How to get your website cited in AI search answers

By HubGrubPublished 4 min read

On this page
  1. 1. Let the crawlers in
  2. 2. Put the important text in the HTML
  3. 3. Answer the question first
  4. 4. Add structured data
  5. 5. Describe your company the same way everywhere
  6. 6. Earn mentions elsewhere
  7. What not to bother with
  8. How to check your progress
  9. Questions

The short answer

There's no special trick. AI search tools cite pages they can crawl and understand, so let their crawlers in, keep important text in the HTML, answer specific questions near the top of the page, mark up your organization, articles and FAQs with structured data, and describe your company the same way everywhere. Mentions on other trusted sites then give them a reason to choose you.

AI search tools such as ChatGPT search, Google's AI Overviews and Perplexity answer questions by finding pages, reading them and citing the ones that answer best. Getting cited is mostly good SEO done carefully, with a few details that matter more than they used to.

1. Let the crawlers in

A page that can't be fetched can't be cited. Check two places.

robots.txt. Each AI company has its own crawler, and some have more than one:

Crawler

Used for

OAI-SearchBot

Finding pages for ChatGPT search answers

GPTBot

Collecting content to train OpenAI's models

PerplexityBot

Perplexity's search index

Claude-SearchBot

Finding pages for Claude's search answers

ClaudeBot

Collecting content to train Anthropic's models

Googlebot

Google Search, including AI Overviews

Google-Extended

Use of your content in Google's Gemini models, separate from Search

Blocking a training crawler such as GPTBot or ClaudeBot doesn't remove you from that company's search answers, and blocking Google-Extended doesn't remove you from Google Search. Decide on each one separately.

Your firewall or CDN. Bot protection can block AI crawlers even when robots.txt allows them. Check your server or CDN logs for these user agents to confirm they're getting through.

2. Put the important text in the HTML

Some crawlers don't run JavaScript, or run it only sometimes. If your page content only appears after scripts load, they may see an empty page. Use server-side rendering or static generation for anything you want cited, and check by viewing the page source rather than the rendered page.

3. Answer the question first

AI answers are built from passages, so write pages that contain clear, self-contained ones:

  • Put a direct two to four sentence answer near the top, before the detail.

  • Use headings that match how people ask, such as "How long does a HubSpot implementation take?".

  • Use tables for comparisons and numbered lists for steps.

  • Write specific facts, such as timelines, requirements and limits, rather than general claims.

  • Keep one topic per page, so it's clear what the page answers.

4. Add structured data

Structured data states facts in a form machines read without guessing.

  • Organization on your site, with your name, legal name, logo, location, contact details and sameAs links to your profiles on LinkedIn and directories.

  • Article or BlogPosting on guides, with the publish and update dates.

  • FAQPage where a visible FAQ is on the page.

  • BreadcrumbList to show where each page sits in your site.

Validate it with Google's Rich Results Test or the Schema.org validator, and only mark up what's visible on the page.

5. Describe your company the same way everywhere

AI tools combine what your site says with what other sites say about you. If your homepage, legal pages, LinkedIn page and directory listings describe you differently, they have to guess which is right, and may confuse you with another company of a similar name.

Write one sentence that says who you are, what you do and where you're based, and use it everywhere: your homepage, about page, legal pages, structured data and every external profile.

6. Earn mentions elsewhere

Being cited by AI depends on being trusted, and trust comes partly from other sites: directory profiles, reviews, partner pages, interviews and articles that mention you. Your own site can't create that, but it can make it easy, with clear facts, a stable name and pages worth linking to.

What not to bother with

  • Pages written only for AI. Hidden pages or text aimed at crawlers rather than people tend to backfire.

  • Keyword stuffing. It reads badly to people and machines.

  • Expecting llms.txt to do the work. It's cheap to add but optional. Crawlable, well-structured pages matter far more.

How to check your progress

  • Ask ChatGPT, Perplexity and Google the questions your buyers ask, and note who gets cited.

  • Check your server logs for visits from the crawlers above.

  • Watch Search Console for the pages and questions you're showing up for.

We build websites with all of this in place from the start. See our websites service, or read how to connect your website forms to HubSpot so the visitors AI search sends you become leads.

Questions

Does an llms.txt file help?

It's cheap to add and does no harm, but treat it as optional. Crawlable, clearly structured pages and consistent facts about your company matter far more.

Should we block GPTBot?

That's a business decision. GPTBot collects content for training OpenAI's models, while OAI-SearchBot is what ChatGPT search uses to find pages. You can block one and allow the other in robots.txt.

Is FAQ structured data still worth adding?

Google now shows FAQ rich results for only a small set of authoritative sites, but the markup still states your questions and answers in a form machines can read. Add it where the FAQ is visible on the page.

How long until AI tools pick up our changes?

It depends on how often each crawler revisits your site, which can be days or weeks. Keeping your sitemap up to date and linking new pages from existing ones helps them get found sooner.

More guides