The proof · measured, not promised

Does a higher AEO score mean more AI citations?

Not yet as a published rate — AEOForged measures that correlation in citation_evidence and only headlines a percentage after ≥20 assets across ≥2 niches; until then this page stays directional.

The Proof is AEOForged's public measurement page for that question. Every number here comes from real answer-engine observations in our evidence ledger — not from search rankings, and not from invented case studies.

Engines monitored: ChatGPT, Perplexity, Google AI Overviews, Gemini, and Copilot — each check is a direct observation, including “not cited.”

Methodology owner: Ryan Kings, Founder & CTO, AEOForged (founded 2026, Stratford-upon-Avon, UK). Evidence ledger: rebuildable with pnpm proof:dossier.

Updated

Why the proof dossier is still directional

We are still gathering measured citation observations across enough assets and niches to publish a defensible rate. We deliberately do not show a headline percentage until the sample clears our gate (20+ measured assets across 2+ niches). When it does, the score-vs-citation correlation will appear here, frozen at publish time. Status: early / directional — no published rate yet.

What is the AEOForged proof dossier?

The proof dossier is the aggregated ledger of citation_evidence — direct observations of whether a published asset was cited in a real AI answer. It is separate from search-rank tools and from our Brave-based visibility proxy. Public claims on this page come only from a published dossier snapshot; until the sample clears the gate, we show a directional state instead of a headline percentage.

What does the proof page measure?

Answer Engine Optimization (AEO) scores measure how extractable and machine-readable a page is. The Proof asks the next question: among assets we have actually checked in answer engines, do higher scores show up more often as citations?

  • Whether a published asset was cited in a real AI answer for a tracked prompt — yes or no, with engine and date.
  • AEO score at observation time — so we can compare high-scoring assets (85+) against those below the threshold.
  • Days from publish (or first track) to first citation, when a citation exists.
  • Per-engine breakdowns — ChatGPT, Perplexity, Google AI Overviews, Gemini, Copilot — never blended into a fake "AI rank."

How does AEOForged keep citation proof honest?

  1. Measured, not proxy: observations are direct answer-engine checks (Perplexity, ChatGPT, Google AI Overviews, Gemini, Copilot), including negative "not cited" results — never a Brave-search rank proxy.
  2. Gated: we only headline a rate once the sample clears at least 20 measured assets spread across 2+ niches. Below that we show a directional state with no headline percentage.
  3. Reproducible: the underlying aggregation is rebuildable from our evidence ledger with pnpm proof:dossier.
  4. Negatives count: a check where the asset was not cited is recorded and included in the rate — wins-only dashboards are marketing, not measurement.

Why is there sometimes no headline percentage?

A thin sample is still data — it is not a published rate. We refuse to invent a confident citation percentage before the dossier clears 20+ measured assets across 2+ niches. When the sample is below that gate, this page says so plainly and stays directional. That is intentional product design, not a missing chart.

Is this a guarantee that AI engines will cite my content?

No. Nobody honestly can. A higher AEO score means the page is more eligible to be extracted — structure, direct answers, schema, freshness. Whether an engine quotes you still depends on authority, accuracy, and competition. The Proof reports measured correlation on our cohort; Monitor and Dominate measure your own citation outcomes as “cited N×” or honestly “not yet cited.”

See packages on /pricing or start a free diagnosis on /start.

Summary

What this page proves

  • The Proof is AEOForged's public page for measured AI citation evidence — correlation between AEO score and real answer-engine citations.
  • Every public citation claim must come from citation_evidence (direct engine observations), never from a search-rank proxy.
  • A headline rate only publishes after ≥20 measured assets across ≥2 niches; thinner samples stay labelled directional.
  • Negative "not cited" observations are part of the data — they keep the rate honest.
  • A higher AEO score measures extractability and readiness; it does not guarantee that an engine will quote you.

As of 28 July 2026. Instrument: citation_evidence → proof dossier publish gate.