Schema markup is code you add to your website that labels facts about your firm, like your name, address, and practice areas, in a format machines can read without guessing. It helps AI search tools confirm facts about your firm. It does not make them recommend you. Content, mentions, and clarity do that work.
What Is Schema Markup, and Does It Help AI Search Find Your Firm?
Schema markup is a standardized way of labeling facts on a web page so that software, not just a human reader, can understand what they mean. Instead of a machine guessing whether "123 Main Street" on your page is your office or a courthouse you mentioned in a blog post, schema markup tags it explicitly as your firm's address.
The word "schema" here refers to a shared vocabulary of labels (called schema.org) that search engines, AI tools, and other software agree to recognize. When people say "structured data," they mean the same thing: information organized into a fixed, predictable format rather than left as plain prose a reader has to interpret.
AI tools use this labeled information to double-check facts they already suspect are true, not to decide who deserves a mention in the first place. A tool still needs a reason to consider your firm: content worth drawing from, and signals elsewhere that your firm exists and practices what it says it does. Schema markup makes the facts behind that content easier to confirm once a tool is already looking. For more on what actually drives those mentions, see our guide to AI SEO for law firms.
How Does Schema Markup Actually Work on a Page?
In practice, schema markup usually takes the form of a block of code called JSON-LD, sitting in the page's source code. A visitor never sees it. It does not change how the page looks. It sits quietly in the background describing the page's content in a format a program can parse directly, without having to read and interpret the visible text the way a person does.
The code names an "entity," meaning a specific thing the page is about: your law firm, an individual attorney, a physical office. It then lists that entity's properties: a name, an address, a phone number, a set of practice areas, hours of operation. Every property uses a label defined by the shared schema.org vocabulary, so a search engine or AI crawler reading your page and one reading a competitor's page both interpret "address" the same way.
Structured data, in this sense, is a standardized format for providing information about a page and classifying its content, which a search system can then use to better understand what's on the page (factId PS2). The same labeled facts are available to any crawler that chooses to read them, not just one company's.
Which Schema Types Matter Most for a Law Firm Website?
A handful of schema types cover most of what a law firm site needs to label.
You do not need every type on every page. A firm's homepage might carry Organization or LegalService information, individual attorney bios carry Person markup, and a practice area FAQ page carries FAQPage markup. Layering in types your page does not actually have content for is one of the mistakes covered further down.
Does Adding Schema Markup Get You Cited by ChatGPT, Gemini, or Perplexity?
No, not by itself. Schema markup helps a tool verify facts it already has reason to look at. It is not a mechanism for getting noticed in the first place. ChatGPT, Google Gemini, and Perplexity each work differently in terms of what they draw from and when, but none of them treats the presence of schema markup as a reason to recommend a firm. Our FAQ on whether ChatGPT cites law firm websites covers how citation actually works for that tool specifically.
What tends to matter more is whether a firm has clear, specific, well-organized content that answers the kinds of questions people actually ask, and whether other sources on the web describe the firm consistently. Schema markup supports that work by making the facts inside your own content unambiguous. It does not replace the content itself. If you are weighing where to put your effort first, our plain-English guide to GEO for law firms walks through what tends to matter more broadly.
Structured data is commonly described as one of several ways a site can help a search system understand a page; it is not described as something that determines inclusion in AI-generated answers on its own (factId PS1). Treat schema markup as groundwork, not as a lever that moves outcomes by itself.
How Do You Add Schema Markup to Your Firm's Site?
Adding schema markup involves a few concrete steps.
- Decide which entities on the page actually need labeling. Usually that's your firm as an organization, the attorneys named on the page, and the physical office location.
- Write or generate the JSON-LD code for each entity, using the property names defined in the schema.org vocabulary so the labels mean the same thing to any system reading them.
- Place that code in the page's head section, or wherever your content management system allows a custom code block. It does not need to sit near the visible content it describes.
- Validate the result with a structured data testing tool before publishing. Fix any errors it flags, and check the output by eye against the actual page to confirm nothing contradicts what a visitor sees.
Most law firm website platforms and page builders include some version of steps one through three built in, particularly for LocalBusiness-type information like address and hours. Custom-built sites usually need a developer, or someone comfortable editing a site's underlying code, to add and maintain it.
What Mistakes Make Schema Markup Useless or Risky?
The most common problem is markup that contradicts the visible page. If your schema lists an address or a practice area that doesn't match what a visitor actually reads, that mismatch undermines the very thing schema markup is supposed to provide: a reliable, confirmable fact. A tool that catches the contradiction has no reason to trust either version.
A second common problem is copy-pasted boilerplate. Templates and plugins often ship with placeholder values or generic descriptions that firms never go back and edit. Markup that still says "Example Law Firm" or lists a generic practice area your firm doesn't handle is worse than no markup at all.
A third is over-marking: adding schema types for content that isn't genuinely on the page, in an attempt to appear more comprehensive than the page actually is. This tends to create exactly the kind of mismatch the first problem describes.
One area deserves specific caution: Review and AggregateRating markup. Any claim about client reviews, ratings, case results, or testimonials on your site, marked up or not, needs to follow your state bar's advertising and professional conduct rules. Those rules vary by state and we are not the authority on them. Check your own state bar's current guidance before publishing anything that describes outcomes or client feedback.
Is Schema Markup the Same Thing as llms.txt?
No, they are different tools that solve different problems. Schema markup labels specific facts inside a page, using a shared vocabulary so a machine can read "this is the phone number" or "this is the attorney's name" without guessing. It operates at the level of individual pieces of information.
llms.txt is a different idea entirely: a proposed plain-text file, written in markdown, that sits at the root of a website and gives AI crawlers a short, curated list of the site's most important pages and what they cover (factId PS5). Where schema markup describes facts within a page, llms.txt is meant to help a crawler decide where to look across an entire site.
Neither one is required, and adoption of llms.txt specifically is still inconsistent across AI tools. A firm can use schema markup, an llms.txt file, both, or neither, depending on how its site is built and how much upkeep it can commit to. If you're deciding where to spend limited time, schema markup on your core pages (home, attorney bios, practice areas) is the more established starting point.
Check What AI Tools Currently Say About Your Firm
Schema markup, like most of what's in this guide, is groundwork. It doesn't tell you whether it's working. The only way to know that is to look at what ChatGPT, Google Gemini, and Perplexity currently say when someone asks about a firm like yours, in your practice area and your city. Our guide on how to monitor ChatGPT mentions of your law firm walks through how to check this yourself on an ongoing basis.
If you want a faster read on where your firm stands right now, a free AI visibility check shows what these tools currently say about your firm, before you spend time on markup, content, or anything else.