--- title: "Write down the business facts AI cannot guess before you ask for a draft" description: "Write your industry, offers, constraints, clients, voice, and examples into one file the AI reads, so its drafts stop sounding generic." canonical: "https://scalewithsearch.com/articles/ai-doesnt-understand-my-business" 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) # Write down the business facts AI cannot guess before you ask for a draft. You have tried ChatGPT for proposals, for email, and for strategy. The output is generic. It suggests pricing models that do not fit your market. It recommends outreach tactics you would never use. It writes in a tone that makes you sound like every other consultant who learned marketing from social media carousels. It does not understand your business, because nobody told it. ## Why the model defaults to generic A large language model learns from a very large body of public text: forum threads, blog posts, corporate sites, and academic papers. It learns common patterns, common advice, and statistical averages. It did not learn your business: your industry, your clients, your offers, your constraints, or your voice. Ask it for a proposal, and it writes what a proposal usually looks like across thousands of examples. Ask it for outreach advice, and it gives the playbook for software startups, even if you are a fractional CFO who serves construction companies. The output is not wrong. It is not yours. ## See the gap in one proposal Take a fictional example. You run a video editing service for real estate agents. You sell a fixed monthly package: four property videos a month with a 48-hour turnaround. You ask ChatGPT: "Write a proposal for a new client." The proposal includes a section on "brand storytelling" and "emotional connection with audiences." You do not do that. You edit property tours, not narratives. It presents pricing as "custom packages tailored to your needs." Your prices are fixed. There is no customization and no negotiation. It describes a 90-day onboarding timeline. Your onboarding is one video call and a shared folder link. The proposal reads well. It is professional and well structured. It does not describe how you work, so you spend thirty minutes on a rewrite. ## List what the model needs to know Your business has specifics, and the model needs them. Six categories cover most of the gap. | Category | What to write down | |---|---| | Industry context | How your market differs from the default. Real estate video differs from software demo production: different clients, deliverables, and objections. | | Offer structure | Retainer, project, or hourly. Package contents. The rule for which package to offer when. | | Constraints | What you do not do, who you do not serve, and the deal terms you refuse. | | Client profile | Who you serve, who you do not, industries, company size, and the problems you solve. | | Voice | Formal or casual, "I" or "we," jargon rules, and email length. | | Examples | Proposals that won, emails that got replies, and positioning that landed clients. | The model has none of this unless you supply it, every time. ## Why the context dump fails The usual workaround is a paragraph at the top of every prompt: "I run a video editing service for real estate agents. Four property videos a month, 48-hour turnaround, fixed package. No brand storytelling or long-form content. My clients are solo agents and small teams in suburban markets. Write a proposal for..." It works. The output improves. You now do this for every proposal, every email, and every strategy session. You are not using the AI; you are briefing it from scratch each time. The guide to [stopping re-explaining your business to AI](/articles/stop-re-explaining-your-business-to-ai) shows how to replace that paragraph with a maintained record. ## Put the facts in one file The fix is a file that holds the business facts and loads before every task. In Claude Code, the file is `CLAUDE.md`, and Claude Code reads it at the start of each session. Anthropic's page [How Claude remembers your project](https://code.claude.com/docs/en/memory) describes that load. In a chat product, put the same text in project instructions. Here is the video editor's file. The business is fictional, and the prices live in the business's own offer sheet, not in this example. ```markdown # Business profile: property video editing ## Offer - Monthly retainer for real estate agents. - Base package: 4 property videos a month, 48-hour turnaround. - Larger package: 8 videos a month. - Same-day delivery only with the rush fee. - Current prices: see offers.md (the offer sheet is the authority). ## Target client - Solo agents and small teams (2 to 5 agents) in suburban markets. - Mid-market listings. Not luxury, not commercial. ## Deliverables - 60 to 90 second property tours with music, transitions, agent branding. - No drone footage (agents supply their own). - No interviews, testimonials, or long-form content. ## Constraints - No equity deals. No trade. - No payment plans below the minimum contract value in offers.md. - No work for iBuyers or investor flippers. - No same-day requests without the rush fee. ## Voice - Casual but professional. Emails under 150 words. - Use "I", not "we". No corporate jargon. - Banned: "streamline your workflow", "elevate your brand". ## Tools - Clients upload footage to Dropbox. - Deliverables go out as Vimeo links. - Communication by email only. Invoices through Stripe. ## Objections - Price: compare the per-video cost of the retainer with a one-off editor's per-video price. - Quality: send the portfolio link with 3 example videos. - Timeline: the 48 hours start when footage is uploaded, not when the listing is mentioned. ``` The full version of a file like this runs to about 800 words. You write it once and update it when the business changes. ## See what changes in the draft You open Claude Code and type: "Write a proposal for a new agent who wants the base package." No context dump. No preamble. The file gives the model the package contents, the deliverables, the constraints, the voice, the tools, and the common objections. The draft matches your tone. It lists the right package. It answers the objections you actually hear. You edit it lightly and send it after your review. Next task: "Draft a follow-up email for a prospect who went quiet after seeing pricing." The file supplies the email style and the price objection response: the per-video comparison with one-off editors. If your file says every outreach email includes a calendar link, the draft includes it. ## Compare the file with custom instructions Custom instructions are one built-in option, and they are short. As checked on 2026.09.19, OpenAI allowed 1,500 characters for Free and Go accounts and 5,000 for paid plans. The guide to [ChatGPT project and custom instruction limits](/articles/chatgpt-custom-vs-project-instructions-limits) lists the current figures and scope. A context file has no fixed character limit, though everything it holds uses context space. You can include full examples, templates, and objection responses. Custom instructions also live inside one vendor's platform. If the vendor changes the feature, your text goes with it. A context file lives on your machine or in your vault. You control it, keep its history, and back it up. The file stays on your disk, but the text a session reads goes to the model provider for processing. Retention and training terms depend on your plan. ## Build your file in seven sections Start with these sections: 1. Who I am: name, business name, role, location, industry. 2. What I sell: services, deliverables, packages, and a pointer to the current price sheet. 3. Who I serve: target client, industries, company size, deal size, geography. 4. Constraints: what you do not do, who you do not serve, deal terms you refuse. 5. Voice: tone, style, banned phrases, sentence structure, email length. 6. Tools: CRM, communication channels, invoicing, file sharing, calendar. 7. Examples: two or three proposals that won, two or three emails that got replies, one positioning statement that landed a client. Write in plain language. You do not need prompt syntax. Explain how your business works, as you would to a new hire. Keep prices, discounts, and contract terms in their authoritative source and point to it. When a price changes, you change one file, and every draft picks up the change. The [guide to briefs and prompts](/articles/brief-vs-prompt) explains how a task brief then names which source wins in a conflict. ## Test the file before you trust it Run a cold test. Start a fresh session with the file loaded. Before you ask for a draft, ask: "List the constraints you will apply to a proposal for a new client." Compare the list with the file. Then ask for the proposal. Check every package detail, deliverable, and exclusion against the file. If the draft invents a service or misses a constraint, the file is unclear or the section is buried. Fix the file and test again. Here is a worked failure. The draft offers aerial shots as an add-on. The file says "No drone footage" under Deliverables. One of the winning proposals in the Examples section, written before that rule, mentions aerial shots. Two sections conflict, and the model followed the example. Remove the line from the example, or add a rule at the top of the file: "When an example conflicts with a rule, the rule wins." Then run the cold test again. The file passes when the constraint list matches the file line for line. The proposal must name only the base package, the 48-hour turnaround, and the exclusions the file states. ## Related: AI memory - [Document your business processes so an AI can follow them without a briefing](/articles/how-to-document-business-processes-for-ai) - [Turn each business document into a data feed your AI memory can trust](/articles/documents-as-data-feeds-ai-memory) - [Route AI context across work, business, and personal domains without mixing them](/articles/multi-domain-ai-context-architecture) ---- ```text .|########||. .|########||. .|########||. |##||. .||##|. |##||. .||##|. |##||. .||##|. |#|. .|#|. |#|. .|#|. |#|. .|#|. |#| |#| |#| |#| |#| |#| .#| |#. .#| |#. .#| |#. .#. .#| .#. .#| .#. .#| |#. .#| |#. .#| |#. .#| |# ...... #| |# ...... #| |# ...... #| .# ||#########| #| .# ||#########| #| .# ||#########| #| .#. |######||######|. .#. .#. |######||######|. .#. .#. |######||######|. .#. #. .##|###|##|#|######| .# #. .##|###|##|#|######| .# #. .##|###|##|#|######| .# || |##|#||||||||||||#||#| || || |##|#||||||||||||#||#| || || |##|#||||||||||||#||#| || # |#||||||||||||||||||||#| #. # |#||||||||||||||||||||#| #. # |#||||||||||||||||||||#| #. || |#||||||||||||||||||||||#| || || |#||||||||||||||||||||||#| || || |#||||||||||||||||||||||#| || # .#||||||||||||||||||||||||#| # # .#||||||||||||||||||||||||#| # # .#||||||||||||||||||||||||#| # || ....|||#||||||##|#|||#|#||##||||....|. || || ....|||#||||||##|#|||#|#||##||||....|. || || ....|||#||||||##|#|||#|#||##||||....|. || #. . ....|# ....#||| ||| .#||| ....#. .# #. . ....|# ....#||| ||| .#||| ....#. .# #. . ....|# ....#||| ||| .#||| ....#. .# # . ||||#| .#####|| . .#| .#||| ||||# #. # . ||||#| .#####|| . .#| .#||| ||||# #. # . ||||#| .#####|| . .#| .#||| ||||# #. .| ||||| #. |#|||||. || #. |#||. .||||| || .| ||||| #. |#|||||. || #. |#||. .||||| || .| ||||| #. |#|||||. || #. |#||. .||||| || || |.... #. ....|#. |. ...|. ....|| || || |.... #. ....|#. |. ...|. ....|| || || |.... #. ....|#. |. ...|. ....|| || #. .||||||##||||||##||####|||||||#||||||#| .# #. .||||||##||||||##||####|||||||#||||||#| .# #. .||||||##||||||##||####|||||||#||||||#| .# # .||####|###########################|. # # .||####|###########################|. # # .||####|###########################|. # # .#||||||.#.|| # |. ..# #| #|||||#| # # .#||||||.#.|| # |. ..# #| #|||||#| # # .#||||||.#.|| # |. ..# #| #|||||#| # # .#|||||| .. |# ## |#| ...#|#|||#| # # .#|||||| .. |# ## |#| ...#|#|||#| # # .#|||||| .. |# ## |#| ...#|#|||#| # # .#|||#|# .# .#| #| ##|.#..#|||||#| # # .#|||#|# .# .#| #| ##|.#..#|||||#| # # .#|||#|# .# .#| #| ##|.#..#|||||#| # # .#||||||############|######|||||#| # # .#||||||############|######|||||#| # # .#||||||############|######|||||#| # # |...||#||||||#||||#|#||||||#|||||#||| |.| # # |...||#||||||#||||#|#||||||#|||||#||| |.| # # |...||#||||||#||||#|#||||||#|||||#||| |.| # #. .. ||||# .|||##|. |#| .|| || .|||#. #| # .# #. .. ||||# .|||##|. |#| .|| || .|||#. #| # .# #. .. ||||# .|||##|. |#| .|| || .|||#. #| # .# || | ...|| ...##| | #| .|. |. ##### .. .| || || | ...|| ...##| | #| .|. |. ##### .. .| || || | ...|| ...##| | #| .|. |. ##### .. .| || || ||||| || ||||#| . |. |. |#. ||||| |#| || || || ||||| || ||||#| . |. |. |#. ||||| |#| || || || ||||| || ||||#| . |. |. |#. ||||| |#| || || .# |.....#|....|#.||||.|||##.|#|....||.#||.#. #. .# |.....#|....|#.||||.|||##.|#|....||.#||.#. #. .# |.....#|....|#.||||.|||##.|#|....||.#||.#. #. #.|||||||#############################| ||| .# #.|||||||#############################| ||| .# #.|||||||#############################| ||| .# || ##||#||#||||||#||#||#||#||##. || || ##||#||#||||||#||#||#||#||##. || || ##||#||#||||||#||#||#||#||##. || # .#|||||||||#||#|||||||||||#| # # .#|||||||||#||#|||||||||||#| # # .#|||||||||#||#|||||||||||#| # || |#||||||||||||||||||||||#| || || |#||||||||||||||||||||||#| || || |#||||||||||||||||||||||#| || .# |#||||||||||||||||||||#| #. .# |#||||||||||||||||||||#| #. .# |#||||||||||||||||||||#| #. || |#||||||||||||||||||#| || || |#||||||||||||||||||#| || || |#||||||||||||||||||#| || #. .######|#|#########| .# #. .######|#|#########| .# #. .######|#|#########| .# .#. |#####||||#####| .#. .#. |#####||||#####| .#. .#. |#####||||#####| .#. |# ||########|| #. |# ||########|| #. |# ||########|| #. |# ...... #| |# ...... #| |# ...... #| |#. .#| |#. .#| |#. .#| |#. .#. |#. .#. |#. .#. .#| |#. .#| |#. .#| |#. |#| |#| |#| |#| |#| |#| .|#|. .|#|. .|#|. .|#|. .|#|. .|#|. |##||. .||##| |##||. .||##| |##||. .||##| .||########||. .||########||. .||########||. Scale With Search 2026 [scalewithsearch.com](https://scalewithsearch.com) ```