Search is splitting in two: the ten blue links, and the answer that ChatGPT, Perplexity, Copilot, and Google's AI Overviews synthesize before anyone clicks. This is the system I built to compete on both — and a live board tracking the climb from day zero.
I run five live products across three deploy stacks. Each was invisible to AI answer engines for the same reason most sites are: no machine-readable identity. An LLM can read a page, but to cite a source confidently it wants structured signals — who published this, what the product is, how the entity connects to a real person and their other work. Getting that right across five sites, on Next.js/Vercel and static/Netlify, by hand, doesn't scale. So I built it to ship and measure itself.
schema.org JSON-LD — Organization, WebSite, SoftwareApplication, Article — deployed and verified against the live HTML, not just the build.
A sameAs graph links the founder identity across GitHub, X, and LinkedIn so "PJ Geldenhuis" resolves to one person across every property.
A fixed-action bridge per repo: verify → commit → push → build, across Vercel and Netlify — each deploy confirmed on the served page.
A daily probe checks rank and AI-citation for each site's target query against a fixed baseline. The numbers below are that loop.
Baseline set 16 Jul 2026. Rankings and AI citations take weeks to move — this is honest day-zero, and it climbs from here.
| Site | Target query | Rank | AI-cited | Schema |
|---|