Do search engine rankings still matter in the age of AI? What generative engines prioritize

For two decades, Google decided which content people saw. Ranking in the top ten meant traffic, authority and customers. Today, ChatGPT, Gemini, Claude, Perplexity and Google’s own AI Overviews and AI Mode often give users a ready-made answer built from several sources instead of a list of links. That raises two questions: do rankings still matter, and if so, how much? And what do generative systems look at when they decide which content to use? This article answers both with the latest research.

Do search engine rankings still matter?

Short answer: yes. Rankings still drive most website traffic and still decide whether AI systems can find your page. But a top position no longer guarantees that AI will cite you.

• Rankings still drive traffic. In SE Ranking’s 2026 data, organic search accounted for 42.75% of all website traffic, compared with 0.32% from AI assistants. Google also reported its fastest query growth in five years in Q3 2026.
• Search indexes are the entry ticket for AI. Google’s AI Overviews and AI Mode draw on Google’s index. ChatGPT’s search uses third-party providers, with Bing named by OpenAI, plus its own OAI-SearchBot crawler. Perplexity runs its own index. A page that isn’t crawlable and indexed can’t be retrieved.
• Rankings for the right queries win retrieval. AI systems retrieve pages that rank for the questions they search, and retrieved pages are the only ones that can be cited.

SEO remains the foundation for everything else. In the AI era, it’s necessary but not sufficient.

How closely do AI citations follow Google rankings?

Less closely than they used to, and much less for standalone assistants than for Google’s own AI features:

AI system	Share of cited pages that rank in Google’s top 10 for the same query	Source
Google AI Overviews (July 2025)	76%	Ahrefs
Google AI Overviews (early 2026)	38%; 31% of citations don’t rank in the top 100	Ahrefs via DesignRush
Perplexity	28.6%	Ahrefs, Aug 2025
ChatGPT, Gemini, Copilot	around 6–9% each; 12% average across assistants	Ahrefs, Aug 2025

The main reason is query fan-out. AI systems split a question into several sub-queries, search for each, and combine the results. A page that ranks around position 6 for three related sub-questions can beat a page that ranks first for only one of them. The drop in AI Overviews coincided with Google making Gemini 3 the default model for AI Overviews on 27 January 2026.

Ranking still helps within that process. In AirOps’ study of 16,851 ChatGPT queries, pages in the first position of ChatGPT’s own retrieval were cited 58.4% of the time, compared with 14.2% for pages in position 10. Rankings matter, but for many queries at once, not just the head term.

How is generative search different from traditional search?

• Conversational queries. Users ask full questions (“How can I learn data analysis in 6 months?”) instead of typing keywords (“data analysis course”). Google says AI Mode queries are on average three times longer than traditional searches.
• Answer synthesis. The model reads many sources and writes one answer, citing only some of them.
• A shorter user journey. The answer appears in the AI interface, so fewer people click. Pew Research Center found that users clicked a regular result in 8% of visits when a Google AI summary appeared, compared with 15% without one.

For example, searching “best project management tools” in classic Google returns a list of comparison articles. In AI Overviews or ChatGPT, the user immediately gets a shortlist of tools with descriptions. Being one of the sources behind that shortlist, or one of the tools on it, is the new goal.

What is GEO (generative engine optimization)?

Generative engine optimization (GEO) is the set of practices that increase the chance that AI systems find, understand, mention and correctly cite your brand and content. It builds on SEO rather than replacing it. For the full picture, see our GEO guide.

	SEO	GEO
Main goal	Rank pages in search results	Be mentioned and cited in AI answers
Unit of competition	The page, for a keyword	The passage or fact, for a question and its sub-questions
Key signals	Relevance, links, technical health, user experience	Retrievability, focused answers, brand mentions, freshness
How you measure it	Rankings, clicks, conversions	Mention rate, citations, share of voice, AI referrals
Shared foundation	Crawlable, indexed, trustworthy, genuinely useful content	Same

Skip theory and jump straight to action: get the free AI Search Visibility checklist

The first step to GEO: assess and improve your brand’s visibility in AI search results.

Download free checklist

*By downloading the checklist, you enter into an agreement for the provision of digital content in exchange for your consent to receive marketing communications from us. Details are provided in the Terms and Conditions and the Privacy Policy. You may withdraw your consent at any time.

What do generative engines prioritise?

Language models don’t judge content intuitively. Their choices reflect retrieval systems and signals designed to approximate human judgments of reliability. Research so far points to seven priorities:

1. Retrievability. The page must be crawlable by search and AI bots and rank for at least one of the sub-queries the system searches. OpenAI states that sites blocking OAI-SearchBot won’t appear in ChatGPT search answers.
2. Passage-level relevance. In the AirOps study, pages whose headings closely matched the query were cited 41% of the time, compared with 29% for weak matches. Pages that covered part of a topic well were cited more often than pages trying to cover everything.
3. Corroboration across the web. In Ahrefs’ study of 75,000 brands, branded web mentions correlated with AI Overview visibility far more strongly (0.664) than backlinks (0.218). Information repeated by several trustworthy sources is easier to state confidently.
4. Freshness. Pages cited by AI assistants are on average 25.7% fresher than pages in Google’s organic results, with ChatGPT showing the strongest preference.
5. Accuracy and originality. Google’s AI features guide advises against recycling what others have said or what an AI model could easily produce. Specific, verifiable facts, original data and first-hand experience stand out.
6. Transparency and authority. Clear authorship, publication and update dates, and references help both people and systems judge reliability. See E-E-A-T in AI search.
7. Readable structure. Clear headings, lists, tables and short answer-first sections make passages easy to extract. Google notes you don’t need to break content into tiny chunks: write for readers, with a clear structure.

What doesn’t move the needle on its own:

• Adding schema markup: a controlled Ahrefs study of 1,885 pages found no meaningful citation lift.
• llms.txt and other AI-specific files: Google says you don’t need them to appear in generative AI search.
• Stuffing synonyms and keyword variations: Google’s systems understand meaning without them.

GEO strategies: how to prepare content for AI visibility

1. Build a clear structure

Use H2 and H3 headings phrased like real questions, open each section with a direct answer, and use lists, tables and step-by-step instructions where they help the reader.

2. Show who stands behind the content

Add author bios with documented expertise, cite sources, show publication and update dates and publish an editorial policy.

3. Be focused on the page and complete across the site

Make each page the best answer to one main question. Cover the related sub-questions with linked pages in a topic cluster, so the site as a whole is complete. See topical authority vs keyword density.

4. Keep structured data accurate, but don’t rely on it

Use Article, Organization, Product and Review markup where it fits, and keep it consistent with the visible page. Note that Google retired HowTo rich results in 2023 and FAQ rich results in May 2026, and there is no special markup for AI features. More in the role of structured data in AI search.

5. Keep content fresh

Review priority pages regularly, update facts and dates when you make substantial changes, and show a changelog for key resources. See how often to refresh content.

6. Publish first-hand expertise

Proprietary research, industry reports, case studies and practitioners’ experience give AI systems something worth citing that they can’t find elsewhere.

7. Earn mentions beyond your website

Digital PR, expert commentary, reviews and genuine participation in communities build the corroboration AI systems rely on. See backlinks and mentions in GEO.

What are the challenges of GEO?

• Opacity. AI providers don’t publish how they choose sources, so most knowledge comes from large-scale studies and testing.
• Volatility. Model updates change citation patterns quickly. The top-10 overlap in AI Overviews halved in the months around their switch to Gemini 3, and Reddit’s share of Perplexity citations fell from 25% to 7% between February and April 2026.
• Limited attribution. ChatGPT runs a web search for only part of its answers (34.5% of queries in February 2026), so many brand mentions come without links.
• Risk of distortion. AI may simplify or misstate your content, which is why consistent facts across the web matter.
• Big brands’ head start. Well-known brands are mentioned more often, though niche expertise and original data can still break through.

Why is it worth investing in GEO now?

• Reach. AI answers already reach billions of users. See the latest AI search statistics.
• Authority. Regular citations in AI answers strengthen your brand at the moment people form opinions.
• Competitive advantage. Many companies don’t yet measure how AI describes them, so early movers can shape the answers in their category.
• Future-proofing. Google is merging AI Overviews, AI Mode and classic search, so the skills you build now will be needed everywhere.

Before you start, read 5 questions worth asking before investing in AI search optimization.

How do you measure GEO effectiveness?

• Mention rate and share of voice: how often your brand appears, compared with competitors, for a fixed set of prompts across several AI tools.
• Citations: which of your URLs are cited, and for which questions.
• AI referral traffic and conversions in GA4 (chatgpt.com, perplexity.ai, gemini.google.com, copilot.microsoft.com).
• Branded search growth in Google Search Console.
• Direct and referral leads and mentions on high-trust domains.

Because AI answers vary between runs, repeat each prompt several times and track trends. Read more about metrics that matter in AI search.

SEO and GEO: coexistence, not rivalry

Google’s May 2026 guide states that, from its perspective, optimising for generative AI search is still SEO. In practice, the two work together:

• SEO provides the foundation: crawling, indexing, technical health, rankings for the questions people ask, and authority.
• GEO increases the chances of being chosen: focused answers, corroborating mentions, fresh facts and measurement inside AI tools.

Brands that combine both are visible in classic results and in conversational answers.

Summary

Search rankings haven’t lost their significance. They still bring most traffic and decide whether AI systems can find your content. But ranking first for one keyword no longer guarantees a citation: only 38% of AI Overview citations now come from Google’s top 10. To be chosen, content has to be retrievable for many related questions, focused, accurate, fresh and confirmed by other sources. GEO is the natural extension of SEO that addresses exactly that.

FAQ: search rankings and AI search

Do search engine rankings still matter for AI search?

Yes. AI systems retrieve pages through search indexes, and pages that rank for the questions they search are far more likely to be cited. But rankings are no longer enough: only 38% of pages cited in Google AI Overviews rank in the top 10 for the same query (Ahrefs, early 2026).

Do AI Overviews only cite top-ranking pages?

No. In early 2026, 38% of AI Overview citations came from top-10 pages, and 31% came from pages outside the top 100 for that keyword. AI Overviews search for several related sub-queries, so pages that rank for those can be cited too.

How does ChatGPT choose which sources to cite?

When ChatGPT searches the web, it rewrites the question into sub-queries, retrieves results through search providers such as Bing, and cites pages that answer the question clearly. Pages ranked higher in its retrieval are cited far more often, and pages with headings that match the question perform better.

What is the difference between SEO and GEO?

SEO aims to rank pages in search results. GEO aims to get your brand and content mentioned and cited in AI-generated answers. GEO builds on SEO, adding focused answers, off-site mentions, freshness and measurement in AI tools.

Does schema markup help with AI visibility?

Not directly. A controlled Ahrefs study found no meaningful citation lift after adding schema, and Google says there is no special markup for AI features. Keep accurate structured data for search features, but focus on content and authority.

How can I check whether AI systems cite my website?

Ask the questions your customers ask in ChatGPT, Gemini, Perplexity and Google AI Mode, several times each, and record whether your brand and URLs appear. For ongoing tracking, use an AI visibility monitoring tool and AI referral reports in GA4.

Sources

• Ahrefs: Only 12% of AI cited URLs rank in Google’s top 10 for the original prompt (Aug 2025)
• DesignRush: Ahrefs study, only 38% of AI Overview citations come from top 10 pages (Mar 2026)
• ALM Corp: AI Overview citations from top-10 pages dropped from 76% to 38% (2026)
• Search Engine Land: ChatGPT citations reward ranking and precision over length (AirOps) (2026)
• Google Search Central: Optimizing your website for generative AI features (May 2026)
• Ahrefs: AI Overview brand visibility factors, 75K brands (2025)
• Ahrefs: AI assistants prefer to cite fresher content (Jul 2025)
• Ahrefs: We tracked 1,885 pages adding schema (May 2026)
• Search Engine Journal: Google drops FAQ rich results from Search (May 2026)
• Pew Research Center: clicks when an AI summary appears (Jul 2025)
• SE Ranking: AI traffic grew 16x from 2024 to 2026 (Jun 2026)