Structured data and llms.txt: what they do and don't do

Watch · 3:00

• 0:00Why this topic attracts myths
• 0:11Structured data
• 1:25llms.txt
• 2:09A sensible order of work
• 2:31How AI Rank Lab helps

In shortStructured data (schema.org markup, usually written as JSON-LD) labels the facts on a page so machines can read them without guessing. It helps with Google rich results and clear entity facts, but Google says no special schema is needed for its AI features. llms.txt is a proposed Markdown file that lists a site's key pages for AI tools. Google Search doesn't use it, and no file or tag guarantees citations.

Why this topic attracts myths

Structured data and llms.txt are easy to sell: add a file, get cited. The truth is narrower. Both are ways of making facts machine-readable. They help when a system actually reads them, and they do nothing for content that isn't worth citing in the first place. This lesson separates what each one is documented to do from what is claimed.

Structured data in plain English

Structured data is a block of code on a page that states facts in a fixed vocabulary, schema.org. A product page can say, in a form any parser understands, "this is a Product, its price is 49 USD, it is in stock, its rating is 4.6 from 212 reviews". An article can state its author and publish date. An organization can state its legal name, logo and official profiles. JSON-LD is the most common format: a <script type="application/ld+json"> block that sits apart from your visible HTML.

What it reliably does: it makes your pages eligible for Google rich results (product, review, breadcrumb and others), and it removes ambiguity about facts such as your brand's exact name. AI Rank Lab's own writer, for example, reads Organization.name from your JSON-LD when it works out how to spell your brand.

What Google says about AI: "Structured data isn't required for generative AI search, and there's no special schema.org markup you need to add." It still recommends using it as part of normal SEO, because it helps pages become eligible for rich results (Google Search Central (opens in a new tab)). Google's general structured data guidelines add that you shouldn't mark up content readers can't see on the page (guidelines (opens in a new tab)).

llms.txt in plain English

llms.txt is a proposal published by Jeremy Howard of Answer.AI in September 2024 (llmstxt.org (opens in a new tab)). It is a Markdown file at /llms.txt that gives AI tools a short map of a site. The only required part is an H1 with the site's name. Then comes an optional one-line summary in a blockquote, optional notes, and H2 sections that list links with a short description each. A section called "Optional" marks links that can be skipped.

Its status today: it is a community proposal, not a standard. Google says Google Search doesn't use llms.txt files, and that creating one "will neither harm nor help your site's visibility or rankings in Google Search". We have not found documentation from OpenAI, Anthropic or Perplexity saying their search or citation systems read it. Where it can help is with tools and agents that are explicitly pointed at it, such as coding assistants reading developer documentation.

What each file or tag actually does

A sensible order of work

1. Make sure search crawlers can reach your key pages (robots.txt, firewall, sitemap).
2. Add Organization markup to the homepage and the markup that matches each page type: Product, Article, LocalBusiness, BreadcrumbList.
3. Check that every marked-up fact matches the visible text. Mismatched or invented markup is a problem, not a shortcut.
4. If you like, add an llms.txt. It takes minutes. Then spend your real effort on the content itself (next lesson).

How AI Rank Lab helps

Under Content → Schema and AI files, the LLMs.txt generator builds a file that follows the llmstxt.org format, and the FAQ schema generator writes FAQPage JSON-LD from your questions and answers; validating existing markup there is free. The AEO and GEO checker reads a page's JSON-LD and checks whether /llms.txt exists and follows the format. A passing llms.txt check means the file is well formed, not that any engine will use it.