Be the sourceAI engines cite.
Your buyers research via ChatGPT, Claude, Perplexity, and Google AI Overview before they ever talk to your sales team. VectorCite gets you cited inside those answers — through an open 47-signal rubric, a forward-deployed execution team, and continuous citation-share tracking.
Built for the engines your buyers actually ask
Search is collapsing into a single answer.
A ranked page of ten blue links is becoming one synthesized paragraph with a handful of citations. AI Overviews already appear on roughly a third of Google results. In that world there is no page two — you are the cited source, or you are invisible.
Ten blue links. You fought for a rank and hoped for the click.
One synthesized answer. A few citations. You are quoted — or you do not exist.
Answer Engine Optimization
Being cited inside the generative answer itself — the paragraph ChatGPT, Perplexity, Gemini and Google AI Overviews hand your buyer. The citation is the new click.
Generative Engine Optimization
The wider discipline of being retrievable, quotable and trusted by generative engines. AEO is its sharp end; GEO is the whole surface area of being machine-citable.
SEO won you the link. AEO wins you the answer.
Probe. Measure. Prescribe. Fix.
Four movements turn “why aren’t we cited?” from a guess into a measured answer with an executable fix attached.
Probe
Ask the answer engines the questions your buyers actually type. Each query is sampled several times; every page cited in the synthesized answer is captured.
Measure
Your content and the cited set are embedded into one 1,536-dimensional space. The centroid shift ‖Δ‖ is the distance between you and the answer — quantified, not guessed.
Prescribe
A gpt-4o pass turns the geometry into a ranked action plan — what to write, restructure and mark up — ordered strictly by citation impact.
Fix
Not advice — artifacts. Paste-ready schema.org JSON-LD and rewritten copy, generated against your real page and the competitors that won the citation.
Then verify — the loop closes.
Mark the fix shipped and VectorCite re-probes. A Beta-Binomial test tells you whether the citation lift is real, not anecdotal.
Most AEO tools count keywords. VectorCite measures geometry.
Citation is not a string-match problem. It is a question of distance — how far your content sits from the centroid of what the engine chose to quote. So we measure that distance directly, in the same space the model reasons in.
One vector space
Your chunks and the cited competitors' chunks are embedded with the same model — so “relevance” stops being a guess and becomes a measurable distance.
The centroid shift
‖Δ‖ is the gap between your content and the centroid of what got cited — classified as a coverage, relevance or evidence gap by geometry, not by vibes.
A closed loop
Ship the fix, mark it applied, and VectorCite re-probes. A Beta-Binomial significance test says whether the citation lift is real or just noise.
dimensions in the vector diff
answer engines probed in parallel
Wilson-interval confidence on every citation rate
The honest answers.
Stop hoping you’re cited. Measure it.
Point VectorCite at a page and a buyer prompt. Get the gap, the plan and the paste-ready fix — in one probe.
The methodology is open.The hosted product is the convenient version.
Same play as Lighthouse for web perf or OWASP for security. The rubric is published. The CLI is one npx away. Install plugins where you already work.
npx aeo-audit https://…
Zero-dep Node CLI + the AEO-SPEC-v1 formal spec.
Admin page · [aeo_badge] shortcode · weekly auto-audit
Tools → AEO Audit. Install in 60 sec.
{% render 'aeo-badge' %}
Liquid snippet + drag-and-drop section.
Embed · static card · dynamic JS card
Drop into any Webflow Embed.