Author: Paul J Bruemmer
Sources: BOND AI Trends (May 2025), Chris Long, Koray Tugberk GUBUR, and John Lyotier (LinkedIn Posts), and ChatGPT
The Surge of ChatGPT Referrals: What’s Driving It?
ChatGPT (Open AI's Generative Pre-trained Transformer) is a type of large language model (LLM).
Perplexity.ai doesn’t just rely on what it’s been trained on—it actively searches the web in real time and pulls information from multiple sources, making it a smart mix between a chatbot and a search engine.
Google’s AI Overviews (formerly known as the Search Generative Experience, or SGE) are AI‑generated summaries that appear at the very top of Google Search results—often referred to as “position zero.”
These and others including Microsoft Copilot (formerly Bing Chat) and Claude – by Anthropic are no longer novelty tools. They are transforming how users discover brands, services, and expertise.
Each of these tools has distinct use case applications. Despite their differing strengths, there are key commonalities between ChatGPT, Perplexity, Claude, and Microsoft Copilot. Their shared traits help define the broader category of AI research and reasoning assistants. They’re all intelligence amplification tools, not just search replacements. They help users think better, faster, and with broader reach—across domains. in a nutshell, they are productivity accelerators.
John Lyotier, Co-Founder & CEO of TravelAI™, noted in 2025 that AI-driven referral traffic is now matching—or even outpacing—both organic and paid channels. What’s more, these AI referrals are delivering up to three times the conversion rates of traditional sources. That’s a major shift in how traffic—and value—is being generated.
The key variable? IMHO I believe trust has everything to do with it, and perhaps question-matching relevance as well.
As BOND’s May 2025 report documents, the explosion of generative AI usage is powered by accelerating data pipelines and a shift from static search to dynamic, personalized information retrieval. Plaintiff trial lawyers, and everyone else from A-Z, can no longer afford to focus solely on SEO tactics. They must adapt to how LLMs retrieve and prioritize content.
The AI Referral Funnel: It Starts With the Right Input
AI systems like ChatGPT are trained on a vast corpus of content—but their output leans heavily on sources that show consistent relevance, authority, and question-matching fidelity. This is where my theory of "Long-Tail Intake Listening" becomes central.
Rather than relying on traditional keyword research, plaintiff law firms, and everyone else from A-Z, should harvest the exact queries their clients are asking in:
- Call logs
- Chat transcripts
- Intake forms ("What did you Google before calling us?")
- Internal site searches
- Reddit, Quora, and Avvo phrasing
These inputs map precisely to the kinds of long-form, conversational prompts users feed into ChatGPT and similar tools.
According to Chris Long VP of Marketing at Go Fish Digital, using regex filters in Google Search Console can surface queries with 15+ words, offering a window into conversational intent—the very DNA of AI query logs.
Entity Level Authority: The New Power Play
As I pointed out in AI Overviews for Plaintiff Trial Lawyers, traditional rank tracking is now unreliable due to anonymized GSC data and AI-generated summaries that compress multiple sources into a single answer. What matters more is entity-level authority.
AIOverviews (which BTW has some major learning to do) and ChatGPT (light years ahead of AIO) will increasingly prioritize law firms, and everyone from A to Z, that:
- Are mentioned across court (industry) documents, PDFs, and press releases
- Are cited in legal trade (industry) publications (e.g., Law360, NLJ)
- Appear in news coverage, CLE decks, and expert commentary
- Are structured as machine-readable (selectable text in PDFs, schema-rich HTML)
BOND’s report shows that AI models reward content not only based on quality but get this... density of structured citations across trusted nodes. This is the algorithmic version of reputation!
For All SEO's You Are Not Dead. You Are Now Multi-Modal.
Rick Egan CMO Growth Strategist insight hits hard: the winners in 2025 SEO are those optimizing for knowledge retrieval, not keyword and content stuffing. For plaintiff attorneys this means:
- Structuring FAQ constellations from broad to hyper-specific
- Turning expert deposition quotes into PDF highlight reels
- Using HARO and Qwoted to seed expert commentary into news stories
- Embedding relevant verdict documents into webpages as downloadable, indexed evidence
Koray Tugberk GUBUR CEO and Founder of Holistic SEO & Digital adds that mentions are the new links. Being named—even without a backlink—across multiple high-trust sites teaches the LLMs your brand is relevant. He calls it "semantic dominance."
The more your law firm, and everyone from A yo Z, is semantically entangled with key legal (or industry) concepts ("wrongful death + Chicago + Romanucci"), the more often you’ll surface in ChatGPT and possibly AIOverview and other intelligence amplification tools.
Actions Plaintiff Law Firms, and everyone, Should Take Now
- Build a "Query Intelligence" Feedback Loop
- Add a field to your forms: "What question did you type into Google or ChatGPT before you found us?"
- Use this data to shape your blog titles, meta descriptions, and H2s.
- Deploy Machine-Readable Evidence
- When possible, upload structured, searchable PDFs of verdicts, filings, or expert reports.
- Host them on your firm’s site or trusted third-party platforms.
- Saturate Trusted Domains
- Secure citations in Law360, Trial Magazine, LegalTech News, and bar association journals.
- Each citation is a data signal that contributes to your LLM visibility score.
- Repurpose Old Content for AI Mode Discovery
- Turn case studies into FAQ clusters
- Convert slide decks into shareable PDFs
- Extract client Q&A into carousel posts on Instagram (now indexed by Google)
- Track AEO (Answer Engine Optimization) Signals
- Use tools like Profound, Evertune or Waikay to monitor if your brand is showing up in ChatGPT responses and topics.
- Combine this with CRM data to verify if AI mentions lead to consultations.
Final Word: AI Referrals Are Already Here. Are You Visible in Them?
If ChatGPT, Google’s AIOverview, Perplexity and all the others are the new front doors to your practice, then entity prominence, mention density, and long-tail intake data are your new marketing weapons. As BOND’s research makes clear, we’re in an era of "compounding technology"—where machine learning + real-world citations = accelerated trust.
In Lyotier’s words, the referrals you’re getting now are likely driven by work you did 6–12 months ago. The work you do today—aligning with how AI systems think—will determine if you’re mentioned, linked, or left behind.
Good Luck!