THE MECHANIC

Query Fan-Out

When a client asks AI for a law firm recommendation, it doesn't run one search — it fans out into 8 to 12 parallel queries simultaneously. Your firm needs to answer many of them, not just one.

The search AI runs before answering

When a corporate counsel asks ChatGPT "who's the best securities litigation firm in New York," the AI doesn't run a single search. It generates and executes multiple parallel sub-queries before composing its response. This process — query fan-out — is one of the main reasons a firm can rank #1 on Google and barely appear in AI responses.

A typical set of sub-queries for that single question might include: "top securities litigation firms New York City," "best securities class action attorneys NY," "SEC enforcement defense lawyers New York," "securities litigation firm client reviews," "Am Law 100 securities litigation practice New York," "securities fraud attorney rankings," and more. Each sub-query pulls different results. AI synthesizes across all of them to form a recommendation.

"Who's the best securities litigation firm in New York?"
top securities litigation firms NYC
best securities class action attorneys NY
SEC defense counsel New York
securities litigation client reviews
Am Law securities practice rankings
securities fraud attorney NY

One client question fans out into 6–12 parallel sub-queries. Each pulls different sources. AI synthesizes across all of them.

Why most law firms only answer one

Traditional SEO optimizes for one keyword at a time. A firm invests in ranking for "New York securities litigation attorney" — and succeeds. On Google, that's a win. In AI search, it means the firm answers one of the twelve sub-queries well and is invisible on the other eleven.

Put simply: AI rewards firms that show up consistently across many related queries, not just firms that dominate one.

AI uses a scoring model called Reciprocal Rank Fusion (RRF) that aggregates rankings across multiple queries. The formula:

Reciprocal Rank Fusion Formula

RRF Score = Σ (1 / (60 + rank position))

A firm ranked #1 for one query scores approximately 0.016. A firm ranked between #4 and #7 across five related queries scores approximately 0.077 — nearly five times higher in AI's aggregated recommendation logic.

The firms that appear consistently across the sub-queries aren't always the ones with the best credentials. They're the ones with the broadest, most structured content across the topic.

Firm A

SEO-optimized

1 page: "Securities Litigation Attorney NYC" — ranked #1

0.016

RRF Score

Appears in some responses

How to close the gap for your practice

Closing the query fan-out gap means building structured content across the full landscape a client might explore — not just the primary practice area keyword. For a securities litigation firm, that means pages covering: securities class action representation, SEC enforcement defense, securities fraud attorney, investor disputes, and securities arbitration — each with FAQ schema, specific credentials, and explicit geographic context.

The goal isn't ten thin pages. It's five to eight substantive pages that each answer a specific client question directly, linked to each other, and structured with schema markup that lets AI extract specific facts rather than having to infer them from marketing copy.

Frequently asked questions

What is query fan-out in AI search?

Query fan-out is the process by which AI expands a single user question into multiple parallel sub-queries before forming a response. When someone asks an AI assistant for a law firm recommendation, the AI runs 8–12 related searches simultaneously and synthesizes the results. A firm that appears consistently across those sub-queries gets recommended more often than a firm that dominates just one.

How many sub-queries does AI generate from one question?

Typically 6–12, depending on the platform and query complexity. ChatGPT and Perplexity tend to generate more sub-queries for competitive categories (litigation, M&A, IP) than for highly specialized niche queries. The exact sub-queries aren't visible to users, but can be inferred from the pattern of sources AI cites in its response.

How do law firms benefit from understanding query fan-out?

Understanding query fan-out shifts the content strategy from "rank for one keyword" to "cover the topic." A firm that builds structured, answer-ready content across 6–8 related queries in its practice area will appear in more AI recommendation responses than a firm with one highly-ranked page — even if the second firm has better credentials or a stronger brand.

FREE · 15 MINUTES · NO OBLIGATION

Does AI recommend your firm right now?

The audit maps exactly where your firm stands across ChatGPT, Perplexity, Claude, Gemini, and Copilot — and what's keeping you off the shortlist.

Get Your Complimentary Audit