Book a call

Can schema get me into ChatGPT answers?

llms.txt and schema, in plain words

Both take about half a day, and llms.txt alone takes 15 minutes. No study yet links llms.txt to being cited. What each one does, in plain words.

By Saiful Islam Saif, Founder
· 5 min read

A woman in a red jacket writing code on a laptop by a window
Photo: Christina @ wocintechchat.com, Unsplash

What is schema?

A small block of code on a page that states the facts in a form machines can read.

A person sees “By Saiful Islam Saif, Founder”. Schema tells the search engine: this is an Article, the author is this Person, the publisher is this Organization, published on this date.

It is written in a format called JSON-LD and sits in the page head. Visitors never see it.

A short example for a company:

Organization schema, JSON-LD
{"@type": "Organization", "name": "Your Company", "url": "https://yourcompany.com", "sameAs": ["https://www.linkedin.com/company/yourcompany"]}

Which schema does a B2B site need?

  • Organization on the home page. Name, website, logo, LinkedIn page.
  • Person for the founder or authors. Name, role, LinkedIn profile.
  • Article on every blog post. Title, author, date.
  • FAQPage where the page has real questions and answers on it. Google retired FAQ rich results on 7 May 2026, so this helps machines read the page, not your search listing.
  • BreadcrumbList so the engine knows where the page sits on your site.
  • Service on each service page, if you want to go one step further.

Every article on this site carries Article, FAQPage and BreadcrumbList schema.

What is llms.txt?

A plain text file at yourcompany.com/llms.txt. It says who you are in a line or two, then lists your key pages with a short note on each.

The idea: an AI tool reading your site knows where the good stuff is, without wading through menus and scripts.

llms.txt
# Your Company
> Outbound lead generation for B2B companies in Southeast Asia, the GCC and Europe.

## Key pages
- [Services](https://yourcompany.com/services): what we do and how it is priced
- [Case studies](https://yourcompany.com/case-studies): results by market

Here is the honest answer.

Google Search Central, AI features and your website1 says there are no extra requirements to appear in AI Overviews or AI Mode. It says you do not need new machine-readable files, AI text files or special schema. It still asks that structured data match the visible text on the page.

SE Ranking, llms.txt study of 300,000 domains2 checked about 300,000 domains in November 2025. 10.13% had an llms.txt file. The study found no link between having the file and being cited by AI models.

Rankability, llms.txt adoption data 20263 found 9.3% of the top 1,000 sites served one in September 2026. Its earlier June 2025 scan found 0.3%, but it used looser checks and a different list, so the two are not a like-for-like trend. The same report says publishing one does not prove any gain in rankings or AI citations.

Exhibit 1About 1 in 11 of the top 1,000 sites now has an llms.txt fileShare of the top 1,000 websites with an llms.txt file, %

Separate snapshots. SE Ranking found 10.13% of about 300,000 domains had one, with no link to being cited.

Sources: Rankability, llms.txt adoption data 2026; SE Ranking, llms.txt study of 300,000 domains

Show the data as a table
SnapshotTop 1,000 sites with llms.txt
June 20250.3%
September 20269.3%

So should I bother?

Yes to schema. It is how search engines read your name, your authors and your dates correctly. That feeds everything else.

Yes to llms.txt too, but only because it takes about 15 minutes. Do not expect it to move anything on its own.

The real work is elsewhere. Pages that answer buyer questions clearly. Your name mentioned on other trusted sites. The same facts everywhere. We cover that in how to get recommended by ChatGPT.

It is a half-day job, not a monthly retainer.

What is the half-day setup?

  1. Add Organization schema to your home page.
  2. Add Article schema to your blog posts, with a real author and date.
  3. Add FAQPage schema only where the questions are visible on the page.
  4. Test a page in Google’s Rich Results Test. Fix any errors.
  5. Write an llms.txt with one line about you and your 5 to 10 best pages.
  6. Make sure your robots.txt does not block the AI crawlers you want to be read by.

What mistakes should I avoid?

  • Schema that says things the page does not. Fake reviews or star ratings in schema can get you penalised.
  • FAQ schema on every page. Only where there are real questions.
  • Paying for “AI optimisation” that is just these two files. It is about half a day of work, not a monthly retainer.

An AI compliance platform in Ireland4 was up against established vendors. In markets like that, clean facts and real proof on your own site matter, because the buyer will check you against them.

Questions people ask

Is llms.txt an official standard?

No. It is a proposal that some sites and tools have adopted. No major AI company has said it relies on it for citations.

Can schema get me into ChatGPT answers?

Not on its own. It helps machines read your facts correctly. Being cited depends on content and mentions.

Do I need a developer?

Not always. Most website builders have schema settings or plugins. llms.txt is a text file you upload.

Should I block AI crawlers?

Only if you do not want to be found in AI answers. For most B2B companies, being read is the point.

Sources

  1. 1 Google Search Central, AI features and your website
  2. 2 SE Ranking, llms.txt study of 300,000 domains
  3. 3 Rankability, llms.txt adoption data 2026
  4. 4 AI compliance platform in Ireland

The numbers in the text point here. Every figure above is from these sources, as the article cites them.

Saiful Islam Saif

Saiful Islam Saif, Founder, Fable InLead. Has led 200+ B2B outbound projects across Southeast Asia, the GCC, Europe, Australia and the USA. LinkedIn

Want a straight read on your own outbound?

A 30-minute call with Saif.

Book a free Outbound Review