Save time, make money and get customers with FREE AI! CLICK HERE →

Claude AI SEO Case Study: From Zero to 39,000 Clicks Without Logging In

This Claude AI SEO case study shows how a website I never once logged into went from zero to 39,000 clicks. Claude did the building, the writing and the publishing inside my Agent OS. The site hit page one of Google in about two weeks, and from my Search Console it now gets over 1,000 clicks a day. In the video below I show the exact stack, the URL rules, the page template and how the agent picks the next keywords. Here’s the full breakdown.

Key takeaways

  • From my Search Console: zero to 39,000 clicks.
  • Page one of Google in about two weeks.
  • I never logged into the site — Claude did the work.
  • Book a free SEO strategy session to apply this to your site.

The results

Here’s what the numbers look like. These are my own self-reported numbers from Google Search Console, not an outside audit.

Metric Result
Starting point A brand-new site with zero clicks
Time to page one About two weeks
Total clicks 39,000
Current daily clicks 1,000+ a day
Rankings #1 positions in Google and in AI search answers
Times I logged in Zero

The point of this case study isn’t the number. It’s the method. Claude ran the whole thing inside my Agent OS, and I gave it rules instead of doing the work myself.

Why I never logged in

This sounds like a gimmick but it’s the most important part. If you can log in and fix things by hand, you will. And then you’re the bottleneck. By never logging in, I forced the AI to do the work end to end: build the pages, write the content, publish and fix its own mistakes.

That changes how you think. Instead of editing a page, you edit the rules the agent follows. Every fix to a rule fixes every future page, not just one. That’s how you get a self-running site.

Want Claude running SEO for your business? Book a free SEO strategy session and we’ll look at your site, your Search Console data and the keywords you’re missing.

Prefer to pick a time straight away? Choose a slot on my calendar.

Want to run AI agents like Claude Code yourself? Join the AI Profit Boardroom for the Agent OS and step-by-step training.

Claim Your Free SEO Strategy Session →Book a Free Strategy Session →Join the AI Profit Boardroom →

The stack: simple on purpose

There’s nothing fancy here. The stack is boring, and that’s why it works with AI.

Part Why I use it
Plain HTML pages Fast, easy for search engines to read, easy for an AI to write
Eleventy A simple static site generator that turns templates into pages
Netlify Hosts the static site and deploys new pages quickly
Static sites No database, no plugins, no login screen to break

An AI agent is very good at writing files. A static site is just files. So the agent can build, change and deploy the whole site without a dashboard, a plugin or a person clicking buttons.

URL rules: one keyword per URL

Two rules, and I never break them:

  • One keyword per URL. Every keyword gets its own page. No stuffing five keywords into one article.
  • Never change a URL. Once a page is live, its address stays the same forever, so it keeps the rankings and links it earns.

These rules stop the agent from making pages that compete with each other, and stop it from breaking pages that already rank. They’re in the playbook the agent reads every time.

The page template

Every page follows the same template. That’s what makes the quality consistent across hundreds of pages.

  • A clear title with the keyword near the front.
  • An answer-first intro. The first lines answer the question straight away, which helps both Google and AI search tools pull the answer.
  • FAQ schema. A short FAQ section marked up so search engines understand it.
  • E-E-A-T signals. Real experience, a named author and real proof, so the page is trustworthy.

You can see the same idea in how I write about Claude GEO SEO: answer first, structure clearly, mark it up.

How the agent finds the next keywords

The agent doesn’t guess. It uses Google Search Console data. It looks for searches where the site already shows up but doesn’t get clicks, then builds a page for that exact search. Search Console plus AI means the next keyword always comes from real data, not opinions.

This is a loop. Publish, wait for impressions, find the gaps, publish again. Each round the site covers more of what people actually search for. The full version of this loop is in my Claude AI SEO system guide.

Hand over the playbook

The last step is the one people skip. I wrote a playbook — the stack, the URL rules, the template, the keyword process — and handed it to the agent. Then I let Claude Opus 5.5 build the site from it.

A good playbook is the difference between an AI that writes random articles and an AI that runs an SEO campaign. If you want my prompts to start with, grab my free SEO prompts. There’s also a version of this case study on the AI Profit Boardroom blog.

What you can copy this week

  • Pick one topic and set up a simple static site on Netlify.
  • Write your rules down: one keyword per URL, never change URLs, the page template.
  • Connect Search Console so your agent works from real data.
  • Let the agent publish a small batch of pages, then check results in two weeks.
  • Fix the rules, not the pages, when something looks off.

Results depend on your niche, your competition and how good your rules are. This is what worked for me; it’s not a promise of the same numbers.

Common mistakes when copying this

  • Logging in to fix things by hand. It feels faster, but the rule that caused the problem is still there, so the same mistake comes back on the next page.
  • Changing URLs to chase a new keyword. You throw away the rankings the old page earned. Make a new page instead.
  • Picking keywords from a tool instead of your own data. Search Console shows what Google already connects to your site, which is where quick wins come from.
  • Skipping the template. Without one, quality drifts from page to page and you can’t tell which change helped.

Each of these is easy to avoid if the rules live in the playbook the agent reads, not in your head.

My verdict on this Claude AI SEO case study

The lesson from this Claude AI SEO case study is that the AI isn’t the magic. The rules are. A simple static stack, strict URL rules, one solid page template and real Search Console data let Claude run a site from zero to 39,000 clicks without me logging in once. If you want help building that kind of setup for your business, book a time on my calendar.

About Julian Goldie

I’m Julian Goldie — founder of Goldie Agency and the AI Profit Boardroom. I test new AI agents almost every day for 432K+ YouTube subscribers and run Hermes, OpenClaw, OpenCode, Claude and Codex side by side in my own Agent OS. Claim your free strategy session or join the AI Profit Boardroom.

FAQ: claude ai seo case study

What is the Claude AI SEO case study?

A site I never logged into, built and run by Claude inside my Agent OS, that went from zero to 39,000 clicks according to my Search Console.

How fast did the site rank?

It reached page one of Google in about two weeks and now gets over 1,000 clicks a day, based on my own Search Console data.

What stack does it use?

Plain HTML static pages built with Eleventy and hosted on Netlify, with no database or plugins.

What are the URL rules?

One keyword per URL, and never change a URL once it is live.

Can I get help setting this up?

Yes. Book a free SEO strategy session and we will look at your site and the keywords you are missing.

Related reading

Last updated October 2026. This page is a living guide to claude ai seo case study — the facts here move fast and it is updated as they do.