--- title: "ChatGPT Memory: Keep Business Context in Owned Files." description: "If ChatGPT forgets business context, keep it in a maintained source file. Test what a new chat receives, update corrections, and reuse the record." canonical: "https://scalewithsearch.com/articles/chatgpt-forgets-everything" date: "2026-01-27" 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) # Move your business context into a file so you explain it once. You spend ten minutes explaining your business to ChatGPT: the industry, the tone, the client, the format you want. The draft comes back usable. The next day you open a new chat, and most of that explanation is gone. You explain it again. This is not a fault in your account. It follows from how chat products hold context. Once you see the mechanism, the fix is a change of where your business context lives, not a better prompt. ## Know why a new chat starts nearly empty ChatGPT works inside a context window: a fixed amount of text that the model can read at one time. Each conversation runs inside its own window. When you start a new chat, the model does not receive the full text of your earlier chats. The guide on [why AI does not remember context between sessions](/articles/ai-doesnt-remember-context) explains this stateless design. OpenAI added two features that carry some context forward. Saved memories store short facts, such as "works in marketing" or "prefers short replies." Since 2025, ChatGPT can also refer to past chats on eligible plans. Both features give the model a summary that the product selects. Neither one gives it your business model, client roster, pricing structure, or the tone you built over years. The [guide to ChatGPT memory that does not apply](/articles/chatgpt-memory-not-applied-every-conversation) covers why a saved memory can also fail to show up in a given chat. Chat history and memory are different things. History is a log of past conversations that you scroll through and re-read. Memory, in the useful sense, is structured context that loads into every new session without a search. A long history does not become memory on its own. The same limit applies inside one conversation. A very long chat pushes early messages out of the window, so the model can lose a detail you gave it an hour ago. The [guide to ChatGPT forgetting in the same chat](/articles/why-chatgpt-forgets-same-chat) covers that case. ## Count what the restatement costs The cost is easy to underestimate, because each restatement is short. Work it out with your own numbers. The example below uses the figures from the original version of this article, January 2026. | Input | Example value | |---|---| | Minutes of context per conversation | 10 | | Conversations per working day | 3 | | Minutes per day | 30 | | Minutes per 5-day week | 150 | | Hours per 48-week year | 120 | At 5 to 15 minutes per conversation, the same method gives roughly 50 to 250 hours per year. The figure depends on how often you use the tool. Multiply your hours by your own hourly rate to see the cost in money. Time is the smaller loss. A model with generic context returns generic output, and you edit it into shape each time. A model that reads your frameworks, terms, and past decisions starts closer to a finished draft. That quality gap repeats in every session. ## See why the usual workarounds fall short Owners try three workarounds before they change the design. **Custom instructions.** ChatGPT keeps a short block of text that applies to every chat. As checked on September 19, 2026, OpenAI lists 1,500 characters for Free and Go accounts and 5,000 characters for paid plans. That fits an output format and a few preferences. It does not fit a client list, pricing tiers, content frameworks, and brand voice rules together. **Copy and paste.** Some owners keep a context document and paste it into each important chat. This works, but it is manual data entry several times a day. The document also goes stale, because an update to it is one more chore. **Custom GPTs and Projects.** A custom GPT holds longer instructions and uploaded files. A Project keeps files and instructions for one body of work. Both take time to build and time to update, and both still hit limits on how much knowledge they hold. When a fact changes, you edit it in each GPT or Project that holds a copy. Each workaround treats the symptom. The context still lives inside the product, in a form that you rebuild by hand. ## Move the context into a file an agent reads The durable fix moves your business context into files on storage that you control. An agent then reads the files at the start of each session. Claude Code works this way. You point it at a folder on your computer, and the folder's files become available as context. One Markdown file, `CLAUDE.md`, states who you are, what you do, and how you work. Claude Code reads it at the start of every session. There is nothing to paste and nothing to restate. You can use the same file with ChatGPT. Attach it to a Project or paste it at the start of a chat. That still takes a manual step, but the file stays the single source, so an update happens in one place. The [CLAUDE.md template for business context](/articles/claude-md-template-business-context) gives a fuller structure. ## Inspect a starter `CLAUDE.md` A first version can be short. Every value below is synthetic. ```markdown # Business context ## Who I am Owner of a two-person bookkeeping practice. Clients are restaurants and retail shops with 5 to 40 staff. ## How I write Plain, direct, no jargon. Short paragraphs. No exclamation marks. ## Current clients - Johnson Hardware: monthly close, reports due on the 5th. - Rivera Cafe: payroll setup in progress, owner prefers phone. ## Standard outputs - Proposal: problem, scope, timeline, next step. One page. - Client email: three short paragraphs, one clear request. ## Terms - "Close" means the month-end close, not a sale. ``` The file answers the questions you used to answer in every chat. It also records terms that the model would otherwise read the wrong way. ## See what changes in daily requests With the file in place, a request can refer to your business without an explanation. "Draft an email to the Johnson account" works, because the agent reads who Johnson Hardware is, what the project involves, and how you write to clients. "Write this in my voice" gets closer, because the agent reads your voice rules and, if you add them, samples of your own writing. "Create a proposal using our standard structure" returns the structure you defined, not a generic template. Conversations become continuations instead of cold starts. The [guide to re-explaining your business to AI](/articles/stop-re-explaining-your-business-to-ai) goes further into which facts belong in the file and which belong in a single task request. ## Set it up in one sitting You do not need programming skills. You need a text editor, Claude Code, and about one working session. 1. Install a Markdown editor such as Obsidian, or use any plain text editor. 2. Create a folder for your business context. 3. Create `CLAUDE.md` in that folder. 4. Write who you are, your domains, and your preferred output formats. 5. Add your key clients and projects, one line each. 6. Add your frameworks, terms, and voice rules. 7. Install Claude Code and start it from that folder. Claude Code needs a paid Claude plan. 8. Ask a question that needs the file, and check that the answer uses it. Keep secrets, passwords, and regulated client records out of the file. Claude Code sends the file's contents to the model in each session, so your plan's data terms apply. ## Keep the file current The file does not maintain itself. It stays useful because you edit it when the business changes. Add a client when you sign one. Remove a project when it closes. Change a price when the price changes. Here is one update from the starter file. The payroll setup for Rivera Cafe finishes. Change its line to "Rivera Cafe: payroll live since 2026.09.01, monthly review on the 10th." The next session drafts the monthly review email without a reminder, and it no longer offers setup help that the client does not need. Each edit reaches every later session. Notes that you add to the folder become available to the agent, and an updated client document replaces the old one. The work changes. You no longer restate context in each chat; you keep one record accurate. ## Related: AI memory - [Fix the context, not the prompt, so ChatGPT gives consistent answers](/articles/chatgpt-gives-different-answers) - [Trace what ChatGPT forgot and store it where it stays](/articles/chatgpt-keeps-forgetting-what-i-told-it) - [Identify which ChatGPT memory limit you hit and work around each one](/articles/chatgpt-memory-limit) ---- ```text .|########||. .|########||. .|########||. |##||. .||##|. |##||. .||##|. |##||. .||##|. |#|. .|#|. |#|. .|#|. |#|. .|#|. |#| |#| |#| |#| |#| |#| .#| |#. .#| |#. .#| |#. .#. .#| .#. .#| .#. .#| |#. .#| |#. .#| |#. .#| |# ...... #| |# ...... #| |# ...... #| .# ||#########| #| .# ||#########| #| .# ||#########| #| .#. |######||######|. .#. .#. |######||######|. .#. .#. |######||######|. .#. #. .##|###|##|#|######| .# #. .##|###|##|#|######| .# #. .##|###|##|#|######| .# || |##|#||||||||||||#||#| || || |##|#||||||||||||#||#| || || |##|#||||||||||||#||#| || # |#||||||||||||||||||||#| #. # |#||||||||||||||||||||#| #. # |#||||||||||||||||||||#| #. || |#||||||||||||||||||||||#| || || |#||||||||||||||||||||||#| || || |#||||||||||||||||||||||#| || # .#||||||||||||||||||||||||#| # # .#||||||||||||||||||||||||#| # # .#||||||||||||||||||||||||#| # || ....|||#||||||##|#|||#|#||##||||....|. || || ....|||#||||||##|#|||#|#||##||||....|. || || ....|||#||||||##|#|||#|#||##||||....|. || #. . ....|# ....#||| ||| .#||| ....#. .# #. . ....|# ....#||| ||| .#||| ....#. .# #. . ....|# ....#||| ||| .#||| ....#. .# # . ||||#| .#####|| . .#| .#||| ||||# #. # . ||||#| .#####|| . .#| .#||| ||||# #. # . ||||#| .#####|| . .#| .#||| ||||# #. .| ||||| #. |#|||||. || #. |#||. .||||| || .| ||||| #. |#|||||. || #. |#||. .||||| || .| ||||| #. |#|||||. || #. |#||. .||||| || || |.... #. ....|#. |. ...|. ....|| || || |.... #. ....|#. |. ...|. ....|| || || |.... #. ....|#. |. ...|. ....|| || #. .||||||##||||||##||####|||||||#||||||#| .# #. .||||||##||||||##||####|||||||#||||||#| .# #. .||||||##||||||##||####|||||||#||||||#| .# # .||####|###########################|. # # .||####|###########################|. # # .||####|###########################|. # # .#||||||.#.|| # |. ..# #| #|||||#| # # .#||||||.#.|| # |. ..# #| #|||||#| # # .#||||||.#.|| # |. ..# #| #|||||#| # # .#|||||| .. |# ## |#| ...#|#|||#| # # .#|||||| .. |# ## |#| ...#|#|||#| # # .#|||||| .. |# ## |#| ...#|#|||#| # # .#|||#|# .# .#| #| ##|.#..#|||||#| # # .#|||#|# .# .#| #| ##|.#..#|||||#| # # .#|||#|# .# .#| #| ##|.#..#|||||#| # # .#||||||############|######|||||#| # # .#||||||############|######|||||#| # # .#||||||############|######|||||#| # # |...||#||||||#||||#|#||||||#|||||#||| |.| # # |...||#||||||#||||#|#||||||#|||||#||| |.| # # |...||#||||||#||||#|#||||||#|||||#||| |.| # #. .. ||||# .|||##|. |#| .|| || .|||#. #| # .# #. .. ||||# .|||##|. |#| .|| || .|||#. #| # .# #. .. ||||# .|||##|. |#| .|| || .|||#. #| # .# || | ...|| ...##| | #| .|. |. ##### .. .| || || | ...|| ...##| | #| .|. |. ##### .. .| || || | ...|| ...##| | #| .|. |. ##### .. .| || || ||||| || ||||#| . |. |. |#. ||||| |#| || || || ||||| || ||||#| . |. |. |#. ||||| |#| || || || ||||| || ||||#| . |. |. |#. ||||| |#| || || .# |.....#|....|#.||||.|||##.|#|....||.#||.#. #. .# |.....#|....|#.||||.|||##.|#|....||.#||.#. #. .# |.....#|....|#.||||.|||##.|#|....||.#||.#. #. #.|||||||#############################| ||| .# #.|||||||#############################| ||| .# #.|||||||#############################| ||| .# || ##||#||#||||||#||#||#||#||##. || || ##||#||#||||||#||#||#||#||##. || || ##||#||#||||||#||#||#||#||##. || # .#|||||||||#||#|||||||||||#| # # .#|||||||||#||#|||||||||||#| # # .#|||||||||#||#|||||||||||#| # || |#||||||||||||||||||||||#| || || |#||||||||||||||||||||||#| || || |#||||||||||||||||||||||#| || .# |#||||||||||||||||||||#| #. .# |#||||||||||||||||||||#| #. .# |#||||||||||||||||||||#| #. || |#||||||||||||||||||#| || || |#||||||||||||||||||#| || || |#||||||||||||||||||#| || #. .######|#|#########| .# #. .######|#|#########| .# #. .######|#|#########| .# .#. |#####||||#####| .#. .#. |#####||||#####| .#. .#. |#####||||#####| .#. |# ||########|| #. |# ||########|| #. |# ||########|| #. |# ...... #| |# ...... #| |# ...... #| |#. .#| |#. .#| |#. .#| |#. .#. |#. .#. |#. .#. .#| |#. .#| |#. .#| |#. |#| |#| |#| |#| |#| |#| .|#|. .|#|. .|#|. .|#|. .|#|. .|#|. |##||. .||##| |##||. .||##| |##||. .||##| .||########||. .||########||. .||########||. Scale With Search 2026 [scalewithsearch.com](https://scalewithsearch.com) ```