Google Starts Strengthening Website Links in AI Answers: What Does This Mean for GEO?

Yesterday, I saw Google's October 15 AI search update and noticed a very interesting change.

What Google emphasized this time isn't just how many questions AI can answer, but whether users can more easily find the websites behind those answers.

For the past two years, we've been discussing whether AI search would reduce website clicks. Now, Google is making links in AI answers more prominent.

This reminds me of twenty years ago, when we were still figuring out how to get a website to appear in search results.

Today, the question has changed again.

According to Google's official announcement published on October 15, "More ways to connect with the web from Search," Google is adding more inline links in Google AI Mode and AI Overviews, and adjusting the placement of some links to make it easier for users to visit source webpages while reading AI answers.

At the same time, Google also announced testing a dedicated display area for official websites, gradually expanding the regional and language coverage of news article carousels, and testing clickable creator and publisher profiles.

It's important to distinguish here: the display adjustments for inline links are being rolled out, while the official website dedicated area and creator profiles are still in testing, and the news carousel is in an expanded rollout phase. These features are not being fully released to all users at the same time.

Behind these changes is a question worth businesses rethinking.

If AI can already answer users' questions directly, why is Google still emphasizing bringing users back to websites?

My understanding is that AI search hasn't made websites lose their value—it's changing why users need to visit websites, and what kinds of websites are worth continuing to visit.

From Yahoo to Google to AI search, what has changed about the role of websites?

I started doing SEO at a digital marketing company in Hong Kong in 2005.

At that time, Yahoo was still a search platform we frequently encountered in our daily work. We studied directories, webpage content, and keywords, hoping users could find corporate websites through the search entry point.

Back then, websites bore a very complete information explanation task.

When users wanted to learn about a company, they needed to enter the website, look at the company introduction, product catalog, contact information, and sometimes even download a PDF to find specific specifications.

Search platforms helped users find the entry point, but the actual information reading and business judgment mainly happened within the website.

Later, Google gradually became an important platform for overseas search marketing.

We began to focus more on keyword rankings, organic search traffic, page content, and external links. Businesses wanted their webpages to appear in search results for relevant queries, and then bring users to their websites through clicks.

At this stage, the commercial value of SEO was relatively easy to explain: search demand brought impressions, impressions could bring clicks, and clicks further generated inquiries and sales opportunities.

Of course, the actual process was never that simple, but businesses could at least observe results along this path.

In the AI search era, the order in which users obtain information has changed.

In Google AI Mode, AI Overviews, and AI search experiences like ChatGPT and Perplexity, users can first see system-organized answers, then decide whether to open the source webpages.

For example, a European procurement manager wants to understand the selection criteria for industrial equipment suppliers.

In the past, he might have needed to open multiple search results and read supplier websites and industry articles separately.

Now, AI can first organize technical requirements, certification standards, procurement considerations, and supplier evaluation dimensions.

Users may have already formed initial judgments before visiting any website.

But when they need to verify a specific model's technical parameters, download certification documents, or contact suppliers for quotes, websites still bear tasks that AI answers cannot fully replace.

Looking back at the three stages of Yahoo, Google, and AI search, I find that what has truly changed isn't whether websites still need to exist, but at which stage of the decision-making process users need to enter websites.

The search entry point has changed, and the information responsibilities that websites bear are also changing.

AI can already provide answers, so why does it still need to highlight information sources?

What caught my attention in Google's update is the relationship between AI answers and original webpages.

Previously, the links on search results pages were themselves the main content. Users judged which pages were worth visiting based on titles and snippets.

Now, AI answers present organized information first, and links become the entry point for users to further explore, verify, and execute tasks.

I believe there are at least three noteworthy changes here.

Users need answers, and they also need to verify answers

AI can help users quickly understand a question, but understanding doesn't equal completing a decision.

For example, if a user asks what application scenarios an energy storage product is suitable for, AI can introduce basic uses and common configurations.

But if the user needs to confirm whether the product meets local certification requirements, whether the current model is still on sale, or whether a certain parameter applies to a specific project, they need to check reliable materials further.

Official websites, technical documents, and other verifiable sources still have clear roles here.

The value of source links isn't just letting users know where answers come from—it's also helping users continue the work beyond AI answers.

Different information needs require different types of websites

Google's announcement of testing a dedicated display area for official websites is especially worth businesses' attention.

Official webpages are suitable for tasks with clear source requirements, such as viewing formal product information, accessing designated service entry points, or completing certain operations.

News article carousels are more suitable for users to learn about recent events and reports from different media outlets.

Creator and publisher profiles can help users further understand the individuals or organizations behind the content.

These designs don't mean Google has established a fixed website value hierarchy—rather, they reflect that different search tasks have different requirements for information sources.

I believe businesses need to understand this difference, rather than treating all content as material for earning AI citations.

More prominent sources don't mean sources automatically become credible

There's another easily overlooked distinction.

A link appearing in an AI answer means users have the opportunity to see this source, but you can't judge whether the content is accurate, complete, or reliable based solely on display position.

True credibility still needs to be supported by content facts, publishing entities, evidence, and update status.

Google's related updates in May and October 2026 continue to focus on source display and webpage exploration experience. I prefer to understand it as a product direction: AI can help users organize information, while webpages continue to carry deeper information and tasks.

AI answers and website visits aren't necessarily an either-or relationship.

Does Google strengthening source links mean website traffic will recover?

This is a question many marketing leaders naturally ask after seeing this news.

A typical discussion scenario is a marketing leader looking at a complete AI-generated answer and asking: "Users already know the answer, so why would they still click on the official website?"

I usually first acknowledge that this concern is valid.

Some questions indeed no longer require clicks.

If users just want to know the meaning of a term, the result of a simple calculation, or a common fact, the search task may end after AI answers directly.

But if users need to compare different suppliers, verify product specifications, understand service details, or further contact businesses, visiting websites still has practical significance.

So I don't think Google adding source links means traditional organic search traffic will return to past levels.

At least for now, we can't draw such a conclusion from a single interface update.

In GEO performance evaluation, I've always advocated distinguishing three stages.

First, the brand is mentioned by AI. Users see the company name or product name in answers.

Second, the official website is displayed as a source. Users can discover corporate webpages through links in AI answers.

Third, users actually visit the website and generate subsequent behaviors such as reading, downloading, or inquiries.

These three stages are related but not the same thing.

A brand may frequently appear in AI answers but rarely earn official website citations. A webpage may also frequently be displayed as a source, but users may not necessarily click.

Even if clicks happen, you still need to determine whether visitors found the information they needed.

This means the value of AI search traffic can't be measured by visit count alone.

For overseas B2B companies, a visit from a clear procurement need and a visit from a general knowledge query may be completely different in commercial significance.

Of course, this doesn't mean traffic from AI is naturally more valuable. Whether it has higher commercial intent needs to be judged with real data.

Google's update this time brings more observation opportunities, not already-validated traffic growth results.

In the AI search era, what roles do corporate official websites, third-party sources, and brand content each play?

If the value of websites is changing, businesses need to reconsider where content resources should be placed.

In the past, when many companies developed SEO content plans, they started from keyword volume and tried to cover related search demands as much as possible.

At the GEO stage, some companies have shifted focus to third-party media publishing, hoping to increase the chances of brand appearance in AI answers by adding external content.

I'm not against content coverage, nor against third-party publishing.

But if all content repeats the same product introduction, users may not gain more valuable information even if they see multiple sources.

I prefer to divide a company's information assets into three roles.

These three types of content are not substitutes for each other.

The official website is responsible for clearly stating facts the company can confirm.

Third-party sources provide different perspectives, giving users the opportunity to cross-verify information rather than simply repeating corporate promotional materials.

Brand content helps potential customers understand why a certain product or service suits a specific scenario, and what differences exist between choices.

For example, an industrial equipment company targeting European and American markets should clearly explain equipment specifications, certifications, and service scope on its official website.

Industry media can discuss related product categories from the perspective of technology trends or market demand.

Selection guides published by the company itself can explain what parameters procurement staff should focus on under different application conditions.

When this information remains consistent at the factual level while each providing independent value, users have more complete research materials.

This is also my understanding of GEO content strategy.

It's not just about getting AI to mention the brand a few more times—it's about helping companies build information assets that can be discovered, understood, verified, and further visited.

In the AIPO methodology proposed by me and my team, content planning, platform publishing, and performance monitoring are also interconnected work stages. Content first solves the information value problem, then considers how to reach users through suitable channels and how to observe subsequent performance.

However, a reasonable division of roles among sources doesn't mean AI systems will consistently cite a certain type of website. Different platforms, questions, and search contexts can all affect actual display results.

Facing AI search changes, what metrics should businesses refocus on?

If you only look at total organic search clicks, businesses may find it hard to fully understand the changes brought by AI search.

But I also don't agree with giving up measurement just because AI search attribution is difficult.

For companies already doing overseas SEO and GEO, I suggest focusing on the following questions first.

• In AI queries related to the brand and core products, is the company mentioned? Is the official website displayed as a source?
• Which pages do the links displayed by AI point to? Is it the homepage, or product pages, solution pages, and technical document pages that help users complete tasks?
• Have impressions, clicks, and search query performance for related pages in Google Search Console changed?
• Do users entering the website continue reading, download materials, submit forms, or generate valid inquiries?
• Under the same region, language, query conditions, and testing period, does AI source display have some consistency?

There's also a practical issue to note here: Google Search Console doesn't provide complete independent reports for all AI answer citations and brand mentions, and website analytics tools may not accurately identify every visit from AI search.

Therefore, traditional search data, website behavior data, and reproducible AI query observations need to be analyzed together.

The same question may have different AI answers across different regions, devices, account states, or times. A single test showing or not showing a certain brand isn't enough to indicate long-term performance.

I focus more on trends than on a single screenshot.

At the same time, AI visibility and actual business contribution should be kept separate. Increased brand mentions is an observation result, while improved inquiry quality needs support from other evidence.

After AI search changes the entry point, where is the long-term value of corporate websites?

Looking back at the three search stages I've experienced, one thing has never changed.

Users need to find information that helps them solve problems.

What has changed is how information is organized and how users decide their next action.

In the Yahoo era, users found websites through entry points.

In the Google era, users selected webpages through search results.

In the AI search era, users may understand answers first, then decide whether to visit sources as needed.

For businesses, this means websites can't be built solely around getting clicks.

They also need to bear the tasks of fact verification, professional explanation, product evaluation, and business communication.

If you're a CMO responsible for overseas market budgets, I suggest re-examining the investment relationship between official website content, SEO, and GEO. Don't evaluate the official website only by total organic traffic—also focus on its role in brand research and customer decision-making.

If you're a marketing director, I suggest first mapping out what information the official website, industry content, and third-party sources each provide, reducing repetitive publishing without independent value, and placing resources on questions customers truly need to understand and verify.

If you're an SEO or GEO lead, I suggest continuing to do well in webpage discoverability, technical standards, and content maintenance, while observing brand mentions in AI answers, official website source display, and actual user behavior after visits.

SEO hasn't lost its meaning because of GEO. Webpages being discoverable, accessible, and understandable by search systems remains an important foundation. GEO further makes us focus on how content enters AI-generated answers and how users move from answers to sources.

Regarding Google's October 15 update, I think it's worth continued observation, but now is not the time to announce a website traffic recovery.

What's truly worth businesses investing in isn't chasing a certain link display format, but building websites with clear facts, independent content value, and the ability to help customers continue decision-making.

AI search hasn't made websites lose their value—it's gradually turning websites from places that provide answers into important destinations where users verify answers, deepen understanding, and complete decisions.

Related Questions

What's the difference between Google AI Mode and AI Overviews?

AI Overviews typically provide AI-generated summaries in Google search results to help users quickly understand questions. Google AI Mode provides a more complete conversational search experience, supporting continuous follow-up questions and in-depth exploration. Both may display related website links, but the interaction methods differ.

Will Google adding links in AI answers increase website traffic?

Clearer source links may increase opportunities for users to discover and visit websites, but it doesn't mean all websites will see click growth. The actual impact also depends on query intent, link position, and whether users need to further read the original webpage.

What's the difference between GEO and SEO?

SEO focuses on webpage discoverability, relevance, and organic search performance in search systems. GEO further focuses on how brand information appears in AI-generated answers, citation sources, and related user behavior. The two work in coordination—there's no need to choose one over the other.

How can corporate official websites increase their chances of being cited by Google AI search?

Companies can continuously maintain accurate, clear, and accessible product and service information, provide content and evidence with independent value, and follow Google search-related technical standards. These efforts help build an information foundation that search systems can understand, but they can't guarantee AI citations.

Does doing GEO require relying on third-party media publishing?

No. Corporate official websites, industry media, professional communities, and independent content each have their roles, and channels should be chosen based on target users' research habits. The value of third-party content lies in providing meaningful external information, not just increasing brand appearance frequency.

How should businesses measure the value brought by AI search?

You can observe by combining brand mentions in AI answers, official website source display, website visits, and valid behaviors after visits. When evaluating, distinguish between visibility changes and actual business contribution, and observe trends through continuous, reproducible queries and website data.