Why I'm moving KinetixSEO from SEO to AI readiness

I launched KinetixSEO on 7 August as an SEO + GEO audit tool. Two weeks in I'm changing what the product leads with, and I want to write down why while it's still an open decision rather than a tidy story after the fact.

The trigger was Cloudflare. In April they shipped a free Agent Readiness score and published adoption data across the 200,000 most-visited domains. 78% have a robots.txt — and almost all of it is written for search engines, not agents. Around 4% declare AI preferences via Content Signals. Fewer than 15 sites in the entire set expose an MCP Server Card.

My first read was "free competitor with enterprise backing, I'm dead." I now think that was wrong. What Cloudflare measures is protocol adoption: whether your site exposes the right machine-readable surfaces. That's a real thing to measure, and they're commoditising it for free at a scale I could never buy. But a site can score well on it and still be cited by nobody, because nothing in its content is liftable.

So I'm conceding that layer, folding their standards into the audit, and moving the product to the two questions underneath: does a model actually cite you, and if not, what changes that.

The uncomfortable part is that I can score you today but I can't yet tell you when a fix shows up in ChatGPT. The timeline model rests on a four-week base constant with no source behind it. A confident date I can't stand behind is worse than no date, so it ships as a range until I can derive it from my own scan history.

Has anyone here repositioned two weeks after launching? Did you rewrite the landing page in one go, or run both messages in parallel for a while?