When Someone Asks ChatGPT for a Lawyer, Are You Named?
Prospective clients increasingly ask an AI model before they ever run a search. "Who's a good personal injury lawyer in Atlanta" now returns a short list of named firms, and that list is not your Google rankings. Most firms have never checked whether they appear on it.
What changes at this scale
The shortlist got shorter
A results page shows ten options and the searcher compares. An AI answer names two or three firms and the prospect often stops there. Being absent removes you before any comparison happens, which is a harsher failure mode than ranking fourth.
Directories and reviews feed the answer
Models lean heavily on corroborating sources when naming professionals: legal directories, review platforms, bar listings. Inconsistent or thin presence across Avvo, Justia, FindLaw and Google suppresses citation even when the firm's own site is strong.
Bar rules still apply to what models repeat
If your site makes outcome claims a model then repeats in an answer, the underlying advertising rules have not gone away. Content gets structured so what is extractable is also what is compliant.
Attorney entities, not just the firm
Individual attorneys are entities models try to resolve, and prospects often ask about a named lawyer rather than a firm. Consistent attorney markup, bios and profile signals are what let a model answer that confidently.
Everything you get
- Citation baseline across ChatGPT, Claude, Perplexity and AI Overviews
- Prompt set built from how prospects actually ask for a lawyer
- Crawler accessibility audit for GPTBot, ClaudeBot and PerplexityBot
- Attorney and firm entity architecture with consistent identifiers
- Legal directory and review platform consistency audit
- Practice-area content restructured for passage-level extraction
- Compliance review of anything a model is likely to quote
- Schema for firm, attorneys, practice areas and locations
- llms.txt and AI crawler directives
- Competitor citation share by market and practice area
- Monitoring on a fixed schedule as model behaviour shifts
- Reporting on citations and mentions, not rankings
Common questions
How do I find out if we're being cited today?
Ask the models the questions your prospects ask and record what comes back. There is a free AI visibility scan on this site that does a version of it. Most firms are surprised, in both directions.
Our Google rankings are strong. Doesn't that carry over?
Partly, but less than you would expect. Strong rankings alongside zero AI citations is common, most often because the site renders client-side and the AI crawlers receive an empty page. Worth checking rather than assuming.
Can this create bar compliance problems?
It can if nobody is watching what is being made extractable. If your site carries outcome claims, a model may repeat them in an answer. Reviewing what is likely to be quoted is part of the work rather than an afterthought.
Is this worth doing now or is it too early?
For most practice areas it is early enough that competition is thin, which is precisely why it is cheap to win now. It is also small enough as a channel that I would not fund it ahead of fixing intake or a site that cannot convert.
Does this overlap with our SEO work?
Substantially, and that is a good thing. Crawler accessibility, schema and directory consistency serve both. The additions are entity disambiguation for individual attorneys, passage-level structuring and measurement of a channel your current reporting does not cover.
Want a straight read on where you actually stand?
Book a free discovery call. I'll come having already looked, and you'll get the findings either way.