# Does SEO Still Matter When Buyers Ask ChatGPT Instead of Google?

> Yes. Assistants that search cite pages their own crawlers can reach and index, but a Google ranking does not carry over to ChatGPT, Claude or Perplexity.

Published: 2026-11-19
Source: https://www.itgrows.ai/business/blog/does-seo-still-matter-for-ai-search
Publisher: ItGrows (https://www.itgrows.ai/business) — AI visibility for businesses

---

Yes, SEO still matters: when ChatGPT, Claude, Perplexity or Google's AI Overviews search the web before answering, they cite pages their crawlers have reached and indexed, so crawlable pages with useful content remain the entry ticket. What SEO no longer does is cover you everywhere, because a good Google ranking does not carry over to ChatGPT, Claude or Perplexity, which crawl with their own bots, and on many questions an assistant answers from what it already knows without searching at all.

That split decides where your effort goes. Some SEO work now pays off in Google and in the assistants at once, some pays off only in Google, and one part of the job, finding out whether assistants actually name you, has no equivalent in a rank tracker.

## Why do assistants that search still need crawlable, indexed pages?

An assistant that searches works like a search engine with a writer attached. It sends one or more queries to an index, pulls back pages, reads them and writes an answer that links to some of them. A page that is not in that index cannot be retrieved, however good it is.

Google describes this mechanism openly. Its guide to [optimizing for generative AI search](https://developers.google.com/search/docs/fundamentals/ai-optimization-guide) explains that AI Overviews and AI Mode rely on Google's core ranking systems to retrieve relevant pages from the Search index (retrieval-augmented generation, also called grounding), and that the model issues several related "fan-out" queries to fetch more results for one question. The answer is built from what that retrieval returns.

ChatGPT, Claude and Perplexity run their own crawlers for search, and each maker states what blocking its crawler costs. Google states the equivalent requirement for AI Overviews and AI Mode:

| Answer engine | Crawler behind its search results | What the maker says |
| --- | --- | --- |
| Google AI Overviews and AI Mode | Googlebot, feeding the Google Search index | A page must be indexed and eligible to be shown in Google Search with a snippet to appear as a supporting link ([Google](https://developers.google.com/search/docs/appearance/ai-features)) |
| ChatGPT search | OAI-SearchBot | Sites opted out of OAI-SearchBot will not be shown in ChatGPT search answers, though they can still appear as navigational links ([OpenAI](https://developers.openai.com/api/docs/bots)) |
| Claude | Claude-SearchBot | Disabling it prevents Anthropic from indexing the content for search and may reduce the site's visibility and accuracy in user search results ([Anthropic](https://support.claude.com/en/articles/8896518-does-anthropic-crawl-data-from-the-web-and-how-can-site-owners-block-the-crawler)) |
| Perplexity | PerplexityBot | It is designed to surface and link websites in Perplexity's search results, and Perplexity recommends allowing it ([Perplexity](https://docs.perplexity.ai/docs/resources/perplexity-crawlers)) |

So for AI search, the old technical SEO question "can the search engine reach and index this page?" has to be answered four times instead of once. A robots.txt rule, a firewall setting or a CDN bot filter that stops one of these crawlers can keep you out of that assistant's search results, whatever your Google position. Google's own page on AI features lists "ensuring that crawling is allowed in robots.txt, and by any CDN or hosting infrastructure" among the SEO fundamentals that remain worthwhile. How each assistant discovers pages in the first place is covered in [how AI crawlers and AI search find a website's pages](https://www.itgrows.ai/business/blog/how-ai-crawlers-find-your-website).

## What does Google's guidance say about SEO for AI features?

Google's answer to the question in the title is unusually blunt. Its generative AI guide asks "Is SEO still relevant for generative AI search?" and answers "In short, yes!", because those features are rooted in Google's core Search ranking and quality systems. It adds that, from Google Search's perspective, optimizing for generative AI search is optimizing for the search experience, "and thus still SEO", whether people call it AEO or GEO.

The technical bar is the ordinary one. According to the AI features page, there are no additional technical requirements beyond being indexed and eligible for a snippet, and meeting every requirement still does not guarantee that Google will crawl, index or serve a page. The generative AI guide adds one setting to check: the site must be included in Search generative AI features in Search Console.

The same guide lists tactics you can skip for Google Search:

- llms.txt and other "AI text files": Google Search ignores them, although the guide says it is fine to keep them for other services that use them.
- Breaking content into small "chunks" so AI can parse it.
- Rewriting pages in a special style for AI, or chasing every long-tail variation of a query, since its AI systems understand synonyms and general meaning.
- Seeking inauthentic mentions across the web.
- Special schema.org markup, which generative AI search does not require.

Two limits on this guidance matter. First, it covers Google Search, including AI Overviews and AI Mode, not ChatGPT, Claude or Perplexity, which crawl and rank on their own. Second, it describes eligibility, not selection: being indexed lets a page be considered, but it does not make the model pick it. The practical side of the Google part is in [how to show up in Google Gemini answers and AI Overviews](https://www.itgrows.ai/business/blog/how-to-show-up-in-gemini-and-ai-overviews).

## Where does classic ranking work stop helping?

Ranking work is built around one index and one results page. Three things break that link once the buyer asks an assistant instead.

**The assistant may not search at all.** In our 23 September 2026 test of 20 German business questions, the model answered 10 without searching at all and cited nobody; where it searched, it cited primary sources (laws, ministries, OECD, large accounting firms) on 9 of 10; firms were named only on 'who should I hire / how to choose' questions; two runs of the same question named two different sets of firms. When an answer comes without a search, no page position takes part in it; the model draws on what it learned in training.

**A Google index entry does not carry over.** On 19 October 2026 we asked assistants for the exact titles of 18 articles, each older than three weeks, published on three client sites. ChatGPT found 1 of 18; Claude found 7 of 18. Google had indexed 29 of 30 articles of one of those sites. Being in Google's index did not mean an assistant could find the page.

**Rank tracking measures a different output.** A position report tells you where a URL sits for a keyword in Google. It says nothing about whether an assistant names your business when a buyer asks which supplier to hire, and the 23 September test shows the named firms can change between two runs of the same question. Producing extra pages for every phrasing does not fix this either. Google's guide warns that creating content for every fan-out variation primarily to manipulate rankings or AI responses violates its scaled content abuse spam policy, and calls it an ineffective long-term strategy.

That test also shows where business names appear: only on "who should I hire" and "how to choose" questions. A page that ranks for an informational keyword can still be read and cited, but the buyer-intent question is where you need to check what the assistants say.

## Which SEO work still pays off in AI answers?

Most of the technical checklist now does double duty. In order of what usually breaks first:

1. **Crawl access for every search crawler.** Check robots.txt, the CDN and the firewall for Googlebot, OAI-SearchBot, Claude-SearchBot and PerplexityBot. Perplexity's documentation notes that a site behind a web application firewall may need to allow its bots explicitly, and gives rules for Cloudflare and AWS. A step-by-step version is in [how to allow GPTBot and other AI crawlers in robots.txt](https://www.itgrows.ai/business/blog/how-to-allow-gptbot-in-robots-txt).
2. **Proof that each crawler reaches your key pages.** Because ChatGPT, Claude and Perplexity crawl separately, a clean Google Search Console report is not enough. Look in your server logs for their search crawlers on the pages you want cited: service pages, pricing, case pages.
3. **Important content in text.** Google's AI features page lists making sure important content is available in textual form. Prices hidden in images, services described only in a video, or text that appears only after a click give every crawler less to read.
4. **Pages with first-hand substance.** Google's guide says that content people find unique, compelling and useful will likely influence a site's presence in generative AI search in the long run more than any other suggestion in it, and contrasts commodity content with experienced, expert takes. For a business, that means how you work, what you charge and why, which cases you take and which you refuse: specifics a model cannot produce from common knowledge.
5. **Less duplication and clean structure.** The guide notes that duplicate content can waste crawling resources, and the AI features page asks for structured data that matches the visible text. Neither is new; both now affect more than one engine.

## How should you split effort between SEO and AI visibility?

Treat it as one shared base and two separate scorecards rather than two competing budgets.

The shared base is everything in the previous section: access, indexing, text, useful pages. Do it once and it serves Google's blue links, AI Overviews and the search features of ChatGPT, Claude and Perplexity. If this base is broken, fix it before anything else, because no AI-specific tactic works on a page no crawler can read.

The Google-only part is position tracking, snippet tuning and keyword-variant pages. Keep what still brings organic visits, but do not read a strong position report as proof that assistants recommend you; the 19 October 2026 findability check shows how far the two can diverge.

The AI-only part is measurement against buyer questions. Write down the questions a buyer asks before choosing someone like you, put them to each assistant, and record who is named and which sites are cited. Since two runs of the same question can name different firms, ask each question more than once and compare only runs on the same question set. For its own features, Google offers a Generative AI performance report in Search Console; for the other assistants there is no console, so you have to ask them. [How to measure AI visibility over time](https://www.itgrows.ai/business/blog/how-to-measure-ai-visibility) describes a repeatable setup.

For a first look at your own site, the [free check](https://www.itgrows.ai/) generates five buyer questions from your homepage and shows how ChatGPT and Perplexity answer them; all five answers open once you create an account. ItGrows does the work on the client's own website so that ChatGPT, Claude, Gemini and Perplexity can find and cite it, and every two weeks puts the client's buyer questions to those four assistants and counts the answers that name the client and the sites they cite.

## Frequently Asked Questions

**Do I still need SEO if my buyers use ChatGPT instead of Google?**

Yes. When ChatGPT searches, it relies on pages its search crawler, OAI-SearchBot, can reach, and OpenAI states that sites opted out of that crawler are not shown in ChatGPT search answers. Crawlable, indexable pages with useful content are the precondition for being cited in a search answer.

**Does ranking first on Google mean ChatGPT will recommend me?**

No. A Google ranking reflects Google's index, while ChatGPT, Claude and Perplexity run their own crawlers, and on many questions an assistant answers without searching at all. Check the assistants directly with the questions your buyers ask.

**Do I need llms.txt or special schema markup to appear in AI answers?**

Not for Google: its guide says Google Search ignores llms.txt and similar files and that generative AI search requires no special schema.org markup. Google still recommends structured data because it helps a page become eligible for rich results, and an llms.txt kept for other services neither helps nor harms your visibility in Google Search.

**Can I block AI training crawlers and still appear in AI search?**

With OpenAI, Anthropic and Perplexity, yes. OpenAI treats OAI-SearchBot for search and GPTBot for training as independent settings, Anthropic separates ClaudeBot (training) from Claude-SearchBot (search), and Perplexity says PerplexityBot is not used to crawl content for AI foundation models.
