike. 2. Ask for a Client A draft and search it for Client B's signature phrases. Pass: none appear. Fail: tighten the rule that limits reading to one client folder. 3. Ask for a draft that needs a statistic. Pass: the assistant marks the gap. Fail: it invents a number. Keeping clients apart is the core discipline. The method is in [separate client context before an agent touches the work](/articles/separate-client-context-before-agent-work). ## Setup and limits Plan about an hour for the structure: install Claude Code, create the folder, and write the first client's files. Claude Code needs a paid Claude plan or an API account; the free plan does not include it ([Claude Code: Advanced setup](https://code.claude.com/docs/en/setup), checked 2026.09.25). Add to the files as you work: new preferences, campaign results, messages that converted or failed. Claude Code sends the content it reads to the model provider over the network. [Claude Code: Data usage](https://code.claude.com/docs/en/data-usage) Check your client contracts before unreleased product details or customer data enter the folder. The assistant drafts. You and the client approve. Advertising claims need proof before they run, and that check stays with you. Freelancers who also need separate briefs and handoffs per client can use the fuller pattern in [AI memory for freelancers](/articles/ai-memory-for-freelancers). ---- ```text .|########||. .|########||. .|########||. |##||. .||##|. |##||. .||##|. |##||. .||##|. |#|. .|#|. |#|. .|#|. |#|. .|#|. |#| |#| |#| |#| |#| |#| .#| |#. .#| |#. .#| |#. .#. .#| .#. .#| .#. .#| |#. .#| |#. .#| |#. .#| |# ...... #| |# ...... #| |# ...... #| .# ||#########| #| .# ||#########| #| .# ||#########| #| .#. |######||######|. .#. .#. |######||######|. .#. .#. |######||######|. .#. #. .##|###|##|#|######| .# #. .##|###|##|#|######| .# #. .##|###|##|#|######| .# || |##|#||||||||||||#||#| || || |##|#||||||||||||#||#| || || |##|#||||||||||||#||#| || # |#||||||||||||||||||||#| #. # |#||||||||||||||||||||#| #. # |#||||||||||||||||||||#| #. || |#||||||||||||||||||||||#| || || |#||||||||||||||||||||||#| || || |#||||||||||||||||||||||#| || # .#||||||||||||||||||||||||#| # # .#||||||||||||||||||||||||#| # # .#||||||||||||||||||||||||#| # || ....|||#||||||##|#|||#|#||##||||....|. || || ....|||#||||||##|#|||#|#||##||||....|. || || ....|||#||||||##|#|||#|#||##||||....|. || #. . ....|# ....#||| ||| .#||| ....#. .# #. . ....|# ....#||| ||| .#||| ....#. .# #. . ....|# ....#||| ||| .#||| ....#. .# # . ||||#| .#####|| . .#| .#||| ||||# #. # . ||||#| .#####|| . .#| .#||| ||||# #. # . ||||#| .#####|| . .#| .#||| ||||# #. .| ||||| #. |#|||||. || #. |#||. .||||| || .| ||||| #. |#|||||. || #. |#||. .||||| || .| ||||| #. |#|||||. || #. |#||. .||||| || || |.... #. ....|#. |. ...|. ....|| || || |.... #. ....|#. |. ...|. ....|| || || |.... #. ....|#. |. ...|. ....|| || #. .||||||##||||||##||####|||||||#||||||#| .# #. .||||||##||||||##||####|||||||#||||||#| .# #. .||||||##||||||##||####|||||||#||||||#| .# # .||####|###########################|. # # .||####|###########################|. # # .||####|###########################|. # # .#||||||.#.|| # |. ..# #| #|||||#| # # .#||||||.#.|| # |. ..# #| #|||||#| # # .#||||||.#.|| # |. ..# #| #|||||#| # # .#|||||| .. |# ## |#| ...#|#|||#| # # .#|||||| .. |# ## |#| ...#|#|||#| # # .#|||||| .. |# ## |#| ...#|#|||#| # # .#|||#|# .# .#| #| ##|.#..#|||||#| # # .#|||#|# .# .#| #| ##|.#..#|||||#| # # .#|||#|# .# .#| #| ##|.#..#|||||#| # # .#||||||############|######|||||#| # # .#||||||############|######|||||#| # # .#||||||############|######|||||#| # # |...||#||||||#||||#|#||||||#|||||#||| |.| # # |...||#||||||#||||#|#||||||#|||||#||| |.| # # |...||#||||||#||||#|#||||||#|||||#||| |.| # #. .. ||||# .|||##|. |#| .|| || .|||#. #| # .# #. .. ||||# .|||##|. |#| .|| || .|||#. #| # .# #. .. ||||# .|||##|. |#| .|| || .|||#. #| # .# || | ...|| ...##| | #| .|. |. ##### .. .| || || | ...|| ...##| | #| .|. |. ##### .. .| || || | ...|| ...##| | #| .|. |. ##### .. .| || || ||||| || ||||#| . |. |. |#. ||||| |#| || || || ||||| || ||||#| . |. |. |#. ||||| |#| || || || ||||| || ||||#| . |. |. |#. ||||| |#| || || .# |.....#|....|#.||||.|||##.|#|....||.#||.#. #. .# |.....#|....|#.||||.|||##.|#|....||.#||.#. #. .# |.....#|....|#.||||.|||##.|#|....||.#||.#. #. #.|||||||#############################| ||| .# #.|||||||#############################| ||| .# #.|||||||#############################| ||| .# || ##||#||#||||||#||#||#||#||##. || || ##||#||#||||||#||#||#||#||##. || || ##||#||#||||||#||#||#||#||##. || # .#|||||||||#||#|||||||||||#| # # .#|||||||||#||#|||||||||||#| # # .#|||||||||#||#|||||||||||#| # || |#||||||||||||||||||||||#| || || |#||||||||||||||||||||||#| || || |#||||||||||||||||||||||#| || .# |#||||||||||||||||||||#| #. .# |#||||||||||||||||||||#| #. .# |#||||||||||||||||||||#| #. || |#||||||||||||||||||#| || || |#||||||||||||||||||#| || || |#||||||||||||||||||#| || #. .######|#|#########| .# #. .######|#|#########| .# #. .######|#|#########| .# .#. |#####||||#####| .#. .#. |#####||||#####| .#. .#. |#####||||#####| .#. |# ||########|| #. |# ||########|| #. |# ||########|| #. |# ...... #| |# ...... #| |# ...... #| |#. .#| |#. .#| |#. .#| |#. .#. |#. .#. |#. .#. .#| |#. .#| |#. .#| |#. |#| |#| |#| |#| |#| |#| .|#|. .|#|. .|#|. .|#|. .|#|. .|#|. |##||. .||##| |##||. .||##| |##||. .||##| .||########||. .||########||. .||########||. Scale With Search 2026 [scalewithsearch.com](https://scalewithsearch.com) ```