Playbook

ChatGPT visibility for law firms: how the engine actually picks attorneys

ChatGPT picks lawyers on verifiable credentials: bar records, board certification, case-type fit, and peer-review directories — and in our measurement it has stated those criteria in its own answers, including the phrase "not advertising alone." It checks the state bar's discipline record and prints the result. It rarely mentions review counts at all. If your firm's marketing was built for billboards and search ads, this surface plays by different rules — and they're rules a credentialed firm can win.

Updated August 11, 2026Reading time 6 min

The engine told us its criteria

In one of our personal-injury panel runs, ChatGPT opened its recommendation list by stating its method: candidates chosen on “board certification, peer recognition, relevant experience, and Texas Bar records — not advertising alone.” That sentence, from the engine itself, is the whole strategic picture for legal marketing on this surface. The firms that dominate billboards and television in the same market appeared in none of our runs. Every firm ChatGPT named was board-certified.

In Nameworthy’s personal-injury panels, ChatGPT printed its own selection criteria inside the answer — board certification, peer recognition, relevant experience and state bar records, and explicitly not advertising alone.

The verification behavior is new, and it matters

Across our legal runs, ChatGPT didn’t just rank — it verified, and said so:

  • Bar discipline checks, printed in the answer. “The State Bar currently reports no public disciplinary history” appeared for pick after pick, cited to the bar’s own site. A clean record has become quotable evidence; a checkable one is table stakes.
  • The certification explained to the client. Answers now teach what board certification means and how rare it is, then filter the market by it.
  • Deadlines flagged from primary sources. Statute-of-limitations warnings cited to the state’s statutes site, down to municipal notice deadlines. The engine behaves like a cautious referral partner, not a directory.
  • Case-type routing. Different sub-specialists recommended for trucking, medical malpractice, and catastrophic-injury questions, with the follow-up offer to narrow further.

What separates the named from the unnamed

Holding a board certification was necessary in our runs but rarely sufficient. The firms ChatGPT actually named paired it with corroborated depth: peer-review directory recognition (Best Lawyers years, Super Lawyers spans, AV ratings, Avvo scores), decades of tenure, or a documented trial record. In family-law markets we’ve also measured directory-profile depth carrying strong ChatGPT presence for firms with well-maintained Avvo profiles. The pattern across both: ChatGPT trusts what third parties can confirm. A credential that lives only on your own site is an assertion; the same credential echoed by a directory and a registry is evidence.

The honest caveat: individual answers churn. Our published durability research shows recommendation lists reshuffling run to run on every engine, and ChatGPT is no exception — the stable thing in our legal panels wasn’t any single ranking, it was the selection logic. Optimize for the criteria, not for a screenshot.

What a firm can actually do

  1. Publish your certifications in liftable form — the exact certification name, the certifying body, the attorney it attaches to, with a link to the registry where it can be verified.
  2. Build the peer-directory layer. The directories ChatGPT cited in our runs are specific and few. Complete, current profiles there are corroboration; absence there is invisibility on this surface.
  3. Keep the bar record clean and claimable — and make sure your attorneys’ bar profiles are complete, since the engine reads them.
  4. Publish case-type depth. Sub-specialty routing is real: pages that document your trial record and case focus in concrete terms give the engine the “relevant experience” leg of its own stated criteria.
  5. Then measure. This surface’s answers differ sharply from Gemini’s, Perplexity’s, and Google’s AI surfaces in the same market — winning one says nothing about the others.

Common questions

How does ChatGPT choose which lawyers to recommend?
In our measurement of legal markets, ChatGPT's picks are driven by verifiable credentials: state bar records (including a live discipline check it often states in the answer), board certification, case-type fit, and recognition in peer-review directories like Best Lawyers, Super Lawyers, Martindale, and Avvo. In one run it stated its criteria outright: bar eligibility, board certification, relevant case focus, and peer-review directories, 'not advertising alone.' Review volume, which drives other surfaces, was rarely mentioned.
Why doesn't ChatGPT recommend my law firm even though we advertise heavily?
Because advertising isn't in its selection criteria, and it has said so in as many words in our runs. Firms built on TV and billboard spend can dominate other channels and still never appear in ChatGPT's answers, which favor board-certified attorneys with clean bar records and peer-directory recognition. The encouraging flip side: a smaller credentialed firm can outrank a much larger advertiser on this surface.
Does board certification alone get a lawyer recommended by ChatGPT?
Usually not by itself. In our panels, holding the certification was necessary but not sufficient: the firms ChatGPT actually named paired certification with corroborating depth — peer-directory listings, decades of tenure, or a documented trial record. A certification that exists only on your own website carries less weight than one corroborated by the directories and registries ChatGPT checks.

The report is free

See what ChatGPT answers for your firm's market.

Real client questions, run across the major engines, with the per-engine fix list. Free, in 48 hours.

Get your AI visibility report