AI SEO Agency for US Companies: GEO Built Into the Site

An AI SEO agency that works in the code

US search is changing shape. Google shows AI Overviews and an AI Mode on top of its links, and many buyers now ask ChatGPT, Perplexity or Copilot before they open a search page. An AI SEO agency has one job in that landscape: make your site the easiest possible source for those systems to read, trust and quote.

RAIN Design Studio approaches that job as engineering. We are a design and development studio in Casablanca, Morocco, shipping work since 2014, and we work with US companies remotely. We have no US office. What we do have is a portfolio of sites where generative engine optimization is built into the templates, the schema and the build, so it cannot drift when someone edits a page.

What we will not promise

Every serious conversation about AI SEO services should start with the limits:

• No guaranteed citations. Nobody outside OpenAI, Google, Anthropic or Perplexity controls which sources their systems cite.
• No guaranteed rankings. We do not sell positions, and we do not publish ranking or traffic results we cannot verify.
• No content volume for its own sake. Hundreds of thin AI-written pages are a risk, not a strategy.

What we do promise is checkable: each change we ship can be seen in your source code, your schema validator and your analytics.

What AI SEO services from RAIN include

Practice	What we ship	Why it matters for AI answers
Answer-first pages	A 2–3 sentence summary at the top of each key page naming who you are, what you offer and one fact	Gives engines a self-contained passage to quote
Entity graph	JSON-LD Organization, WebSite, Service, BreadcrumbList and FAQPage, linked by @id	Removes ambiguity about who the business is
llms.txt	llms.txt and llms-full.txt generated from the site’s own data	A curated map for language models, cheap to keep accurate
Crawler policy	robots.txt rules for GPTBot, OAI-SearchBot, ClaudeBot, PerplexityBot and Google-Extended	Makes AI access a business decision
Citable facts	Numbers, names and comparisons in tables and lists	Short, verifiable units are easier to extract
Freshness	Visible “last updated” dates, dateModified, sitemap lastmod	One consistent signal for people and machines
Page architecture	Hubs, service pages, guides and internal links generated from one dataset	Lets engines understand the topic structure

For the full method, see our generative engine optimization service, or start with what generative engine optimization is.

Problems we find most often in audits

When we audit US company sites for AI search readiness, the same issues come up again and again:

• Content rendered only by JavaScript, which some AI crawlers do not execute, so key pages look empty to them.
• Accidental crawler blocks, such as a robots.txt or firewall rule that shuts out AI search agents nobody meant to block.
• Schema that contradicts the page, for example an outdated address, price or service list in JSON-LD.
• Vague openings, where the first paragraph of a service page says nothing an engine could quote.
• Orphan pages and thin hubs, which hide the site’s best answers from crawlers and readers alike.

Most of these are fixed in templates rather than page by page, which is why we start with the build.

How AI SEO agencies differ

US companies comparing AI SEO companies will meet three broad types. None is wrong; they solve different problems.

	Content-led AI SEO agency	Traditional SEO agency adding AI	RAIN (engineering-led)
Main lever	Publishing volume, prompt tracking	Links, keywords, audits	Site architecture, schema, crawler policy, answer-first copy
Who changes the code	Usually your developers	Usually your developers	We do, or we rebuild on Astro
Typical output	Articles and dashboards	Reports and recommendations	Shipped templates, schema and pages
Best for	Brands with a strong technical base already	Classic organic growth	Sites whose structure is the bottleneck

If your site is already fast, well structured and rich in schema, a content-led partner may be all you need. If your platform makes structured data, speed or crawlability hard, the fix is in the build, and that is where we work.

Proof from shipped sites

These are sites we designed and built, with facts you can verify in their source:

• Origin Element, a London AI consultancy: 36 SEO and GEO content pages across services, regions (including the US and Canada), industries and guides; an entity graph of 11 JSON-LD blocks on the homepage; llms.txt generated from the same data. The client chose to allow PerplexityBot and Google-Extended while disallowing GPTBot and ClaudeBot, a reminder that crawler policy is a business choice.
• 2RCLOUD, a Salesforce consultant’s bilingual site: Astro with zero runtime JavaScript, Person and FAQPage JSON-LD, and hreflang-annotated English and French routes.
• shi — Smart & High IT: 215 pages, including 155 city pages across 9 cities, generated from one SEO dataset, with ITService and FAQPage schema and llms.txt.
• Media Progetti: JSON-LD on 66 of 69 built pages, FAQPage on 59. The delivery statement reported a Semrush audit of 100% site health and 100% AI search, with 0 errors on 78 crawled pages.

The common thread is Astro: static HTML is the easiest format for any crawler to read.

How we measure AI search work

Search Console does not report AI Overview clicks separately from ordinary web results, and AI tools share little data. So we measure what can be measured honestly:

1. Baseline. A fixed list of buyer questions, checked manually in ChatGPT, Perplexity, Copilot and Google before work starts.
2. Referral traffic. Sessions from AI tools tracked as a separate channel in your analytics.
3. Technical health. Schema validation, crawl errors, Core Web Vitals (LCP, INP, CLS) and audit scores.
4. Re-checks. The same questions re-run monthly, with results reported as observed, not projected.

Working with a remote AI SEO consultant team

• Time zone. Casablanca is five or six hours ahead of US Eastern Time, so our afternoon overlaps your morning. Pacific teams get an early call and written updates.
• Rhythm. One senior lead, one Slack channel and a Loom update every Friday. A human reviews and signs off every deliverable, even where AI tools help with research and drafting.
• Access. We work in your CMS, repository, Search Console and analytics, under your accounts.
• Payment. Invoices in USD.

Agencies that want to resell this work under their own brand can read about our white label web development process.

Pricing and next step

• Fixed-scope projects from $10,000, for example an AI SEO audit plus schema, crawler policy and a rewrite of key pages.
• Growth, $7,500 per month: 45 hours, two active projects, twice-weekly syncs.
• Scale, $15,000 per month: 100 hours, unlimited active projects, daily syncs.

If you are comparing AI SEO agencies and want one that shows its work in code rather than in promises, book a free 15-minute call. Send us your domain and three questions your buyers ask; we will tell you what we would fix first.