Claim
AI search citations and web traffic are decoupled. Measure AI search presence on six separate signals: platforms where AI cites the site, prompt types triggering citations, raw citation count, linked citation count, recommendation count, and downstream business outcomes. Blending them into a single score masks which layer is performing and which is broken.
Mechanism
The citation-to-traffic chain has multiple failure points: a citation can appear without a linked URL; a link can appear without a click; a click can arrive without a conversion. Each failure point requires a different intervention. A blended AEO score hides which link in the chain is broken and produces the wrong corrective action.
Conditions
Holds when: the team has traffic instrumentation capable of attributing sessions to AI search referrers separately from other sources.
Fails when: attribution tools do not separate AI search traffic from dark social or direct traffic, making the downstream signals unmeasurable.
Evidence
"More AI answer links don't automatically mean more traffic." — Aleyda Solis, SEOFOMO, May 10, 2026
Solis specifies six tracked signals: platforms where AI cites the site, prompt types triggering citations, raw citation count, linked citation count, recommendation count, and business outcomes downstream of AI traffic.
Signals
- Citation count rising while linked citation count is flat, a decoupling visible in the data
- Business outcome tracking shows AI traffic converts at a different rate than organic
Counter-evidence
Six-signal tracking requires instrumentation most teams have not built. A blended score is a reasonable starting point for early-stage AEO programs where only the Presence layer is measurable.