Scam alert: Stridec does not recruit through WhatsApp, Signal, Telegram or personal email. Real Stridec email only comes from @stridec.com. Read our notice

How Does AI Change SEO: What’s Actually Different in 2026

AI changes SEO by adding a new layer on top of ranking: an AI engine now reads the pages that rank, decides which brands to name and which pages to cite, and writes the answer itself. Ranking still matters, but it is no longer the finish line. As of November 2026, the brands that win are the ones the AI understands most clearly and can quote most precisely, in Google AI Overviews and AI Mode, ChatGPT, Gemini, Claude and Bing Copilot.

For a working SEO, the change is practical, not philosophical. Indexing now rewards pages that say clearly what a business is, as much as pages that match keywords. Ranking now sits next to being named in the answer as a parallel goal, not a downstream result. The places people search have multiplied beyond Google’s results page. And thin, aggregator-style content has stopped paying, because AI engines build their answers from the substantial sources and ignore the rest.

This article walks through what is actually different, layer by layer, with what it means for your work. It describes the current state rather than predicting the future: what has changed in indexing, ranking, the AI engines, content economics and measurement, as of November 2026.

Key takeaways

  • Ranking and being named in AI answers are now parallel goals. A page can rank well and be absent from the AI answer, or the other way round.
  • Indexing now leans on clear definitions and structured data, alongside the traditional keyword and link signals.
  • Thin aggregator content no longer competes, because AI engines draw on substantial, original sources and skip the rest.
  • Measurement has to cover several AI engines. On 2 Nov 2026 we asked 15 buyer questions from a Malaysian location across five AI engines: 270 answers, and The answers leaned on Malaysian directories, local “best of” articles and firms with a clear Malaysia service page.
  • SEO and GEO are best done together. Neither replaces the other.

What we saw in November 2026

The clearest change we saw this month: ranking and being named are now separate results, and they vary by market, language and engine. Our own checks:

  • 2 Nov 2026, Malaysia. We asked 15 buyer questions from a Malaysian location across five AI engines and collected 270 answers. The answers leaned on Malaysian directories, local “best of” articles and firms with a clear Malaysia service page.
  • 4 Nov 2026, Vietnam and Thailand. We asked “best GEO agency” in English and in the local language from Vietnam and Thailand in Google AI Mode. In Vietnam, only one agency appeared in both the English and the local-language answer. In Thailand, none did. The same question in two languages produced two different shortlists.
  • 7 Nov 2026, Bing Copilot. Our Bing Copilot data for stridec.com showed our share when cited held at about 20 to 24%, while the number of questions citing us fell. A steady share can hide a shrinking footprint.
  • Aug 2026, Singapore. Qwang Coffee, Alva’s own store, was cited in a Google AI Overview 48 hours after launch and listed first on 3 of 5 commercial questions. Two engines, two different results for the same store.

None of this shows up in a rank tracker. That is the practical case for the measurement changes later in this article.

How has indexing changed?

Indexing has been the quietest change but the most basic one. Search engines, and the AI crawlers that feed answer engines, now organise what they know around businesses, products, people and concepts and their attributes, not only around words and the documents that contain them.

Clear records, not bags of words. A page about a product, person, place or concept is read as a structured record: what the thing is, its type, its attributes, how it relates to other things, and where the information came from. Schema markup (JSON-LD) feeds this directly. That record is what the AI draws on when it writes a response.

Structured data has gone from optional to expected. Articles, products, organisations, FAQ blocks, breadcrumbs and reviews ship with schema by default. Pages without schema can still be indexed, but they go in as less structured records and lose out to pages that gave the engine an explicit, machine-readable definition.

Images, video and audio count. They are increasingly read for their content rather than treated as files with alt-text labels. Transcripts, diagrams and screenshots all add to what the index knows about a page.

What this means for your work. On-page work is no longer just title, meta description, H1 and keywords. It includes complete schema, a clear statement of what the page and the business are about, and a clean site structure so the engine doesn’t have to guess.

Does ranking still matter?

Yes. Traditional ranking, the position your page holds in the blue-link results, has not gone away. It is still the largest single source of organic traffic for most sites, and AI Overviews mostly draw on pages that already rank. What has changed is that ranking is no longer the only goal. Being named in AI-generated answers is a parallel goal that works differently.

The two can diverge. A page ranked at position 3 may be absent from the AI Overview for the same query. A page ranked at position 12 may be its main citation. In August 2026, AeroChat ranked third on Google for “best Shopify chatbot” but wasn’t named in the AI Overview. The reason turned out to be the other sources: one article named AeroChat first, but none of the others mentioned it, so the AI didn’t either. Ranking and being named are separate. Find out which one is broken before you fix anything.

Quotable writing is the new on-page property. AI engines build answers by lifting and combining statements from sources. Pages that state claims clearly, one claim per sentence, attributable and in context, are quoted more reliably than pages that bury claims in marketing prose.

Credibility is judged per question. The AI asks whether the source is credible for this specific question, not in general. A government website is credible for statistics, a practitioner blog for methods, a news outlet for events. The same page may be cited for some queries and ignored for others.

What this means for your work. SEO work now has two tracks. The ranking track continues with on-page optimisation, technical health and link building. The AI answer track adds clear definitions of what you are, clear claims, mentions on other sites, and checking each AI engine. We run both together, which is why we describe Stridec as a GEO/SEO agency. More on how the two compare in generative engine optimization vs SEO.

Which AI engines matter now?

The single results page has split into several answer engines, each with its own citation habits and traffic patterns. Working SEO now means knowing what each one does and where it gets its sources.

Google AI Overviews and AI Mode. The AI-generated summary at the top of Google results for many queries, and Google’s conversational AI Mode. Sources usually come from the top organic results, but not always: Google’s model has its own selection logic and sometimes promotes lower-ranked pages that are easier to quote. See how to get cited in AI Overviews.

ChatGPT (with search). When ChatGPT searches the web, it reads the top results and writes an answer with inline citations. It leans on authoritative domains and well-structured content. A growing share of buyers use ChatGPT as their research starting point instead of Google. If you are missing there, start with why isn’t my brand showing in ChatGPT.

Gemini. Google’s own assistant, which grounds its answers in Google’s index. Its sourcing tracks Google search closely, with some differences covered in how to get cited in Gemini.

Claude (with web search). Web-connected Claude searches and cites sources. It favours thorough, well-structured content with clear claims and attribution. See how to get cited in Claude.

Bing Copilot. Microsoft’s answer engine across Bing search and Copilot. It favours authoritative sources but also cites practitioner content for niche queries.

What this means for your work. Tracking has to cover several engines. A brand can be named in ChatGPT and absent from Google AI Overviews, as Qwang Coffee was in Aug 2026. Checking one engine misses most of the picture.

What kind of content works now?

Depth, originality and clarity now matter more than volume. Thin, aggregator-style content has lost its place in AI answers, because the engines prefer original, substantial sources and skip the aggregators that used to win rankings with thin pages on high-volume keywords.

Aggregators lose. Pages that summarised what other sources said, without adding original observation or data, could rank when ranking was the only game. They lose in AI answers, because the AI would rather cite the original source than the summary.

Original observation gets cited. Pages with practitioner observation (what we saw when we did this), original data (numbers from work the author actually did), or a distinct way of explaining the topic are cited more reliably than pages without them.

Depth gets cited. Pages that cover the topic properly, with real substance rather than padded length, are cited more than thin pages that ranked on freshness or keyword density.

Clarity gets cited. Pages that define their subject clearly (a plain opening definition, a clear breakdown of sub-topics, schema that removes ambiguity) are cited more than pages where the engine has to guess what the page is about.

The penalty for thin content has moved. Thin content can still rank for low-competition keywords. It rarely gets cited.

What this means for your work. Content budgets shift toward fewer, deeper articles with original observation and complete coverage of the topic, rather than more thin pages.

How should you measure SEO now?

Rank tracking is no longer enough. Measurement has expanded from rank on a single results page to whether you are named across several AI engines, AI Overview presence, and the effect of zero-click answers. Teams that haven’t expanded their measurement are missing the half of the picture the AI engines now control.

Named rate per engine. For each priority question, check whether each major engine (Google AI Overviews and AI Mode, ChatGPT, Gemini, Claude, Bing Copilot) names your brand in the answer, and separately whether it cites your page. Track this over time, the way you track rank. Named is the headline number. Cited is an early signal.

AI Overview presence. For each priority query on Google, check whether an AI Overview appears, whether you are in it, and how much of the answer comes from your page. Track this alongside organic rank. The free AIO Visibility Checker gives a quick first read.

Zero-click impact. AI answers sometimes settle the user’s question with no click to the source. The traffic effect varies by query: informational queries lose many clicks, while commercial and transactional queries still drive them. Audit your query mix to see your exposure. If your traffic is already falling, see losing organic traffic to AI Overview.

Branded search lift. Being named in AI answers often lifts branded search: people who saw the brand in an answer later search for it directly. AeroChat’s branded search share reached 74% period over period in early 2026, alongside its first AI Overview. Track branded search as a downstream signal of AI visibility.

What this means for your work. The report needs rebuilding. Rank is one column; named rate per engine is another; AI Overview presence is another; branded search trend is another. A rank-only report undersells the work happening in the AI answers. For a full walk-through, see how GEO results are measured.

Conclusion

AI changes SEO in practice, not in theory. Indexing now favours clear definitions and structured data. Ranking sits next to being named in the answer as a parallel goal with different signals. Search has spread beyond Google’s results page to AI Overviews and AI Mode, ChatGPT, Gemini, Claude and Bing Copilot. Content economics favour depth, original observation and clarity. Measurement has grown from rank tracking to named rate per engine, AI Overview presence and zero-click analysis.

The crawl, index and rank pipeline still exists. What has been added is a layer that reads the indexed content, judges whether it is clear and credible enough to quote, and decides which brands to name. Working SEO in November 2026 means doing both tracks: the ranking work that has always existed, and the AI answer work that is now its own discipline. Sites that hold both win. Sites that hold one and ignore the other lose share wherever they stopped looking.

What to do next

If you suspect AI search is already costing you, check before you change anything. Run your brand through the free AIO Visibility Checker, then ask ChatGPT and Gemini the questions your buyers ask. If you rank but enquiries have dropped, read we rank on page one but enquiries dropped. If your boss has asked whether you show up, how to check in 15 minutes and what to report back gives you a plan. When you are choosing how to fix it, hire a GEO agency, train your team, or do it yourself lays out each route.

Frequently asked questions

Is SEO dead because of AI?
No. Organic search remains a major source of B2B and consumer research traffic in 2026, and AI answers mostly draw on pages that already rank. What has changed is that ranking is now one of two parallel goals, with being named in AI-generated answers as the second. The work has expanded, not collapsed. Sites that ignore AI answers lose share there; sites that ignore ranking lose share in the results page. Both still matter.
How do I get my page cited in AI Overviews?
Three things matter most: clarity (the page defines its topic and your business plainly, with structured data), quotable writing (claims stated clearly, one per sentence, attributable), and credibility for the specific claim. Pages that cover the topic properly, use schema markup, and contain original observation or data are cited more reliably than thin or aggregator-style content. Our step-by-step guide goes deeper.
Do backlinks still matter for SEO?
Yes for ranking, and less directly for AI answers. Backlinks remain a primary ranking signal. For AI answers they are an indirect signal: they help establish credibility, but the AI also reads the page itself and weighs how clear and quotable it is. Mentions of your brand on other sites matter too, even without a link. A heavily linked page with unclear writing can still lose to a clearer page with fewer links.
What is the difference between ranking and being named in an AI answer?
Ranking is your position in the blue-link results, driven by traditional SEO signals such as relevance, links and engagement. Being named means your brand appears inside the AI-generated answer, in Google AI Overviews, ChatGPT, Gemini or Claude. Being cited means your page is linked as a source. The three can diverge: in Aug 2026 AeroChat ranked third on Google for “best Shopify chatbot” but wasn’t named in the AI Overview, because the other sources behind the answer didn’t mention it.
Should I optimise for ChatGPT and Gemini, not just Google?
Yes, if your buyers use them, and most now do. Each AI engine has its own citation habits. ChatGPT (with search) leans on authoritative domains. Gemini tracks Google’s index closely. Google AI Overviews draw on organic results but apply their own selection. The work is largely shared (clear definitions, quotable writing, schema, depth, mentions on other sites); the difference is in checking each engine separately.
Has keyword research changed because of AI?
Yes. Single-keyword research now extends to fan-out research: the cluster of related, follow-up and neighbouring questions an AI might consider when answering a topic. A page that covers one query well will rank for that query, but a page that covers the whole cluster gets cited across the family of questions. AI prompts are also more personal and varied than keywords, so write for the buyer’s situation, not just the search term.
Does schema markup matter more now?
Yes. Schema (JSON-LD) is how a page tells the engine what it is and what it covers. Pages without schema go into the index as less structured records and lose out to pages that supplied an explicit, machine-readable definition. Article, FAQPage, BreadcrumbList, Organization and Product schemas are now basic requirements for content that wants to be cited.
How do I check whether AI search is affecting my business?
Start with the free AIO Visibility Checker to see whether Google’s AI Overview names you. Then ask ChatGPT and Gemini the five to ten questions your buyers actually ask, from the market you sell in, and note who gets named. If you rank but aren’t named, that is the gap. This 15-minute check walks through it.
Should I hire a GEO agency or adapt my SEO team?
Either can work. A capable SEO team can learn the on-page side; the off-site side, getting other sites to describe you consistently, takes time and contacts many teams don’t have. Hire a GEO agency, train your team, or do it yourself compares the routes, and our GEO services and pricing shows what the agency route costs with Stridec.

The AI Search Playbook

The full method to get your brand named in AI answers, to run yourself.

Get the Playbook →
AL
Author

Alva Chew

Founder of Stridec, in search since 2001. Stridec is a Singapore-based GEO/SEO agency that helps brands get recommended by AI searches such as Google, ChatGPT and Claude.