We Tracked 2,342 AI Search Citations Across ChatGPT, Claude, and Perplexity — Here's What We Found

The Experiment: 7 Weeks of Automated AI Citation Monitoring

From February 13 to April 3, 2026, we ran automated citation checks across three major AI search engines: ChatGPT, Claude, and Perplexity. We queried each engine with 12 different search queries — ranging from brand-specific queries to generic industry queries — and tracked whether a specific business was cited in the response.

In total, we collected 2,342 data points. Here's what the data actually shows about how AI engines decide which businesses to cite.

Overall Citation Rates by Engine

Not all AI search engines are equal when it comes to citing businesses. Here's how the three engines compared:

Key takeaway: Perplexity cites businesses at 2x the rate of ChatGPT and Claude. This makes sense — Perplexity is designed as a search engine that actively retrieves and cites web sources in real time. ChatGPT and Claude rely more heavily on training data, which means they favor businesses with established third-party authority (press coverage, directory listings, industry publications).

What Types of Queries Trigger Citations?

This is where the data gets really interesting. We tested 12 different query types and found a stark divide:

The pattern is clear: AI engines will cite you for brand queries (when someone asks about you by name) and local/specific queries — but generic industry queries are dominated by companies with massive external authority.

Who Gets Cited for Generic Queries? The Competitor Data

We also tracked which companies AI engines recommend when asked generic queries like "best AI consulting firms" or "top AI agencies 2026." Here are the top 10 most-cited competitors across our monitoring period:

What the Top-Cited Companies Have in Common

Looking at why these companies get cited while others don't, a clear pattern emerges:

• Third-party validation: BCG, Deloitte, Accenture, and McKinsey are mentioned in Forbes, Harvard Business Review, Gartner reports, and countless industry publications. AI engines treat these external mentions as authority signals.
• Named enterprise clients: LeewayHertz and Markovate publicly name Fortune 500 clients on their websites. AI models treat named clients as proof of credibility.
• Directory listings: LeewayHertz appears on Clutch, GoodFirms, G2, and DesignRush with verified reviews. These directories are heavily weighted by AI training data.
• Consistent entity identity: Companies cited across multiple engines have consistent business information (name, services, location) across dozens of web sources.
• Query diversity matters: LeewayHertz appeared in 8 different query types despite having far fewer total mentions than BCG or Deloitte. Their broad web presence across directories, blog content, and case studies made them relevant to more query types.

The Local Advantage: Why "Pittsburgh" Queries Work

One of the most actionable findings: local queries had a 16.5% citation rate compared to 0% for generic national queries. When someone asks "AI consulting companies in Pittsburgh," the competitive field shrinks dramatically. Instead of competing with BCG and Deloitte, you're competing with a handful of local firms.

For small and mid-sized businesses, this means local AEO is where the opportunity is. Optimize your Google Business Profile, get listed in local directories, produce locally-relevant content, and ensure your city and service area are clearly stated across your web presence.

What This Means for Your AEO Strategy

Based on 2,342 data points, here are the actionable takeaways:

1. Prioritize Perplexity. It cites businesses at 2x the rate of ChatGPT and Claude, and it retrieves information in real time (not just from training data). Optimize your website content and structured data for Perplexity first.
2. Build external authority. AI engines heavily weight third-party mentions. Get listed on Clutch, GoodFirms, and industry-specific directories. Pursue press coverage and guest posts. One Forbes mention is worth more than 10 blog posts on your own site.
3. Own your local queries. If you serve a specific region, local AEO is your fastest path to citations. The competition is 10x lower than national queries.
4. Name your clients. Anonymous testimonials don't register with AI engines. Named clients, with their company and industry, are treated as credibility signals.
5. Don't ignore brand queries. Brand queries had a 100% citation rate. Make sure your website clearly communicates who you are, what you do, and where you're located — so when someone does ask an AI about your company, the response is accurate and compelling.
6. Track your citations. You can't improve what you don't measure. Set up automated monitoring to track when and where AI engines cite your business, and iterate based on the data.

Methodology

This analysis is based on 2,342 automated citation checks run between February 13 and April 3, 2026. We queried ChatGPT (via API), Claude (via API), and Perplexity (via API) with 12 standardized queries related to AI consulting, AEO, and related services. Citation detection used pattern matching on company name, domain, and brand variations. Competitor tracking monitored 12 named competitors across the same query set. All data is stored in a PostgreSQL database and queries are run on an automated twice-daily schedule.

PulseSpark AI runs this monitoring system for our own business and for clients. If you want to see how AI engines currently cite your business, contact us for a free AEO audit.