--- title: "Persistent Context Across AI Tools: Files You Keep." description: "Keep persistent context across AI tools with owned files. Give each tool a small adapter, maintain source order, and test one cold-start transfer." canonical: "https://scalewithsearch.com/articles/persistent-context-across-ai-tools" date: "2026-08-18" 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) # How to Keep Context Across ChatGPT, Claude, Cursor, and Other AI Tools. You finish a useful planning session in ChatGPT. The next task belongs in Claude. The implementation work happens in Cursor. Each tool opens with a different slice of the story, so you spend the first twenty minutes rebuilding the same business context. The incident is not that one product forgot everything. Each product has its own memory, project, file, and rules features. The failure is that the working record has no stable home outside those interfaces. ChatGPT, Claude, and Cursor do not read each other's memory, so context does not sync between them on its own. Persistent context across AI tools starts with buyer-owned files. Each tool gets a small adapter that points to those files. The business record stays stable while the interface changes. ## Separate the record from the product feature Product memory can be useful. OpenAI describes ChatGPT memory as a continually updated synthesis of context from past chats, with controls to review, correct, delete, or turn memory off. [OpenAI Memory FAQ](https://help.openai.com/en/articles/8590148-memory-faq) Claude can build memory from chats, search prior conversations, and keep project memory separate. Anthropic also documents memory import and export. [Claude chat search and memory](https://support.claude.com/en/articles/11817273-use-claude-s-chat-search-and-memory-to-build-on-previous-context) [Claude memory import and export](https://support.claude.com/en/articles/12123587-import-and-export-your-memory-from-claude) Cursor stores project rules in files under `.cursor/rules`. Those files can be version-controlled and scoped to a codebase. [Cursor rules](https://cursor.com/docs/rules) These features solve different problems. They do not need identical behavior. Your stable layer should contain the facts, decisions, corrections, source roles, task brief, and approval boundary that must survive a tool change. Product features may personalize or accelerate the session. They should not be the only place where the business defines the job. This is the same distinction behind [replacing a model without rebuilding business memory](/articles/replace-the-model-keep-business-memory). The model and interface are workers. The maintained record defines what they should know for the task. ## Know what persistent context contains Persistent AI context is information about you, your business, and your preferences that loads at the start of each AI session. It survives the end of a session, so you do not explain the business again. Without it, each session follows the same loop. You start a conversation, explain who you are, explain what you do, explain how you like work done, and only then ask the real question. The next day you repeat it. Output stays generic until you have supplied enough context. Useful persistent context has four parts: - **Identity.** Who you are, what you do, and who you serve. This prevents generic answers. - **Preferences.** Voice, tone, frameworks, and working style. This keeps output close to how you communicate. - **Operations.** Current projects, active clients, and recent decisions. This keeps answers tied to your actual situation. - **Constraints.** Rules and boundaries, written as "always" and "never." This stops unhelpful suggestions. A simple structure for a first context file: ``` # WHO [Name] [Role/Business] [Location] [Key identifiers] # WHAT [What you do] [Who you serve] [Core offering] # HOW [Voice preferences] [Frameworks you use] [Working style] # NOW [Current projects] [Active priorities] [Recent context] # RULES [Hard constraints] [Always do X] [Never do Y] ``` Models apply structured sections more consistently than long paragraphs. Here is the file for a fictional operations consultant: ``` # WHO Marcus Webb, Operations Consultant - 15 years manufacturing optimization - Based in Cleveland, OH - Clients: mid-market manufacturers # WHAT I help manufacturers reduce waste and increase throughput. Core service: 90-day operational audits Specialization: Lean implementation in job shops # HOW Voice: Direct, practical, no corporate speak Framework: Toyota Production System adapted for job shops Never: Suggest software solutions as first answer Always: Start with process before technology # NOW Active clients: Precision Metalworks, Hartley Fab Current focus: Lead time reduction project at PM Preparing: Q2 workshop series on setup reduction # RULES - No generic lean advice - Specific examples from manufacturing context - Challenge assumptions about "industry standard" - Math over feelings when evaluating improvements ``` With this file, a model knows the consultant's field, method, current work, and limits before the first question. For a business version with more sections, see the [CLAUDE.md template for business context](/articles/claude-md-template-business-context). ## Build one portable context packet Do not begin by exporting every conversation from every product. Begin with one recurring job that crosses tools. Suppose the job is a weekly account brief. The stable packet can contain: - `account-context.md` for current account facts; - `decisions.md` for approved choices and effective dates; - `corrections.md` for mistakes that must change the next run; - `task-brief.md` for the output and acceptance checks; - `source-order.md` for conflict handling; - `approval-boundary.md` for the final stopping point. Store the packet in a folder you can open without an AI subscription. Use ordinary file names. Keep credentials out of it. If a database is necessary, export the controlling records and schema needed for transfer tests. Plain text is useful here because many tools can read it and people can inspect it. It does not replace access control, backups, or structured storage. [Why plain text is the durable layer for AI memory](/articles/plain-text-ai-memory) explains the role it should play. ## Inspect `context-manifest.yaml` The manifest tells every adapter what it may load and why. ```yaml job_id: weekly-account-brief record_version: 2026.08.18.1 required_sources: - path: account-context.md role: current account facts - path: decisions.md role: approved decisions override older notes - path: corrections.md role: accepted behavior corrections - path: task-brief.md role: output and acceptance contract excluded_paths: - archive/ - credentials/ output: drafts/weekly-account-brief.md stop_before: external send receipt: receipts/weekly-account-brief.json ``` The manifest is not a prompt full of prose. It is a routing decision. It names the smallest source set for one job. The record version matters. A receipt can prove which packet a tool used. Without a version or file hash, two tools can appear to share context while reading different revisions. ## Give each tool a thin adapter ### Example: `adapters/claude.md` This is an unrun adapter specification for a file-based route. Replace the route details after testing the actual Claude product. ```markdown # Claude adapter for weekly account brief status:: synthetic specification; not a tool configuration route:: scoped local file reader in the job workspace supports:: read named text files; return filename citations; draft locally required_manifest:: context-manifest.yaml file_budget:: only the four required sources in this manifest file_limit:: record the observed per-file and total limits before trial excluded:: archive/, credentials/, other account folders write_scope:: drafts/ and receipts/ only source_rule:: approved decisions override older notes stop:: missing required file, ambiguous account, conflicting authority, untestable permission boundary, or any requested external send ``` The route's permissions must enforce the paths. The adapter records the contract; writing `excluded` does not block access. An adapter explains how one tool receives the stable packet. It does not duplicate the business rules. For ChatGPT, the adapter may tell a person which project files to attach or which approved connector path to use. For Claude, it may reference project knowledge or a local instruction file. For Cursor, it may use a project rule that points to the manifest and task brief. Keep adapter files beside each other: ```text adapters/ chatgpt.md claude.md cursor.mdc generic-chat.md ``` Each adapter should name supported capabilities, file limits, known exclusions, and the exact stop rule. Do not claim feature parity. One tool may search prior chats. Another may read repository files. A third may support a connected business system. The transfer goal is narrower: each supported route can read the same approved facts, honor the same source order, produce the required structure, and stop at the same boundary. ### Compare the four delivery methods The same file can reach a model in four ways. They differ in how much the loading depends on you. The details below were current in January 2026. **Manual loading.** Keep the context in a text file and paste it at the start of each session. It works with any AI tool and needs no setup. It costs effort every session, it is easy to skip, and the file can go stale. **ChatGPT custom instructions.** ChatGPT loads custom instructions into each chat. In January 2026, they sat under Settings, then Personalization, then Custom Instructions. Loading is automatic, but each field held about 1,500 characters. That is too little for complex context, and the text does not change with current work. **Claude Projects.** A Claude Project holds uploaded files as knowledge for every conversation in that project. It takes more context and several files. You must create the project and remember to work inside it. **`CLAUDE.md` in Claude Code.** Claude Code reads `CLAUDE.md` from the working directory at the start of each session. You do not paste or select anything. The file lives on your disk, and it can point to other files. The session still sends the text it reads to Anthropic for processing. It needs a paid Claude plan, a Claude Code setup, and a terminal. Its length is not unlimited: the model's context window sets the limit, and a short file with pointers works better than one long file. Only the last method loads with no action from you. Each of the four can carry the same packet. ### Add platform memory and retrieval to the comparison A second January 2026 guide added two more ways to save context between sessions. Platform memory features in ChatGPT, Claude, and Gemini save facts about you as you chat. You mention that you are a lawyer in Chicago, and the product keeps it. The feature is automatic, works across conversations on that platform, and can catch details you would not think to write down. Platform memory fits a person who stays on one platform with simple, stable context. The guide listed six weaknesses: - The product decides what to save, and it can save the wrong detail. - Everything sits in one pile. - The saved list sits in settings, out of view during the chat. - Capacity is limited, so older entries can drop. - A detailed brand voice can shrink to "user prefers professional tone." - ChatGPT memory does not transfer to Claude. Retrieval-augmented generation (RAG) embeds your documents into a vector database. When you ask a question, the system searches the database and adds the relevant passages to the prompt. It scales to thousands of documents and suits organizations with large documentation sets. It needs vector databases, embeddings, and API integration. RAG carries ongoing costs for embeddings, hosting, and retrieval latency, and its retrieval quality varies. For one person with 10 to 50 pages of context, it is more infrastructure than the job needs. [RAG is retrieval, and business memory also needs authority](/articles/rag-vs-business-memory) covers what retrieval does not decide. The guide compared all five methods for an individual or a small team: | Method | Ease | Control | Scale | Portability | Cost | |---|---|---|---|---|---| | Manual copy and paste | Easy | Full | Poor | Full | None beyond the AI plan | | Custom instructions | Easy | Limited | Poor | None | None beyond the AI plan | | Platform memory | Automatic | None | Limited | None | None beyond the AI plan | | RAG with a vector database | Complex | Full | Very large | Full | Highest: hosting, embeddings, and upkeep | | File-based context | Moderate | Full | High | Full | None beyond the AI plan | Frequency decides the method. At one or two AI sessions a week, copy and paste is enough: "I am a real estate broker in Austin. My market is move-up buyers. I prefer direct communication." Custom instructions handle simple voice and format rules, such as "I am a financial advisor for tech employees. Use direct language, no jargon, and bullet lists." They cannot separate client work from personal tasks or hold project detail. At more than three to five conversations a week, a file-based packet repays its setup time. Start the file with about 300 words on who you are, what you do, and how the assistant writes for you. An identity line such as "a freelance marketing consultant for B2B SaaS companies, six years in practice, selling to mid-market buyers" gives the model more to work with than "a consultant." ### Write the first file 1. Open any text editor and create a file named `CLAUDE.md` or `AI-CONTEXT.md`. 2. Write who you are and what you do, in specific terms. 3. Add voice rules and a list of active projects, as in the sample below. 4. Load the file through one route. Claude Code reads it from the working directory. ChatGPT takes it as a file upload at the start of each session. A Claude Project keeps it for every chat in that project. 5. Expand the file as the work grows. Add client profiles, workflows, templates, and decision logs. Split it into files by domain when one file gets long. ```markdown ## Voice - Tone: direct, conversational, contractions allowed - Structure: main point first, then details - Format: bullet lists over long paragraphs - Banned words: leverage, synergy, solutions, cutting-edge ## Active projects - Client A: content strategy redesign (discovery phase) - Client B: SEO audit and keyword research (deliverable due February 15) - Internal: new service offering (beta test) ``` The file stays visible. You can open it at any time and see what the model receives, which platform memory does not show you. ## Keep product memory in its proper role Product memory can hold preferences that improve interaction. Examples include writing brevity, preferred units, or a recurring personal preference. Current prices, client permissions, accepted scope, and live operational status need stronger handling. Put consequential facts in maintained records with an owner and effective date. Let product memory point toward the record when the tool supports that pattern. Do not copy sensitive records into every memory feature for convenience. Use the minimum source set for the task. Keep client and role boundaries explicit. A portable system that spreads unnecessary data across four products is easier to move but harder to govern. ## Keep the packet current A context file is not written once. Stale context produces stale output. - **Weekly.** Update the current-work section with projects and priorities. - **Monthly.** Review the rules and preferences for accuracy. - **Quarterly.** Audit the whole file. Remove old information and add new context. The investment pays most for daily AI users, for owners with several clients and projects, for work where output quality matters, and for people whose expertise should shape every answer. If you ask an AI a simple question once a week, a pasted paragraph is enough. If your processes are not written down yet, write them first, because the context file depends on them. A context file is the first layer. A fuller [AI memory architecture](/articles/ai-agent-memory-architecture-small-business) adds a searchable knowledge base, routing to different context for different work, proposed updates that a person reviews, and connections to outside systems. ## Run a cold-start transfer test The transfer test starts with a fresh session. Do not paste the original conversation. Do not rescue the tool with remembered details. Run this list for each supported tool: 1. Start a new session with no prior thread. 2. Load only the adapter and manifest-approved files. 3. Ask for the defined weekly account brief. 4. Confirm every factual claim traces to an allowed source. 5. Confirm an older archived decision does not override the current decision. 6. Remove one required source and confirm the tool stops. 7. Add a conflicting current source and confirm the conflict is reported. 8. Confirm the output uses the required file structure. 9. Confirm no message is sent. 10. Save a receipt with the tool, adapter, record version, files read, and result. Pass does not mean identical wording. Pass means the job survives the route change without losing its facts or authority rules. ## Diagnose the layer that failed If every tool uses the wrong price, inspect the stable record or source order. If one tool omits a required section, inspect its adapter and output validation. If one tool cannot access a required source, record a capability gap. This separation keeps a failure from becoming a vague argument about which model is better. You can name the failed layer and decide whether to repair, narrow, or reject that route. ## Approval and stopping boundary The preparation workflow may read the approved packet, produce the draft, run validation, and write a receipt. It must stop when a required source is missing, two controlling sources conflict, an adapter requests an excluded path, or the target tool cannot honor the output contract. It must also stop before uploading new sensitive data, connecting an account, changing retention settings, deleting product memory, canceling a subscription, or sending the output. Those actions change access, custody, or external state. They need a named owner and a separate decision. Portability is proven by a bounded cold start, not by moving the entire archive on the first attempt. ## Sources - [OpenAI: Memory FAQ](https://help.openai.com/en/articles/8590148-memory-faq) - [Anthropic: Use Claude's chat search and memory](https://support.claude.com/en/articles/11817273-use-claude-s-chat-search-and-memory-to-build-on-previous-context) - [Anthropic: Import and export your memory from Claude](https://support.claude.com/en/articles/12123587-import-and-export-your-memory-from-claude) - [Cursor: Rules](https://cursor.com/docs/rules) For the next step, read [how to replace the model without rebuilding the business memory](/articles/replace-the-model-keep-business-memory); [Full scope and terms](/work) covers the scoped build. ## Questions about How to Keep Context Across ChatGPT, Claude, Cursor, and Other AI Tools ### Switched from ChatGPT to Claude. Good call, but I lost months of context. How are people handling this? Keep the stable record outside the vendor interfaces in files containing current facts, decisions, corrections, the task brief, source order, and approval boundaries. Give each tool a thin adapter that points to the same approved packet, then run a cold-start transfer test. ### Long ChatGPT chats go bad but starting a new one means losing all your context. How do you actually deal with this? Do not move the entire transcript as the first portability strategy. Create one bounded context packet, load it into a fresh session, verify source use and output structure, and confirm the tool stops when a required file is removed. ### Do you ever get frustrated re-explaining the same context to ChatGPT or Claude every time? Store recurring business context in buyer-owned files and let each tool receive it through a documented adapter. The transfer goal is consistent facts, source precedence, output structure, and stopping behavior, not identical wording from every model. ## Save the visual summary context-manifest.yaml | account-context.md | decisions.md | corrections.md | task-brief.md | drafts/weekly-account-brief.md [Download the PNG](/infographics/persistent-context-across-ai-tools-1200x1500.png) ## Related: Owned Memory - [Give Claude Code a map of your Airtable bases without an API connection](/articles/claude-code-with-airtable) - [Export Google Workspace context to Markdown so Claude Code drafts from it](/articles/claude-code-with-google-workspace) - [Give Claude Code your Linear sprint and decision context in Markdown](/articles/claude-code-with-linear) ---- ```text .|########||. .|########||. .|########||. |##||. .||##|. |##||. .||##|. |##||. .||##|. |#|. .|#|. |#|. .|#|. |#|. .|#|. |#| |#| |#| |#| |#| |#| .#| |#. .#| |#. .#| |#. .#. .#| .#. .#| .#. .#| |#. .#| |#. .#| |#. .#| |# ...... #| |# ...... #| |# ...... #| .# ||#########| #| .# ||#########| #| .# ||#########| #| .#. |######||######|. .#. .#. |######||######|. .#. .#. |######||######|. .#. #. .##|###|##|#|######| .# #. .##|###|##|#|######| .# #. .##|###|##|#|######| .# || |##|#||||||||||||#||#| || || |##|#||||||||||||#||#| || || |##|#||||||||||||#||#| || # |#||||||||||||||||||||#| #. # |#||||||||||||||||||||#| #. # |#||||||||||||||||||||#| #. || |#||||||||||||||||||||||#| || || |#||||||||||||||||||||||#| || || |#||||||||||||||||||||||#| || # .#||||||||||||||||||||||||#| # # .#||||||||||||||||||||||||#| # # .#||||||||||||||||||||||||#| # || ....|||#||||||##|#|||#|#||##||||....|. || || ....|||#||||||##|#|||#|#||##||||....|. || || ....|||#||||||##|#|||#|#||##||||....|. || #. . ....|# ....#||| ||| .#||| ....#. .# #. . ....|# ....#||| ||| .#||| ....#. .# #. . ....|# ....#||| ||| .#||| ....#. .# # . ||||#| .#####|| . .#| .#||| ||||# #. # . ||||#| .#####|| . .#| .#||| ||||# #. # . ||||#| .#####|| . .#| .#||| ||||# #. .| ||||| #. |#|||||. || #. |#||. .||||| || .| ||||| #. |#|||||. || #. |#||. .||||| || .| ||||| #. |#|||||. || #. |#||. .||||| || || |.... #. ....|#. |. ...|. ....|| || || |.... #. ....|#. |. ...|. ....|| || || |.... #. ....|#. |. ...|. ....|| || #. .||||||##||||||##||####|||||||#||||||#| .# #. .||||||##||||||##||####|||||||#||||||#| .# #. .||||||##||||||##||####|||||||#||||||#| .# # .||####|###########################|. # # .||####|###########################|. # # .||####|###########################|. # # .#||||||.#.|| # |. ..# #| #|||||#| # # .#||||||.#.|| # |. ..# #| #|||||#| # # .#||||||.#.|| # |. ..# #| #|||||#| # # .#|||||| .. |# ## |#| ...#|#|||#| # # .#|||||| .. |# ## |#| ...#|#|||#| # # .#|||||| .. |# ## |#| ...#|#|||#| # # .#|||#|# .# .#| #| ##|.#..#|||||#| # # .#|||#|# .# .#| #| ##|.#..#|||||#| # # .#|||#|# .# .#| #| ##|.#..#|||||#| # # .#||||||############|######|||||#| # # .#||||||############|######|||||#| # # .#||||||############|######|||||#| # # |...||#||||||#||||#|#||||||#|||||#||| |.| # # |...||#||||||#||||#|#||||||#|||||#||| |.| # # |...||#||||||#||||#|#||||||#|||||#||| |.| # #. .. ||||# .|||##|. |#| .|| || .|||#. #| # .# #. .. ||||# .|||##|. |#| .|| || .|||#. #| # .# #. .. ||||# .|||##|. |#| .|| || .|||#. #| # .# || | ...|| ...##| | #| .|. |. ##### .. .| || || | ...|| ...##| | #| .|. |. ##### .. .| || || | ...|| ...##| | #| .|. |. ##### .. .| || || ||||| || ||||#| . |. |. |#. ||||| |#| || || || ||||| || ||||#| . |. |. |#. ||||| |#| || || || ||||| || ||||#| . |. |. |#. ||||| |#| || || .# |.....#|....|#.||||.|||##.|#|....||.#||.#. #. .# |.....#|....|#.||||.|||##.|#|....||.#||.#. #. .# |.....#|....|#.||||.|||##.|#|....||.#||.#. #. #.|||||||#############################| ||| .# #.|||||||#############################| ||| .# #.|||||||#############################| ||| .# || ##||#||#||||||#||#||#||#||##. || || ##||#||#||||||#||#||#||#||##. || || ##||#||#||||||#||#||#||#||##. || # .#|||||||||#||#|||||||||||#| # # .#|||||||||#||#|||||||||||#| # # .#|||||||||#||#|||||||||||#| # || |#||||||||||||||||||||||#| || || |#||||||||||||||||||||||#| || || |#||||||||||||||||||||||#| || .# |#||||||||||||||||||||#| #. .# |#||||||||||||||||||||#| #. .# |#||||||||||||||||||||#| #. || |#||||||||||||||||||#| || || |#||||||||||||||||||#| || || |#||||||||||||||||||#| || #. .######|#|#########| .# #. .######|#|#########| .# #. .######|#|#########| .# .#. |#####||||#####| .#. .#. |#####||||#####| .#. .#. |#####||||#####| .#. |# ||########|| #. |# ||########|| #. |# ||########|| #. |# ...... #| |# ...... #| |# ...... #| |#. .#| |#. .#| |#. .#| |#. .#. |#. .#. |#. .#. .#| |#. .#| |#. .#| |#. |#| |#| |#| |#| |#| |#| .|#|. .|#|. .|#|. .|#|. .|#|. .|#|. |##||. .||##| |##||. .||##| |##||. .||##| .||########||. .||########||. .||########||. Scale With Search 2026 [scalewithsearch.com](https://scalewithsearch.com) ```