--- title: "Document each search once so AI outreach names the real role" description: "Write each active search into a file with the role, company, pay range, process, and selling points so Claude drafts outreach that fits one role." canonical: "https://scalewithsearch.com/articles/ai-for-recruiters" 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) # Document each search once so AI outreach names the real role. You recruit for five roles at once. The clients are a fintech startup, a SaaS company, an e-commerce brand, a healthcare company, and a B2B platform. The roles are a senior backend engineer, a sales director, a marketing manager, a data analyst, and a product designer. You ask ChatGPT for LinkedIn InMail for the engineering role. It writes about "exciting opportunities" and "cutting-edge technology." It does not mention the Series B round, the payment infrastructure product, or the pay range. It reads like every recruiter message the candidate ignores. You are on message 47 today. You do not have time to rewrite each draft. ## Why a generic model writes boilerplate outreach ChatGPT can write a recruiter template. It cannot write your message, because it does not know the role or the company. It also does not know why this role differs from the ten other offers the candidate got this week. So it produces the lines that everyone uses: - "I came across your profile and was impressed by your background." - "We're looking for a talented professional to join our growing team." - "This is an exciting opportunity to make an impact." - "Competitive salary and benefits package." Candidates recognize these lines at once. They do not reply, because the message shows that nobody read their profile. The model does not know seven things: - what the client company does; - what makes this role different from similar roles; - which requirements are real and which are preferences; - the pay range and benefits; - the client's culture and who does well there; - the interview process; - why the role is open and what happened to the last person in it. With several searches for several clients, the model also mixes them. It writes engineering outreach that sounds like sales outreach. It uses the wrong company name. ## What a search file holds You need the model to start from the search, not from a blank page. For each search, record: - the client company profile and what the company does; - role details, with must-have requirements and deal-breakers; - the pay range and benefits, with the source of the range; - the culture and the kind of person who fits; - the interview process and timeline; - why the role is open and what success looks like; - the selling points that set this role apart. At 30 messages a day across six searches, you cannot rebuild this context for each draft. ## Write the search file Claude Code reads `CLAUDE.md` from the folder where you start it. Use that file for your standing rules, and keep one file per search in a `searches/` folder. A synthetic search file looks like this: ```text # SEARCH-PF-01: Senior Backend Engineer, PayFlow (synthetic example) location:: San Francisco, hybrid 3 days a week pay:: base range and equity band from the client's approved offer sheet, 2026.01.10 equity:: 0.1 to 0.25 percent stage:: Series B; growth figures from the client's deck, not from memory product:: payment infrastructure for B2B SaaS companies with recurring revenue stack:: Go, Python, PostgreSQL, Redis, AWS, Kafka for event streams, React for internal tools must_have:: 5+ years of backend work; payment systems or high-transaction systems nice_to_have:: Go (client will train); B2B SaaS background culture:: startup pace, not a 9-to-5 job; owners, not ticket-takers; 15 engineers and everyone ships code; CEO and CTO are both engineers why_open:: new headcount after Series B to build a fraud detection system; not a backfill selling_points:: hard technical problems at scale; equity in a growing company; direct work with the founders interview_process:: 1. 30-minute recruiter screen 2. 60-minute technical screen with the engineering manager 3. 4-hour onsite: 2 technical, 1 system design, 1 team interview 4. offer decision usually within 48 hours deal_breakers:: will not work hybrid (the client tried full remote and ended it); does not want startup risk; wants a stable 9-to-5 role ``` Now you ask Claude Code for InMail to a backend engineer with payment-processing experience. It reads the PayFlow file. It knows the stack, the selling points, and the process. The draft fits this role and no other. When you switch to another search, Claude reads a different file. Separate files keep the company names and facts apart. The pattern is in [separating client context before agent work](/articles/separate-client-context-before-agent-work). Several US states require a pay range in job postings. Keep the client's approved range in the file with its source and date. Do not let a draft invent or round a range. ## Three tasks, before and after Outreach. Before: ChatGPT writes, "I came across your impressive profile and wanted to reach out about an exciting Senior Backend Engineer opportunity at a fast-growing fintech company." You rewrite it from scratch. After: Claude reads the search file and drafts a message like this one: ```text You built the payment reconciliation system at [their company]. PayFlow is hiring a Senior Backend Engineer to build fraud detection for its payment platform. Series B, technical founders, Go and Python. Base [approved range] plus equity. Worth a 30-minute conversation? ``` Specific, relevant, and short. You check the candidate details and send it. Job descriptions. Before: you ask for five job descriptions and get five near-copies with different titles. All five say "fast-paced environment" and "competitive compensation." After: Claude reads each search file and writes five descriptions that describe five companies, five sets of requirements, and five sets of selling points. You edit lightly and post. Candidate questions. Before: a candidate asks about the interview process. You dig through email for the hiring manager's notes from three weeks ago and reply 20 minutes later. After: you ask, "What is the PayFlow interview process?" Claude answers from the file: recruiter screen, technical screen, 4-hour onsite, offer decision within 48 hours. You reply in a minute. ## A week with search files Monday morning. You source for three roles. You find a strong backend engineer with payment-processing experience at a large payments company, who now works at a Series A startup. You ask: "Write InMail for the PayFlow backend role. The candidate built payment systems at a large processor and now works at a Series A startup." Claude reads the file. The draft mentions the candidate's payments work. It explains why PayFlow is a logical next step: a later stage and larger scale problems. It lists the stack and the approved pay range. Review and send take two minutes. You find another candidate for a different search: the VP of Sales role at CloudMetrics. The candidate runs West Coast sales at a competitor. You ask for InMail. Claude reads the CloudMetrics file and writes a different message about career progression, team size, and pay structure. Nothing from the engineering file appears in it. Wednesday afternoon. The PayFlow hiring manager asks you to schedule technical screens for two candidates. You ask: "Draft interview confirmations for PayFlow candidates: technical screen with the engineering manager, next Tuesday at 2 p.m." Claude knows the format and length of that screen. The draft explains the format, the duration, and what to prepare. ## Measure the rework before you trust the change Here is the arithmetic on an assumption. If you send 30 messages a day and fix each generic draft for 5 minutes, you spend 2.5 hours a day on rework. That is more than 12 hours a week. Test your own number. Time one day of outreach with generic drafts and one day with search files. Count the drafts you sent without an edit. ## Close each search with a record When a search closes, do not delete the file. Add the outcome first: filled or cancelled, the start date, and which selling points drew replies. Record which requirements the client relaxed during the search, and which deal-breakers held. Then remove any candidate specifics and move the file to an `archive/` folder. The next search for the same client, or for a similar role, starts from that record. You know what the client really accepts, which messages worked, and how long the process took. ## Keep candidates and sends under your control The search file holds role context. Candidate records stay in your applicant tracking system. Put in a draft only the candidate facts that the message needs, and apply the [minimum-necessary test for personal data](/articles/client-data-in-ai-agent-memory). The files stay on your computer, but Claude Code sends the text it reads to Anthropic's servers during a session. Check your client agreements for rules on confidential role details, such as an unannounced backfill. You send every message yourself. LinkedIn's User Agreement bars bots and other unauthorized automated methods that send messages. The draft step is where AI helps. The send step stays with you. [Who approves what an AI agent sends](/articles/who-approves-what-an-ai-agent-sends) describes that boundary. ---- ```text .|########||. .|########||. .|########||. |##||. .||##|. |##||. .||##|. |##||. .||##|. |#|. .|#|. |#|. .|#|. |#|. .|#|. |#| |#| |#| |#| |#| |#| .#| |#. .#| |#. .#| |#. .#. .#| .#. .#| .#. .#| |#. .#| |#. .#| |#. .#| |# ...... #| |# ...... #| |# ...... #| .# ||#########| #| .# ||#########| #| .# ||#########| #| .#. |######||######|. .#. .#. |######||######|. .#. .#. |######||######|. .#. #. .##|###|##|#|######| .# #. .##|###|##|#|######| .# #. .##|###|##|#|######| .# || |##|#||||||||||||#||#| || || |##|#||||||||||||#||#| || || |##|#||||||||||||#||#| || # |#||||||||||||||||||||#| #. # |#||||||||||||||||||||#| #. # |#||||||||||||||||||||#| #. || |#||||||||||||||||||||||#| || || |#||||||||||||||||||||||#| || || |#||||||||||||||||||||||#| || # .#||||||||||||||||||||||||#| # # .#||||||||||||||||||||||||#| # # .#||||||||||||||||||||||||#| # || ....|||#||||||##|#|||#|#||##||||....|. || || ....|||#||||||##|#|||#|#||##||||....|. || || ....|||#||||||##|#|||#|#||##||||....|. || #. . ....|# ....#||| ||| .#||| ....#. .# #. . ....|# ....#||| ||| .#||| ....#. .# #. . ....|# ....#||| ||| .#||| ....#. .# # . ||||#| .#####|| . .#| .#||| ||||# #. # . ||||#| .#####|| . .#| .#||| ||||# #. # . ||||#| .#####|| . .#| .#||| ||||# #. .| ||||| #. |#|||||. || #. |#||. .||||| || .| ||||| #. |#|||||. || #. |#||. .||||| || .| ||||| #. |#|||||. || #. |#||. .||||| || || |.... #. ....|#. |. ...|. ....|| || || |.... #. ....|#. |. ...|. ....|| || || |.... #. ....|#. |. ...|. ....|| || #. .||||||##||||||##||####|||||||#||||||#| .# #. .||||||##||||||##||####|||||||#||||||#| .# #. .||||||##||||||##||####|||||||#||||||#| .# # .||####|###########################|. # # .||####|###########################|. # # .||####|###########################|. # # .#||||||.#.|| # |. ..# #| #|||||#| # # .#||||||.#.|| # |. ..# #| #|||||#| # # .#||||||.#.|| # |. ..# #| #|||||#| # # .#|||||| .. |# ## |#| ...#|#|||#| # # .#|||||| .. |# ## |#| ...#|#|||#| # # .#|||||| .. |# ## |#| ...#|#|||#| # # .#|||#|# .# .#| #| ##|.#..#|||||#| # # .#|||#|# .# .#| #| ##|.#..#|||||#| # # .#|||#|# .# .#| #| ##|.#..#|||||#| # # .#||||||############|######|||||#| # # .#||||||############|######|||||#| # # .#||||||############|######|||||#| # # |...||#||||||#||||#|#||||||#|||||#||| |.| # # |...||#||||||#||||#|#||||||#|||||#||| |.| # # |...||#||||||#||||#|#||||||#|||||#||| |.| # #. .. ||||# .|||##|. |#| .|| || .|||#. #| # .# #. .. ||||# .|||##|. |#| .|| || .|||#. #| # .# #. .. ||||# .|||##|. |#| .|| || .|||#. #| # .# || | ...|| ...##| | #| .|. |. ##### .. .| || || | ...|| ...##| | #| .|. |. ##### .. .| || || | ...|| ...##| | #| .|. |. ##### .. .| || || ||||| || ||||#| . |. |. |#. ||||| |#| || || || ||||| || ||||#| . |. |. |#. ||||| |#| || || || ||||| || ||||#| . |. |. |#. ||||| |#| || || .# |.....#|....|#.||||.|||##.|#|....||.#||.#. #. .# |.....#|....|#.||||.|||##.|#|....||.#||.#. #. .# |.....#|....|#.||||.|||##.|#|....||.#||.#. #. #.|||||||#############################| ||| .# #.|||||||#############################| ||| .# #.|||||||#############################| ||| .# || ##||#||#||||||#||#||#||#||##. || || ##||#||#||||||#||#||#||#||##. || || ##||#||#||||||#||#||#||#||##. || # .#|||||||||#||#|||||||||||#| # # .#|||||||||#||#|||||||||||#| # # .#|||||||||#||#|||||||||||#| # || |#||||||||||||||||||||||#| || || |#||||||||||||||||||||||#| || || |#||||||||||||||||||||||#| || .# |#||||||||||||||||||||#| #. .# |#||||||||||||||||||||#| #. .# |#||||||||||||||||||||#| #. || |#||||||||||||||||||#| || || |#||||||||||||||||||#| || || |#||||||||||||||||||#| || #. .######|#|#########| .# #. .######|#|#########| .# #. .######|#|#########| .# .#. |#####||||#####| .#. .#. |#####||||#####| .#. .#. |#####||||#####| .#. |# ||########|| #. |# ||########|| #. |# ||########|| #. |# ...... #| |# ...... #| |# ...... #| |#. .#| |#. .#| |#. .#| |#. .#. |#. .#. |#. .#. .#| |#. .#| |#. .#| |#. |#| |#| |#| |#| |#| |#| .|#|. .|#|. .|#|. .|#|. .|#|. .|#|. |##||. .||##| |##||. .||##| |##||. .||##| .||########||. .||########||. .||########||. Scale With Search 2026 [scalewithsearch.com](https://scalewithsearch.com) ```