industry guide

Generative Engine Optimization for Law Firms

Last updated

Legal is one of the highest-stakes verticals for AI recommendations: prospects ask assistants sensitive questions they would never type into a public search box, and the answer often names two or three firms before the prospect ever opens a browser tab. Assistants lean unusually hard on established legal directories and bar records to avoid recommending an unlicensed or disciplined attorney — which means your directory footprint and your practice-area pages carry more weight here than in almost any other industry.

What are buyers asking AI about law firms?

They ask for shortlists, honest prices, and help deciding — and the assistant’s reply names specific businesses. The 8 prompts below reflect how real buyers phrase law firms questions to ChatGPT, Gemini, and Perplexity; each one produces an answer that either includes you or a competitor.

  • >. best personal injury lawyer in Phoenix for a car accident that wasn’t my fault
  • >. how much does a DUI lawyer cost and is it worth it for a first offense
  • >. do I need a lawyer for an uncontested divorce or can I file myself
  • >. find me an estate planning attorney near Naperville who does flat-fee wills
  • >. my landlord won’t return my deposit — what kind of lawyer handles this and who’s good in Austin
  • >. which immigration law firms have the best reviews for H-1B cases
  • >. employment lawyer for wrongful termination — free consultation, works on contingency
  • >. who is the top medical malpractice attorney in Ohio and what do they charge

Which sources do AI assistants cite for law firms?

Avvo, Justia, FindLaw do the heavy lifting in grounded law firms answers, alongside your own site when it is machine-readable. Building presence where assistants already look beats polishing anywhere else first.

SourceWhy it shows up in answers
AvvoAttorney profiles with ratings, disciplinary history, and client reviews — assistants cite it for "best lawyer" and cost questions because every profile maps to a licensed attorney.
JustiaLawyer directory plus a huge corpus of plain-English legal explainers; grounded answers frequently quote Justia articles and then surface its directory listings.
FindLawThomson Reuters consumer legal content ranks for nearly every legal question, so retrieval pipelines pull it constantly — firm profiles ride along.
Martindale-HubbellPeer-review ratings (AV Preeminent) give assistants a credential signal they can safely repeat when asked for "top rated" counsel.
Super LawyersSelection-based lists by practice area and metro — the exact shape of "best X lawyer in Y" answers assistants are asked to produce.
State bar association directoriesThe authoritative license record. Assistants use bar directories to verify a firm exists and is in good standing before repeating its name.
Google Business Profile & reviewsLocation-qualified prompts ("near me", "in Phoenix") pull heavily from maps data and review volume/recency.

What schema.org markup fits law firms?

Start with Attorney as valid JSON-LD, then layer the types below. Typed structured data is how assistants disambiguate who you are, what you do, and where — before deciding whether to repeat your name.

  • Attorney

    The specific schema.org type for an individual lawyer or law practice — use it instead of generic LocalBusiness so assistants disambiguate you from same-named businesses.

  • LegalService

    Marks the firm as a legal-services provider; pair with areaServed and a service catalog (hasOfferCatalog) listing practice areas.

  • FAQPage

    Legal buyers ask cost/process/eligibility questions; FAQPage markup makes your answers the quotable unit assistants extract.

  • Person

    Individual attorney bios with alumniOf, knowsAbout, and memberOf (bar admissions) give assistants a verifiable expert entity to attach to the firm.

What GEO actions move the needle for law firms?

5 of the 10 actions below are high-impact. Work top-down: crawler access and machine-readable facts first, then the citation sources assistants already trust, then content shaped like the questions above.

  1. 01Publish a cost page per practice area

    high impact

    "How much does a DUI lawyer cost" is among the most common legal prompts. A page giving honest ranges, fee structures (hourly vs flat vs contingency), and what changes the number is extremely quotable — most firms refuse to publish this, so those who do get cited.

  2. 02Claim and complete Avvo, Justia, and FindLaw profiles

    high impact

    These three directories dominate grounded legal answers. Profiles should match your site exactly on name, address, phone, and practice areas — mismatches read as different entities.

  3. 03One page per practice area per metro

    high impact

    Assistants answer "employment lawyer in Denver" from pages scoped exactly that way. A single "Practice Areas" page cannot win location-qualified prompts.

  4. 04Add Attorney + LegalService JSON-LD with full NAP

    high impact

    Include name, address, telephone, areaServed, and sameAs links to your bar listing and directory profiles so every mention resolves to one entity.

  5. 05Structure attorney bios as credential records

    medium

    Bar admissions, years practicing, notable results, publications — as lists, not prose. Assistants verify before they recommend; make verification trivial.

  6. 06Answer the "do I need a lawyer for X" questions honestly

    medium

    Content that sometimes answers "no, you can file this yourself" earns the trust signals assistants weight — and gets cited for the borderline cases where the answer is yes.

  7. 07Publish case results with concrete numbers

    medium

    Settlement and verdict figures (where bar rules permit) are the statistics generative engines measurably prefer to quote. Follow your state bar’s advertising rules on disclaimers.

  8. 08Keep review velocity on Google and Avvo

    medium

    Review recency and volume drive location-qualified answers. A steady request process after matter close beats a one-time review push.

  9. 09Allow AI retrieval bots in robots.txt

    high impact

    Many firm sites ship WordPress security plugins that block unknown bots wholesale — verify OAI-SearchBot, PerplexityBot, and Claude-User are not blocked, or grounded answers cannot read your pages at all.

  10. 10Mark up FAQs on every practice-area page

    medium

    Four to six real client questions with 1–3 sentence answers, wrapped in FAQPage JSON-LD. This is the single most extractable content shape.

Why does AI visibility matter now?

Because discovery has already shifted: fewer clicks from classic search, more decisions made inside AI answers. Every figure below is independently published and linked — the same sourcing standard this wiki recommends for your own pages.

  • 8% vs 15%Google users clicked a traditional result link on only 8% of searches that showed an AI summary, versus 15% without one — and clicked a source cited inside the summary just 1% of the time. Pew Research Center, July 2025
  • 900M weekly usersChatGPT reached roughly 900 million weekly active users in early 2026, more than doubling from about 400 million a year earlier. TechCrunch, February 2026
  • 4.4x valueThe average visitor arriving from an AI search source converts at roughly 4.4 times the value of a traditional organic search visitor. Semrush, 2025
  • up to +40%Adding citations, quotations, and statistics to pages improved visibility in generative engine responses by up to 40% in the original GEO benchmark study. Aggarwal et al., GEO: Generative Engine Optimization (KDD 2024), 2024

Frequently asked questions

Do AI assistants actually recommend specific law firms?

Yes. Ask ChatGPT or Perplexity for "best personal injury lawyer in [city]" and you get named firms with reasons, usually assembled from Avvo, Justia, Super Lawyers, and Google reviews. If your firm is absent from those sources, you are absent from the answer.

Which directories matter most for legal GEO?

Avvo, Justia, and FindLaw appear most often in grounded legal answers, with Martindale-Hubbell and Super Lawyers supplying the "top rated" credential language. Your state bar listing is the trust anchor assistants use for verification.

Can a small firm outrank BigLaw in AI answers?

In local and practice-specific prompts, yes. Assistants answer "estate planning attorney in Boise" from local signals — directory profiles, reviews, and city-scoped pages — where a focused small firm can be the best-documented option.

Is publishing legal fees a bar-compliance problem?

Publishing typical ranges with clear disclaimers is permitted in most U.S. jurisdictions, but advertising rules differ by state — check your bar’s rules first. Firms that publish honest ranges dominate cost-question citations precisely because so few do.

How do I check what AI currently says about my firm?

Ask the assistants your clients use the questions they would ask — practice area plus city — and record whether you are named. GEOExtension automates this: a frozen question set probed across ChatGPT, Claude, Gemini, and Perplexity, reported as a mention rate with confidence intervals.

Do client reviews influence AI recommendations?

Strongly. Review volume, recency, and rating on Google and Avvo feed both the retrieval sources and the reasoning assistants apply when choosing which two or three firms to name from dozens of candidates.

see where you stand

Is AI already recommending your business?

Run the free audit to score any page against the 19 GEO checks this wiki teaches — no account, no API keys. Then probe real ChatGPT, Claude, Gemini, and Perplexity answers with your own keys to measure your actual mention rate.

Written and maintained by the GEOExtension team. Every statistic on this page links to its source; recommendations mirror the checks in our free GEO audit.