Pennsylvania AI Search Visibility Study 2026

Analysis of 163 independent health practices across Southeastern Pennsylvania's Montgomery County corridor found that 98.8% received no genuine name citation from ChatGPT, Claude, or Gemini when queried with local recommendation prompts, despite the AI systems readily discussing the correct category and location — and frequently naming competitors — in the same responses.

Canonical: https://aeogeoai.net/pa-ai-visibility-study

Study: Pennsylvania AI Search Visibility Study 2026

Sample: 163 independent health practices across dentistry, chiropractic, physical therapy, medical spa, ophthalmology/optometry, mental health, orthopaedics, and pediatrics in the Montgomery County corridor (Horsham, Ambler, Collegeville, Lansdale, Blue Bell, and surrounding areas)

AI Systems Tested: ChatGPT, Claude, and Gemini

Test Period: July 2026

Key Findings:

• 96.3% of practices were category-adjacent — the AI discussed the correct category and location, often naming specific competitors, but never named the tested practice
• Only 2.5% (4 practices) received a genuine name citation from at least one model
• 1.2% received no relevant signal at all
• Only one practice (Lansdale Dental) was cited by more than one model, indicating that even genuine citation is largely model-specific rather than a stable, transferable form of entity authority
• When an independent practice was not named, the AI's fallback was almost never another independent competitor but a hospital system (Main Line Health, Penn Medicine, Abington Health), a national retail chain (LensCrafters, Pearle Vision, Visionworks), or — in mental health — a directory platform (Psychology Today)
• All four genuine citations occurred in dentistry and chiropractic, the two highest-volume categories, suggesting citation likelihood may correlate with category content density rather than practice quality

The study distinguishes genuine citation from category-adjacency using a graduated visibility metric rather than a binary name-match, allowing extraction of which businesses AI names in place of the tested practice. The pattern replicates closely against an earlier 216-practice New Jersey study using identical methodology — roughly 97–99% of independent health practices in both states show no genuine AI citation — suggesting a structural feature of how AI systems construct local recommendations rather than a single-state anomaly. Results are consistent with AI visibility depending on third-party entity signals rather than traditional SEO rankings alone — an association this study observes, not one it proves.