How do law firms show up in ChatGPT results?

By Moshe Dor, FounderLast updated July 7, 2026

A firm shows up in ChatGPT when the model can identify it as a distinct entity, trusts the sources that mention it, and finds quotable answers on its site. There is no placement to buy — OpenAI banned legal ads. Firms with consistent data, schema markup, and question-shaped content get named; the rest stay invisible.

What has to be true for ChatGPT to name your firm?

Three things at once. The model must resolve your firm as one unambiguous entity — identical name, address, and practice areas everywhere it looks. It must trust the sources mentioning you: your own crawlable pages, directories, and press. And it must find text it can quote directly, meaning short, factual, question-shaped answers rather than slogans.

Why don’t bigger or older firms automatically win?

Because ChatGPT rewards machine legibility, not budget or reputation. A twenty-lawyer firm with three name variants across directories, no schema, and JavaScript-rendered content is invisible to the model, while a smaller firm with clean entity data and quotable pages gets cited. Size helps only when it is expressed in signals a model can read.

What is the fastest path for a partner to start?

Measure first, then fix in order. Run your money queries across the major engines to see where you stand, then correct entity consistency, add schema, open AI-crawler access, and publish an answers hub. Retrieval-based engines can respond within weeks; ChatGPT citations follow over one to four months as crawl and refresh cycles catch up.

What ChatGPT needs vs what firms usually have
ChatGPT needs	Typical firm reality	Fix
One consistent entity	Name varies across directories	Entity / NAP cleanup
Machine-readable identity	No schema or broken schema	LegalService + FAQPage markup
Crawlable, quotable text	JavaScript brochureware	HTML answers hub, GPTBot allowed

See where your firm stands.

The $490 AI Visibility Audit answers this question about your firm, with verbatim engine results — 25 prompts, 4 engines, under 48 hours. Credits toward any install.

Run the $490 audit