Rank & Rent | AI Micro Site SEO Specialist

We build rank-and-rent micro sites: small, surgically targeted local service sites that own a city + service keyword set, capture inbound calls and form fills, and are then rented to a single local business on a monthly lead-flow agreement. We are not looking for a general "SEO guy." We are looking for someone who runs an AI-native SEO stack — DataForSEO as the data layer, Claude/LLM tooling as the reasoning and production layer, and automation (n8n / Make / GHL) as the ops layer — and who understands that in 2026 a micro site has to rank in three places at once: classic organic, the map pack, and inside AI answers (ChatGPT, Google AI Overviews, Perplexity, Claude). If your process is "install an SEO plugin and write 20 blog posts," this is not the role. The tactic, in detail 1. Market selection — pick a niche we can beat, not one we like Before a single page is built, the target market gets scored: Transactional keyword extraction. We target the words customers actually type ("EV charger installation cost"), not industry jargon ("level 2 charging infrastructure"). Volume and CPC pulled live from DataForSEO — no guessing, no "I think this gets searched." Competitor fundamentals head-to-head. Score the top 3 incumbents out of 100 across five categories: title tags/meta, heading structure, schema, page speed, image SEO. We only enter a market where we can win at least 4 of 5 on execution alone. Backlink weakness audit. Referring-domain count, per-domain authority, anchor distribution, spam signals. The green light looks like: a handful of rank-0 referring domains and anchor text that is almost entirely the bare brand name — a profile beatable with 6–8 genuinely good links, not volume. AI citation gap. Run an AI-visibility check on the incumbents. A competitor with zero citations across ChatGPT / AI Overviews / Perplexity is an open lane, and that lane is the single biggest reason this play still works. Deliverable: a go/no-go market brief with the scoring, the seed keyword set, projected lead volume, and the rent price the market can carry. 2. Build — a static asset engineered to load in under a second Static framework (Astro or equivalent), deployed to Cloudflare/Vercel, custom domain. Lighthouse 95+ across all four categories on localhost before deploy — not after, not "we'll fix it later." Full structured data: LocalBusiness / Service / FAQ schema, sameAs links to every real profile so AI engines resolve the site as an entity, not a URL. llms.txt published; robots.txt verified not to block GPTBot, ClaudeBot, PerplexityBot, or Google-Extended; key content confirmed server-rendered so AI crawlers can actually read it. Templates, not pages. One service-page template, one location-page template, one article template — then the city × service matrix is generated against them. 3. Content — every page exists to push the money page Money page first. One transactional page carries the primary intent. Its FAQ block answers the highest-value People-Also-Asked questions directly. Fan-out query targeting. Pull the sub-questions AI engines internally generate when answering the seed query, then write one thorough, source-cited piece per chosen fan-out query. This is the mechanism that gets the site cited rather than just indexed. Answer-capsule structure. TL;DR up top, cited live sources, clean heading hierarchy, schema on every piece. Internal link architecture with specified anchors. Supporting content builds topical authority; internal links route that authority to the money page. Done correctly, a 0-backlink page beats a 400-backlink page — we have seen it and we expect you to have too. No generic AI slop. Output must carry real operator-level specificity (pricing ranges, local detail, process steps, first-hand experience). We review for this and reject on it. 4. Local layer — map pack + NAP discipline Google Business Profile built and optimized: primary + secondary categories, services validated against actual search volume (zero volume = dropped), service areas, hours, photo program. Core citation set with a locked NAP block used byte-identical everywhere: Bing Places, Apple Business Connect, Yelp, Facebook, BBB, Yellow Pages, plus niche/regional directories matched to the vertical. Suspension hygiene is non-negotiable: no virtual offices, no mailbox addresses, no shared-address stacking, no cities the business does not serve. 5. AI visibility — the part most operators are still ignoring Baseline AI citation audit across ChatGPT, Google AI Overviews, Perplexity, and Claude, with a competitor scoreboard. Monthly re-check to prove the content work is actually moving citations, not just rankings. Reporting must show which queries the site now owns in AI answers, not a vanity score. 6. Instrumentation — a rentable asset is a measurable asset This is what converts a ranked site into a rentable one, and it is where most rank-and-rent builds fall apart: Call tracking number on the site and GBP, with recorded, timestamped call logs. GA4 key events for form fills, click-to-call, and bookings. Weekly rank tracking (organic + local pack) piped into a dashboard the renter can be shown. A CRO pass on the money page: offer clarity, above-fold click-to-call, short form, visible social proof. A lead we cannot count is a lead we cannot invoice for. Deliverables & milestones # Milestone Output 1 Market validation Scored go/no-go brief, keyword set, competitor weakness report, projected rent price 2 Build & technical Live site, Lighthouse report, schema validated in Rich Results Test + validator.schema.org, llms.txt, sitemap submitted in GSC 3 Content & structure Money page, city × service matrix, supporting fan-out content set, internal link map with anchors 4 Local & citations GBP live and optimized, citation set submitted with NAP audit sheet 5 Instrumentation & handoff Call tracking + GA4 events live, rank/citation dashboard, SOP doc so we can replicate the build 6 90-day performance window Monthly report: rankings, map-pack position, AI citations, calls, form fills What we expect you to bring Required Demonstrable rank-and-rent or local lead-gen sites you built and monetized — URLs, timelines, and lead volume. Case studies beat credentials here. Hands-on DataForSEO (API or MCP) or equivalent raw-data workflow. We want to see live data in the reports, not screenshots from a freemium tool. Working LLM-assisted production process — Claude/Claude Code, custom skills, MCP connectors, or your own equivalent. Tell us what your stack actually is. Schema, Core Web Vitals, and technical SEO you can defend line by line. Google Business Profile optimization and citation building, including suspension-risk judgment. Understanding of generative engine optimization: fan-out queries, citation tracking, llms.txt, crawler access, entity building. Nice to have Automation build experience (n8n, Make, GHL) for reporting and review workflows. Astro / static-site deployment on Cloudflare or Vercel. Multi-location or franchise SEO experience. Hard rules No PBNs, no spun content, no fake GBP listings, no address stacking, no Fiverr-tier link marketplaces. Anything that risks a suspension or manual action kills the asset and the engagement. No guaranteed-ranking promises. We evaluate on leading indicators and lead volume. Screening questions Link one rank-and-rent or local lead-gen site you built. What did it rank for, how long did it take, and what did it rent for? Walk us through how you would decide not to enter a market. What specific numbers would stop you? What is your data source for keyword volume and SERP data, and how does it get into your workflow? How do you get a page cited inside an AI answer, as distinct from ranking it in classic organic? How do you keep AI-assisted content from reading like AI-assisted content? Budget & terms Fixed-price by milestone for the pilot site, with a performance bonus tied to the 90-day lead volume. Propose your own number — include a rough hour split across the six milestones so we can see how you actually work. Strong pilot performance turns into a recurring build pipeline; we have a roster of markets queued. Please include the word "capsule" somewhere in your proposal so we know you read past the first paragraph.