Google's AI search results changed in November 2026 in ways that are genuinely good news for website owners — with one catch that takes some control away from you. AI Mode and AI Overviews now show more clickable links, placed higher in the answer and more often. The citation cards that list your page as a source now sometimes carry a short summary written by Google's AI, not text taken from your page. A new button lets readers add your site to their preferred sources without leaving the results page. And Bing's webmaster tools now show which topics your site is being cited for in AI answers. None of this is a new ranking system. It is a set of display changes that decide how much traffic AI answers send on to websites — and they reward the same things good SEO has always rewarded: clear answers, specific content, and a site people actually remember. This article walks through each November change, what it means for your site in practice, and what to do about it this week. The changes below are drawn from the November 2026 SEO and AI search news roundup, which tracks each update to its original source.
| November 2026 change | Status | What it means for your site |
|---|---|---|
| More clickable links in AI Mode and AI Overviews | Rolling out | Direct answers earn inline links placed next to the claim they support |
| "Quick results from the web" official-site section | Testing | Brand and task queries can surface your official page at the top |
| AI-written summaries inside citation cards | Testing | Google describes your page in its own words; you do not control the text |
| "Source preferences" button in AI Mode citations | Testing | Readers can add your site as a preferred source without leaving the page |
| Related article links with thumbnails | Rolling out | Featured image quality becomes a click factor inside AI answers |
| Bing Topics Portfolio in Webmaster Tools | Beta | See which topics your site is cited for, and which opportunity queries to target |
| AI Mode ads in 10 new countries | Rolling out | Sponsored results arrive in AI Mode; UK, Brazil and Mexico follow later |
More clickable links, higher up in AI Mode and AI Overviews
The headline change is the simplest one: Google is putting more links into its AI answers, higher up, more frequently. For months, the main complaint from site owners about AI Overviews and AI Mode was that the answer swallowed the click — the user got what they needed without ever visiting a source. Google's November updates push in the opposite direction.
Inline links now appear next to the specific claim they support, rather than being gathered in a single list at the bottom. That matters because a link placed next to the sentence it supports is far more likely to be clicked than one buried after the answer ends — and it means content that answers a specific question directly is more likely to earn one. If your page answers the query in its first few paragraphs, in plain language, the inline link has somewhere to point.
Google is also testing a "Quick results from the web" section that prominently surfaces the one or two official websites that matter for a specific task or query, in a dedicated section at the top of the response. For a local business or a brand search — someone searching for a specific gym, restaurant or company by name — Google may now surface that business's official page directly, rather than leaving the user inside the AI answer. The winners here are sites with a clear brand presence: if the query contains your name, your official page is the obvious destination.
Two more display changes extend the same idea. News article carousels, which highlight timely articles, are rolling out to more locations and languages, and they now highlight Preferred Sources — content from websites the reader already follows. And at the end of some AI responses, Google now shows related article links with image thumbnails in a more clickable design, which makes a blog post's featured image a click factor in AI results, not just in traditional search. Google is also testing clickable creator and publisher names on video results in AI Mode, linking to the creator's Google Search Profile — another reason to claim that profile if you publish video.
The pattern across all of these is consistent: Google is investing in making AI search useful for finding and visiting websites, not just for reading answers. The sites that benefit most will be the ones with a clear brand, consistent publishing, and actual content Google can point to.
Google now writes your citation summary for you
The second change is the one that takes control away. Google has started testing AI-generated summary descriptions inside AI Overview citation cards — the source cards that appear alongside an AI Overview response. Instead of showing text pulled directly from your page, the card sometimes shows a short description Google's AI wrote about your page, based on what it thinks is most relevant to the query. SEO analyst Brodie Clark, who spotted the change, noted that the text does not appear exactly like that on the page: it is summarised based on the information Google thinks is most relevant.
You do not control this text. There is no setting to edit it and no tag to influence it. It is the same dynamic that already applies to meta descriptions in traditional search — Google often rewrites those too — now extended into AI citation cards. What you can control is the material Google's AI draws on. A page that explains clearly, in its first few paragraphs, what it is about and who it is for gives the summariser good material. A vague, generic page gets summarised vaguely. Clear headings and well-structured content help for the same reason: if a human can scan your page and understand it in seconds, Google's AI can summarise it accurately.
This is also why the November updates connect to Google's AI content guidelines, which Google tightened in October to say that manual fact-checking of AI-generated content is critical. The same standard now cuts both ways: Google's AI summarises your page, and your page needs to be specific and accurate enough to survive being summarised. Thin content does not just rank badly — it now gets described badly, in Google's own words, on the card that is supposed to send you the click.
The Source preferences button moves into AI Mode
The third change is small on the surface and potentially the most valuable of the three. Google is testing a "Source preferences" button directly inside the AI Mode citations sidebar — the panel that shows which sources were used to generate an AI response. Spotted by Damien Andell, who shared a video of it, the button opens your Preferred Sources list right there in the panel, without leaving the results page. Users can add or remove sources on the spot.
Preferred Sources is Google's personalisation feature that lets a signed-in user mark the websites they want to see more of. A site a user has added gets more prominent placement in their results — including in article carousels and AI Mode responses — and its content can carry a "preferred" badge. Google's help page explains how readers manage their source preferences from their profile settings. Google's own documentation is precise about the scope: the badge appears for users who have selected your site, and the feature works alongside ranking rather than overriding it. It does not improve your rankings for everyone, and it cannot rescue a site that trips Google's spam or quality filters.
Two eligibility details matter. Only domain-level and subdomain-level sites can appear as a preferred source — a subdirectory such as example.com/blog cannot — and to appear inside the AI features specifically, your site must be included in Search generative AI features in Search Console, which you can check under Settings, then Search generative AI. Google's Guide to Preferred Sources covers the full requirements.
Until now, the feature lived mostly in settings pages most users never open. Putting the button inside the citations sidebar changes that: a reader who just found your content genuinely helpful — an answer that solved their problem — can add your site at the exact moment they are looking at your citation. That makes the button a loyalty tool rather than a ranking tool. The sites that gain from it are the ones whose content is good enough that a reader wants more of it, which is a standard no shortcut satisfies.
You can also put Google's "Add to preferred sources" button on your own pages — Google provides the script and a deeplink fallback — so a reader can add your site without going through Google's settings at all. Check first that your domain appears in Google's source preferences tool; if it does not show up there, the button will not work yet.
What AI SEO tools can and cannot tell you now
The November updates also brought a new measurement option, this time from Bing. Bing Webmaster Tools launched Topics Portfolio, currently in beta, inside its AI Performance report. Instead of showing individual queries one at a time, it groups the queries where your site is being cited in Bing's AI answers into broader thematic clusters, so you can see the shape of your presence across topics at a glance. Each topic shows its size (how much AI search demand the topic represents), how many other sites are cited for it, and how many distinct queries sit inside it. Clicking into a topic shows the grounding queries — the specific phrases where your site is already cited — and the opportunity queries, where Bing thinks your site could be cited but is not yet.
That distinction is genuinely useful for content planning: the grounding queries show where you already have a real AI search presence, and the opportunity queries are specific questions your existing and ideal clients are asking where you have the expertise to answer but are not currently being used as a source. The feature is beta and the topic classification is AI-generated, so an occasional query will land in a category that does not quite fit — Bing provides thumbs up and down feedback to improve it — but as a way of seeing your AI citations thematically rather than one query at a time, it is the clearest view either search engine currently offers.
What no AI SEO tool can tell you yet is the number that matters most: how many clicks AI answers are sending you. Google's Search Console has a generative AI report, and Bing has its AI Performance report, but neither is a complete click record. Treat any tool that claims a precise "AI visibility score" with suspicion. The honest measurement stack right now is Search Console for your own clicks and impressions, Bing Webmaster Tools for your AI citation topics, and manual checks of the queries you care about. If you are evaluating AI search optimization tools, ask them which of those three sources their numbers come from — a tool that cannot answer that is selling a guess.
What to do this week
The November changes reward work you probably already know you should be doing. In order of impact:
1. Answer the question in the first few paragraphs. Inline links are placed next to the text they support, and citation summaries are drawn from what your page says early on. A page that opens with a direct, specific answer earns both. A page that opens with three paragraphs of throat-clearing earns neither. If you have not already, check your most important pages: does the first screenful answer the query, or does it introduce the topic?
2. Make every page's purpose obvious. Clear headings, a logical structure, and specific language are what let Google's AI summarise your page accurately in a citation card. Vague content gets a vague summary, and a vague summary does not earn clicks. This is the same discipline that makes content quotable for AI engines in general — and it is worth applying to your guides and tool pages, not just your blog posts.
3. Use strong, relevant featured images. Related article links in AI responses now carry thumbnails, so visual quality is a click factor inside AI answers. A featured image that genuinely represents the article's subject earns the click; a generic graphic does not. The same applies to adding FAQ schema to your pages, which makes your questions and answers eligible for richer display in both traditional and AI results.
4. Add the Preferred Sources button to your site. It costs two lines of script, and with the new in-citations button, readers are more likely than ever to discover the feature. Set your own site as a preferred source first, through your Google account settings on mobile, to see the flow from the reader's side. Then add the button where your most helpful content lives.
5. Check your Bing Topics Portfolio. Log into Bing Webmaster Tools, open the AI Performance report, and look for the Topics Portfolio tab. Start with the topics where you already have citations, expand them, and look at the opportunity queries — those are the most direct content briefs AI search has ever handed you. If the tab is not there yet, the feature may not have enough data for your site; check back in a few weeks.
6. Confirm your site is included in Search generative AI features. In Search Console, go to Settings, then Search generative AI, and confirm your property includes, or inherits, the include setting. Without it, your pages are not eligible to appear as supporting links in AI Overviews or AI Mode at all — and they cannot appear as a preferred source inside the AI features either. This is the one check on the list that is a hard requirement rather than an optimisation.
The bigger picture: more searches, slower monetisation
It is worth understanding why Google is making these changes now. Alphabet's third-quarter results showed search queries growing 12% year over year — the fastest rate in five years — driven by people using AI Mode more than ever. But search revenue missed estimates, and the stock fell roughly 10% on the news. People are searching more, but the searches are migrating toward a format that monetises less well than the one Google spent twenty years optimising, because AI answers synthesise information directly and users click on traditional ads less often. Google's chief executive compared the moment to the shift from desktop to mobile a decade ago, when mobile ads initially made less money until Google worked out how to monetise them properly.
That context explains the shape of the November updates. More clickable links in AI answers, more prominent official-site sections, and richer article carousels are all ways of keeping websites in the loop while Google works out the monetisation — and the rollout of sponsored ads inside AI Mode to ten new countries, including India, Canada, Australia and Singapore, with the UK, Brazil and Mexico to follow, is the other half of the same strategy. For site owners, the message is that Google needs websites in its AI answers: an answer with nowhere to go is a dead end, and dead ends do not keep users searching. The November changes are Google acting on that need. The sites that benefit will be the ones ready to be linked to.
None of this replaces the fundamentals. Getting cited in Google AI Overviews still starts with being indexed, being eligible for a snippet, and answering the question better than the page next to you. What November changed is what happens after you are cited: more links, higher up, a summary you do not write, and a button that lets a satisfied reader choose to see more of you. That is a better deal than site owners had in October. It is worth the hour it takes to act on it.
Frequently asked questions
What changed in Google AI Mode in November 2026?
Google added more clickable links higher in AI Mode and AI Overview answers, a "Quick results from the web" section surfacing official sites, wider news article carousels, related article links with thumbnails, AI-written summaries inside citation cards, a "Source preferences" button test inside the AI Mode citations sidebar, and AI Mode ads in ten new countries. Bing also launched Topics Portfolio in beta to show which topics your site is cited for in AI answers.
Do AI Overviews send traffic to websites now?
More than they did. The November updates put more inline links higher in responses, surface official sites in a dedicated section, and show related articles with thumbnails — all designed to send clicks through. But the AI answer still satisfies many queries on its own, so the sites that gain clicks are the ones whose pages answer specific questions directly and are cited next to the claim they support.
Can I control the summary Google writes about my page in a citation card?
No. Google's AI writes the summary based on what it thinks is most relevant about your page for that query, and there is no setting to edit it. What you can control is the material it draws on: a page that states clearly, in its first few paragraphs, what it is about and who it is for will be summarised more accurately than a vague one.
What are Preferred Sources in Google Search?
A personalisation setting that lets a signed-in user mark the websites they want to see more of. A selected site gets more prominent placement in that user's results and can carry a "preferred" badge in AI Mode, AI Overviews and Top Stories. It is user-scoped: it helps with the people who added you, not with everyone. Only domain-level and subdomain-level sites are eligible, and the site must be included in Search generative AI features in Search Console.
How do I add the Preferred Sources button to my website?
Google provides a standard script plus a container element, with a deeplink fallback if you cannot run JavaScript — the full instructions are in Google's Guide to Preferred Sources. First check that your domain appears in Google's source preferences tool; if it does not show up there, the button will not work yet. The button is optional: a site can be a preferred source without it.
What are the best AI SEO tools for tracking AI search visibility?
There is no single best tool yet, and any tool claiming a precise "AI visibility score" should be treated with suspicion. The honest stack is free: Google Search Console for your own clicks and impressions, including its generative AI report, and Bing Webmaster Tools for your AI citation topics through Topics Portfolio. Paid AI search optimization tools can be useful for tracking citations across queries, but ask them which of those sources their numbers come from before trusting them.
Comments
No comments yet. Be the first to share your thoughts below.