LLM optimization (LLMO) is the work of making your pages reachable, quotable and credible for the AI systems that answer questions, such as ChatGPT search, Perplexity and Google’s AI Overviews. In practice it comes down to three checks: the AI crawlers can fetch your pages, each page states its answer plainly near the top, and the claims on it are specific and sourced.
What LLM optimization covers, and what it doesn’t
Traditional SEO decides whether a page ranks. LLM optimization decides whether an assistant can lift a correct sentence from that page and name you as the source. The two overlap heavily. Google’s guidance for site owners says there are no additional requirements to appear in AI Overviews or AI Mode beyond being indexed and eligible to show a snippet, and that there is no special schema.org markup to add.
So LLMO is not a separate bag of tricks. It is ordinary SEO, plus attention to three things that matter more when a machine is doing the reading: access for each AI crawler, answers written so they can be quoted on their own, and evidence that makes a claim safe to repeat.
Step 1: Check which AI crawlers can reach your site
Each assistant uses its own crawler. Blocking one in robots.txt, or at your CDN or firewall, removes you from that assistant’s answers, and the names are easy to confuse.
| Crawler | Company | What it does | If you block it |
|---|---|---|---|
| OAI-SearchBot | OpenAI | Indexes pages for ChatGPT search | Your pages are not shown in ChatGPT search answers |
| GPTBot | OpenAI | Collects content that may train OpenAI’s models | Training opt-out only; separate from search |
| ChatGPT-User | OpenAI | Visits a page when a user asks ChatGPT about it | ChatGPT can’t open your page on request |
| PerplexityBot | Perplexity | Indexes pages for Perplexity results; not used for model training | You drop out of Perplexity’s sources |
| Googlebot | Crawls for Search, including AI Overviews and AI Mode | You drop out of Google Search entirely |
Sources: OpenAI’s crawler documentation and Perplexity’s crawler documentation. Because GPTBot and OAI-SearchBot are separate, you can keep your content out of model training while staying visible in ChatGPT search.
Try this: open yoursite.com/robots.txt and search for each name above. Then ask your host or CDN whether a bot-protection rule blocks AI crawlers at the edge; robots.txt can allow a bot that the firewall still turns away.
Step 2: Put the answer in the first two sentences
Assistants quote sentences, not pages. A page whose opening paragraph answers its title question hands the model a clean sentence to lift; a page that opens with a statistic or a story makes it dig, and it may quote someone else instead. Apply the same rule under every H2: the direct answer first, then the detail it depends on.
- Write H2s as the questions buyers ask. “How long does a corporate video take to produce?” beats “Production Timelines.”
- Use a table for comparisons. Rows and columns are easy to extract and hard to misread.
- Keep important content in text. Google lists this explicitly; a price list or process shown only in an image or video is invisible to text extraction.
- Keep structured data consistent with the page. FAQ markup that says something different from the visible answer is a trust problem, not a shortcut.
Step 3: Add the evidence AI engines favor
The most useful public test of this is the GEO study by researchers at Princeton and other institutions (first published 2023, revised 2024), which rewrote web pages in different ways and measured how visible each version was in generative engine answers. Adding quotations from relevant sources produced the largest gain, about 40% on the study’s main visibility measure. Adding statistics and citing sources also helped. Keyword stuffing scored below the unedited pages.
For a services business, that translates into four habits:
- Give every figure its source organization, its year and a link to the exact page it came from.
- Quote a named person or document when you make a claim that matters, such as a regulation, a vendor’s spec or a client’s own words.
- Publish what only you know: your actual process, your turnaround times, the questions clients ask you, the mistakes you see on real projects.
- Cut what you can’t support. An unsourced percentage is a reason for an assistant to skip you, not cite you.
Step 4: Make your brand facts consistent everywhere
Assistants build an answer about your company from your site and from the other pages that mention you. When your Google Business Profile, LinkedIn page, directory listings and About page describe different services or locations, the model is working from conflicting inputs and may repeat the wrong one.
Write one description of what you do, who you serve and where you work, and use it everywhere. Add Organization markup to your homepage with sameAs links to your official profiles so search engines can connect them.
Step 5: Measure citations, not just rankings
Rankings won’t tell you whether assistants are using your content. Two checks will:
- Referral traffic. In Google Analytics 4, open Traffic acquisition and look for sessions from chatgpt.com, perplexity.ai and gemini.google.com. The volume is small for most sites, but the trend shows whether citations are growing.
- A monthly citation check. Write down 10 to 20 questions your buyers actually ask. Once a month, ask each in ChatGPT search, Perplexity and Google AI Mode, and record whether you’re cited and which page was used. Pages that are never cited are your rewrite list.
If you want help, our digital marketing team can run the crawler check and set up a citation baseline for your site. Contact iStudios Media or call (669) 287-2700.
Related reading: Predictive SEO 2.0: Map the Questions AI Search Asks Next and Topic Clusters for AI Search: How to Build Hub and Spoke.
Frequently Asked Questions
Is LLM optimization different from SEO?
Mostly not. Google says there are no additional requirements to appear in AI Overviews or AI Mode beyond being indexed and eligible for a snippet. The difference is emphasis: AI crawler access, answers that can be quoted on their own, and sourced evidence.
Should I block GPTBot?
Blocking GPTBot keeps your content out of OpenAI model training. It is a separate crawler from OAI-SearchBot, which controls whether you appear in ChatGPT search, so you can block one and allow the other.
Do I need special schema to appear in AI Overviews?
No. Google states there is no special schema.org structured data needed for AI Overviews or AI Mode. Keep any structured data you do use consistent with the visible text on the page.
How do I know if ChatGPT or Perplexity is citing my site?
Check Google Analytics 4 for referral sessions from chatgpt.com and perplexity.ai, and run a fixed list of buyer questions in each assistant every month, recording which of your pages are cited.
