AI engines like ChatGPT, Perplexity, and Google AI Overviews select businesses to recommend based on entity recognition, semantic completeness, structured data markup, brand mentions on authoritative platforms, and content freshness — not traditional SEO ranking factors like backlinks. A Qwairy analysis of 118,000 AI-generated answers found only 11% of cited domains appeared across multiple platforms, meaning each AI engine trusts fundamentally different sources and businesses need platform-specific optimization strategies.
The critical finding: 11% domain overlap
This single statistic changes everything about how professional services firms should think about AI visibility. Across 118,000 AI-generated answers analyzed by Qwairy (published March 21, 2026), only 11% of cited domains appeared on more than one platform. ChatGPT cited 42,592 unique domains. Google AI Mode cited 38,876. Perplexity cited 37,399. Claude cited 31,244. The vast majority of these domains appeared on only one platform.
This means a firm that optimizes only for Google (traditional SEO) will miss ChatGPT, Perplexity, and Claude. A firm that optimizes only for ChatGPT will miss Google AI Overviews. Comprehensive AEO requires platform-specific strategies.
How ChatGPT finds businesses: Foursquare first
The Foursquare-OpenAI partnership (announced December 2024) makes Foursquare ChatGPT's primary local business data source. Research shows 60-70% of local business results in ChatGPT come from Foursquare's Places API, which covers 100+ million points of interest across 200+ countries. Only when Foursquare data is insufficient does ChatGPT turn to Bing and other sources.
A First Page Sage study of 11,128 queries revealed ChatGPT's recommendation algorithm:
| Signal | Weight | What It Means |
|---|---|---|
| Authoritative list mentions | 41% | Appearing on "best of" lists, directories, awards |
| Awards / accreditations | 18% | Industry recognition, certifications, memberships |
| Online reviews | 16% | Review volume and sentiment (not from Google — walled garden) |
| Customer examples / usage | 14% | Case studies, testimonials, demonstrated expertise |
| Social sentiment | 11% | Reddit, social media discussions, forum mentions |
Critically, ChatGPT cannot access Google Business Profile reviews — Google's review ecosystem is a walled garden. Reviews on Yelp, BBB, Foursquare, and industry directories are what ChatGPT reads.
How Perplexity works: freshness is everything
Perplexity uses Retrieval-Augmented Generation (RAG): it decomposes every query into 3-5 sub-queries, retrieves approximately 10 pages per sub-query from a proprietary index combined with Bing, then synthesizes answers with inline citations. The result is dramatically more citations per response — Perplexity averages 21.87 citations per response compared to ChatGPT's 7.92.
Perplexity's source selection runs through a three-layer reranking system: an XGBoost quality gate filtering on entity clarity, manually curated authority domain boosts, and aggressive time decay. The freshness impact is massive:
For professional services firms, this means monthly content updates aren't optional — they're the primary driver of Perplexity visibility. A CPA firm that publishes monthly tax tips will be cited; one with a copyright 2022 in the footer will not.
How Google AI Overviews work: semantic completeness wins
Google AI Overviews don't rank pages — they select sources to synthesize answers, then add citations post-generation. A Wellows.com study of 15,847 AI Overview results across 63 industries found the dominant signals are fundamentally different from traditional Google rankings:
| Signal | Correlation | Impact |
|---|---|---|
| Semantic completeness | r = 0.87 | Pages scoring 8.5/10+ are 4.2× more likely to be cited |
| Multi-modal content | r = 0.92 | Text + images + video + structured data = 156% higher selection |
| E-E-A-T signals | 96% | 96% of citations come from strong E-E-A-T sources |
| Domain Authority | r = 0.18 | Traditional DA is declining as a factor |
The most striking finding: only 38% of pages cited in AI Overviews appear in the top 10 organic results — down from 76% in earlier analyses. Google's AI is increasingly independent from its traditional search rankings.
The universal signals that work across all platforms
Despite the 89% divergence in source selection, several optimization principles work across all AI platforms:
- Schema markup — JSON-LD structured data increases citation likelihood by 2.1× across all platforms
- Answer-first content — Opening sections of 120-180 words earn 70% more ChatGPT citations
- Original data — Publishing proprietary research makes content 3.7× more likely to be cited
- Entity clarity — Consistent NAP data across directories, schema, and content
- FAQ sections — Q&A formatting increases citation probability by 25% (Princeton GEO study)
- Citations to authoritative sources — Content citing studies and statistics gets 30-40% more AI visibility
A landmark Yext study analyzing 6.8 million AI citations found that 86% come from brand-managed sources: websites at 44%, listings at 42%, and reviews/social at 8%. Forums (including Reddit) account for just 2%. This means businesses have direct control over the vast majority of signals AI uses to make recommendations.
Platform-by-platform optimization cheat sheet
| Platform | #1 Action | #2 Action | #3 Action |
|---|---|---|---|
| ChatGPT | Claim Foursquare listing | Get on "best of" lists | Build review volume on Yelp/BBB |
| Perplexity | Update content monthly | Publish original data | Build niche authority pages |
| Google AI Overviews | Implement schema markup | Create multi-modal content | Maximize semantic completeness |
| All platforms | Allow AI crawlers in robots.txt | Answer-first content format | FAQ sections with schema |
How IECAN optimizes for all three platforms
IECAN builds websites optimized for ChatGPT, Perplexity, and Google AI Overviews simultaneously. Every website includes industry-specific schema markup (increasing citation probability by 2.1×), answer-first content architecture, AI crawler configuration for all platforms, FAQ sections with FAQPage schema, entity graph optimization across directories, and an intelligent intake system that creates structured, citable qualification content.
The intake system is IECAN's unique advantage: while competitors focus only on being found, IECAN ensures the AI-referred visitors who arrive at your website are qualified before they reach your phone — converting the highest-intent traffic on the internet into actual paying clients.
One-time pricing. No monthly subscription. No proprietary lock-in. You own everything.