Law firm websites and AI search: what to fix first
How law firms can make practice area pages, attorney bios and LegalService schema readable for AI search, avoid WAF and geo-blocking mistakes, and stay accurate.
By Crawlito · 6 min read
In short
- AI assistants can only describe a firm's services accurately if practice area pages and attorney bios are reachable and readable in the HTML.
- LegalService is a schema.org subtype of LocalBusiness, with Attorney and Notary as more specific types, so Google's LocalBusiness guidance applies to law firm offices.
- Aggressive firewall rules and country blocking, often added after spam or attack attempts, can stop search and AI crawlers from reaching any page.
- Google treats legal topics as Your Money or Your Life, where content must be highly accurate and consistent with expert consensus.
- No technical change can make an assistant recommend or cite a firm; readiness removes the barriers that stop it reading the site at all.
On this page
For a law firm, AI search readiness comes down to three things: crawlers must be able to reach the site, each practice area and attorney must have a clear page with its facts in the HTML, and that content must be accurate enough to stand on its own when it is summarised. The most common blocker is not content at all; it is a firewall or country block that stops crawlers before they see a page.
People increasingly ask assistants questions like "which firms near me handle landlord disputes" or "who is the employment lawyer at a firm I was referred to". The assistant can only answer from what it can fetch and read. This guide covers the parts of a firm website that decide that, and how to check each one. It will not make any firm more likely to be cited or recommended; that depends on many things outside the site.
Make sure crawlers can reach the site at all
Law firm sites attract spam form submissions, login attempts on client portals and vulnerability scanners, so they are often protected by strict security settings. Three are worth checking first.
Country blocking. A firm that serves one state or country may block traffic from everywhere else. Search and AI crawlers do not necessarily crawl from your country. Google's crawler documentation, for example, says Google "egresses primarily from IP addresses in the United States", and may try other countries only if it detects that US requests are blocked. Block a country, and you may block the crawlers that would have read your pages. Our guide to geo-blocking and IP allowlists explains how to exempt verified crawlers.
Bot challenges on every page. JavaScript challenges and CAPTCHAs stop crawlers, because crawlers do not solve them. Keep challenges on the contact form, client portal and login, not on practice area pages. See CAPTCHAs and crawlers.
robots.txt left over from a redesign. A Disallow: / copied from a staging site blocks every compliant crawler.
A quick check from the command line:
for ua in "Mozilla/5.0 (compatible; Googlebot/2.1; +http://www.google.com/bot.html)" \
"OAI-SearchBot/1.0" "PerplexityBot/1.0" "Claude-SearchBot/1.0"; do
curl -s -o /dev/null -w "%{http_code} " -A "$ua" https://www.examplefirm.com/practice-areas/employment-law/
echo "${ua%%/*}"
done
curl -s https://www.examplefirm.com/robots.txtAnything other than 200 on a public page needs an explanation. A test from your own computer does not reproduce crawler IP addresses, so also check your hosting or CDN logs for crawler requests that were blocked.
Practice area pages: one topic, stated plainly
A practice area page is what an assistant reads when it tries to work out whether your firm handles a type of matter. Pages that bury that answer in generic copy are hard to use.
What a strong practice area page contains:
| Element | Why it matters |
|---|---|
| A specific H1, such as "Employment law for employees in Ohio" | States the service and, where relevant, the jurisdiction |
| A first paragraph saying who you help and with what | Gives a reader, human or machine, the answer immediately |
| The matters you take, and ones you do not | Prevents mismatched enquiries and vague summaries |
| Jurisdictions and courts where you practise | Legal answers depend on location |
| The attorneys who handle this work, linked to their bios | Connects the service to accountable people |
| How an initial consultation works, including any fee | Practical facts people ask about |
| A last-reviewed date for content that describes law | Signals whether the legal content is current |
Keep this text in the server-rendered HTML. Tabs and accordions are fine if the content is present in the HTML when the page loads; content fetched only after a click may never reach a crawler that skips JavaScript.
Avoid thin city pages that repeat the same text with a different place name. If you have a real office in a city, give it a real location page. If you only take cases from that area, say so on the practice area page instead. Our guide to local service pages covers when a separate area page is justified.
Attorney bios as the firm's evidence
Attorney pages carry much of a firm's credibility: who the lawyers are, where they are admitted, what they focus on. Make each bio:
- A separate page with its own URL, not a modal or a card that expands on the team page.
- Factual and checkable: bar admissions with jurisdiction, education, practice focus, languages, professional memberships and published work.
- Linked both ways: from relevant practice area pages, and from the bio back to them.
- Consistent with the attorney's details on bar association directories and the firm's other profiles.
Avoid superlatives in bios and elsewhere. Lawyer advertising rules in many jurisdictions restrict misleading or unverifiable claims about results and expertise, and an AI summary may repeat a claim like "the best injury lawyer in the state" without its context. Check your jurisdiction's rules with whoever handles compliance for the firm. Our guide to author pages and content credibility has more on bio structure.
LegalService and Attorney structured data
schema.org defines LegalService as "a business that provides legally-oriented services, advice and representation, e.g. law firms". It is a subtype of LocalBusiness, and its more specific types are Attorney and Notary. Because LegalService is a LocalBusiness, Google's LocalBusiness structured data guidance applies: name and address are required, and properties such as telephone, url, geo and openingHoursSpecification are recommended.
An office page might carry:
{
"@context": "https://schema.org",
"@type": "LegalService",
"name": "Example & Partners LLP, Columbus office",
"url": "https://www.examplefirm.com/offices/columbus/",
"telephone": "+1-614-555-0100",
"address": {
"@type": "PostalAddress",
"streetAddress": "100 Main Street, Suite 400",
"addressLocality": "Columbus",
"addressRegion": "OH",
"postalCode": "43215",
"addressCountry": "US"
},
"openingHoursSpecification": {
"@type": "OpeningHoursSpecification",
"dayOfWeek": ["Monday", "Tuesday", "Wednesday", "Thursday", "Friday"],
"opens": "08:30",
"closes": "17:30"
}
}For attorney bios, Person with properties such as jobTitle, worksFor, alumniOf, knowsAbout and memberOf describes the lawyer in a structured way. Google does not offer a rich result for attorney profiles, so treat Person markup as a clear statement of facts, not a route to a search feature. Mark up each office separately, and keep every value identical to the visible page. Our LocalBusiness schema guide covers hours, multiple locations and validation in detail.
Do not mark up client reviews of the firm on the firm's own site as aggregateRating. Google's guidance limits review markup on LocalBusiness types to sites that collect reviews about other businesses.
Accuracy for a Your Money or Your Life topic
Google's guidance on creating helpful content describes topics that can significantly affect people's finances, safety or well-being as "Your Money or Your Life" (YMYL), and says content on them must be highly accurate and consistent with established expert consensus. Legal information is squarely in that category.
For a firm, that translates into a few habits:
- Name the jurisdiction on every article that describes law. A filing deadline in one state is wrong in the next.
- Date your legal content and review it when the law changes. Show a "last reviewed" date and the reviewing attorney.
- Separate information from advice. Explain general rules and say clearly that they are not advice for a specific situation.
- Remove or update outdated posts about superseded statutes or old fee rules. An assistant does not know the post is from 2019 unless the page says so.
- Avoid outcome promises. "We win" or "you will receive compensation" claims are risky under advertising rules and misleading when quoted out of context.
These are good practice for clients first. They also mean that when any system summarises a page, there is less room for it to state something the firm would not stand behind.
Frequently asked questions
Should a law firm block AI crawlers?
That is a business decision. Blocking AI search crawlers such as OAI-SearchBot or PerplexityBot stops those products reading your pages, while blocking training crawlers such as GPTBot affects model training, not search answers. A common approach is to allow search crawlers and decide separately on training.
What schema type should a law firm use?
Use LegalService for offices, or Attorney where that more specific type fits, and Person for individual lawyers' bios. Mark up each office with its own address, phone number and hours.
Do AI assistants read attorney bios?
Assistants that search the web can fetch any public page they are allowed to crawl. A bio on its own URL, with facts in the HTML, is far easier to read than one inside a pop-up or loaded by script.
Can we be sure an assistant recommends our firm after these fixes?
No. These changes remove technical barriers and make your facts unambiguous. Which sources an assistant cites depends on many factors outside your website.
Check your firm's site
Start with the pages that describe what the firm does. Run the free check on a practice area page to see whether crawlers can reach and read it, or book an audit for a full review of access, structure and markup across the site.
Filed under Strategy and audits. Topics: local-seo, structured-data, bot-protection, waf, checklist