Buyers ask AI before they ever reach your site. Lumen scores whether the models can find, read, and recommend you — and monitors who gets recommended when they can't.
Nine discoverability signals, scored per scan. Five joined this quarter — the models moved, so the scan moved.
When a model doesn't recommend you, it recommends someone. Lumen asks the questions your buyers ask — "best WebUX monitoring for GTM teams", "alternatives to [category leader]" — and records the names that come back instead of yours.
Uptime monitors page for servers. Nothing paged anyone when the models stopped recommending you — until now.
Zero citations on a buyer-intent prompt. Treated like downtime — because to an AI-referred buyer, you are down.
Lumen surfaces the specific fix — schema, framing, access — as a decision with a confidence score. Pillar: Agent Visibility.
You approve, your team ships, the next monitoring run confirms the citation returned. The card closes itself.