(17 min. read)
Google vs. AI
While Google still unarguably dominates traditional search, the rapid rise of ChatGPT and other chatbots shows that the way users seek answers is shifting before our very eyes. The arrival of AI-powered search features (e.g., Google AI Overviews, Bing Chat) is transforming how information is displayed—alongside classic results, we now see direct answers generated by language models. What does that look like in numbers?
- In June 2025, ChatGPT climbed into the top 5 most-visited websites globally, recording ~5.4 billion monthly visits (Search Engine Land).
- ChatGPT currently processes over 2.5 billion queries (prompts) each day; about 330 million come from users in the United States. On a yearly scale, that amounts to more than 912 billion chatbot interactions (The Verge).
- A rise in zero-click searches has been observed; in some U.S. states, the share of queries where Google displays an AI-generated answer has exceeded 30% (SE Ranking).
- According to Sam Altman, around 10% of the global population already uses OpenAI systems (Forbes).
- Despite AI’s growing role, Google remains the hegemon with a ~90% share (about 14 billion searches daily). For comparison, in June 2025 chatbots generated 5.6% of search traffic (Wall Street Journal).
These numbers send a very clear signal: a new, mass channel for acquiring information is emerging, one that SEO must take seriously. Visibility must now be fought for both in traditional SERPs and in generative model responses. And we can’t forget the growing power of other platforms—especially social media.
Still, optimization for AI is at an early stage—most companies don’t yet have a strategy for visibility in AI answers. In a survey presented at SMX Advanced, only 22% of marketers said they monitor brand visibility in LLMs (ChatGPT, Gemini, etc.) or the traffic they generate. Another 53% are “in early testing”; the rest are doing nothing for now (Search Engine Land). This gives early adopters a significant advantage that’s definitely worth leveraging. As Matt Diggity, a world-renowned SEO expert, put it:
AI isn’t killing SEO. It’s creating the biggest opportunity gap I’ve seen in my 16 years of doing this (Facebook).
SEO as Search Everywhere Optimization
In light of these changes, a new definition of SEO is emerging: instead of Search Engine Optimization, we now talk about Search Everywhere Optimization. In the AI era, it’s no longer enough to optimize for Google alone. SEO 2.0 means managing brand visibility and trust across multiple channels—including AI-generated answers, social platforms, media mentions, and user reviews. In short: everywhere where algorithms pick up quality signals and everywhere where users look for decision-making confirmation.
Today, relying solely on the uncertain flow of Google traffic is no longer viable—it’s valuable, even essential, to diversify traffic sources. Especially considering that 31% of consumers use social media to find answers online, and 1 in 4 consumers aged 18–54 prefer social search over traditional search engines (HubSpot). Moreover, 56% of Americans start product research on Amazon (Emarketer). Each platform has its own “decision code” and psychology, meaning a one-size-fits-all strategy simply won’t work.
Traditional SEO, focused exclusively on Google rankings, is therefore no longer sufficient, as purchase decisions increasingly happen elsewhere. The new strategy is optimization for every channel where decisions are made. The goal is for the brand to be present at the moment of choice, not just at the moment of search. This is the sentiment confirmed by Jorge Castro, who says:
AI isn’t replacing SEO — it’s transforming it. In this new era, success comes from blending classic SEO fundamentals with strategies that make brands visible in AI-generated answers, social platforms, and digital conversations. The brands that adapt fastest will own tomorrow’s visibility.
Jorge Castro, CEO of Growth Marketing, and founder of jorgecastro.ai
And Alongside SEO… GEO
GEO—Generative Engine Optimization—is another concept gaining traction. It’s the optimization of content and brand signals so that language models—ChatGPT, Gemini, Perplexity, or Google’s AI Overviews—can easily recognize, trust, and cite them in their answers. The primary goal of GEO is for a page to appear as a recommended source within generative AI. It’s no longer just about keyword stuffing, but about aligning with user intent and delivering precise answers to their queries.
What Does AI Cite?
The user journey online has shortened. Unfortunately, a click doesn’t always happen, as users are often satisfied with an instant answer alone—something SEO specialists have little control over. A growing percentage of queries now follow a “question-to-ready-answer” model, bypassing traditional browsing of results. Generative summaries in zero-click answers deliver information instantly, which shortens the decision funnel: the user reaches a website only once they’re nearly ready to make a decision. That’s why it’s crucial to optimize content for AI to pick it up as clear, concise, and accurate answers – this practice is known as AEO (Answer Engine Optimization).
Voice assistants are also gaining importance. Already, 30% of users aged 16–64 worldwide use voice assistants every week, and 45% of Americans use voice search on their smartphones (Backlinko). What does this mean? Users expect quick, specific, and simple answers to their questions.
How do LLMs choose which content to cite?
The more expert, concise, and well-structured the content is, the greater the chance an LLM will select it for recommendation. Google itself states that AI Overviews favors pages that meet high E-E-A-T standards and have a clear content structure, because it’s easier to extract key facts from them (Google). Equally important are websites with strong authority and a robust backlink profile.
AI-SERPs are also increasingly citing Shorts, videos, and YouTube playlists—such as product demos or tutorials. It’s worth embedding video into articles, publishing them on YouTube, adding timestamps with relevant keywords, and including rich video descriptions.
How to Optimize Content for LLMs?
AI Overviews and ChatGPT results are becoming a new SEO (or rather GEO) target. And while they don’t guarantee clicks or conversions, they often generate engagement with audiences who have the highest intent to act. To adapt content to the changing rules, it’s important to:
- Provide direct and concise content – AI prefers text that answers the question immediately, without long, low-value introductions. Language should be clear and NLP-friendly (Natural Language Processing).
- Focus on quality and relevance, not length – Google is increasingly effective at analyzing search intent and consistently rewards helpful, unique, relevant, and trustworthy content, aligned with E-E-A-T.
- Use a conversational tone – texts should be written in language close to natural human speech, since AI searches mirror how people actually talk, not rigid keyword strings. A Q&
A format works well. - Maintain heading structure – H1/
H2/ H3 headers haven’t lost value and still play a critical role in how AI systems understand content, so they should never be skipped. - Include TL;DR summaries – concise highlights of the article are frequently cited by AI models.
Instead of obsessing over ranking for specific keywords, it’s better to focus on deeply understanding what the user truly wants and why. Creating expert, data-driven, well-referenced content that genuinely helps people translates into stronger rankings, engagement, and trust. This are all the things Christopher Hofman Laursen deems to be necessary to be successful in this day and age:
There are the v. 1.0 manipulative hacks such as prompt injection, white text on white background, etc. They will work today, but is really not the foundation for a strategy, and will just create bad habits. Best practice SEO where you focus on covering the customer journey with a good semantic structure and focus on cost of retrieval is the way forward. Add an increased spend on digital PR to get digital footprints elsewhere than your site, then you have a LLM strategy. It's back to building brand. WHat is it that we want to be known for, and push that message with your digital PR strategy.
On another note with Chat GPT Agent it's important that your site is structured, sends the right message (a brand), provides the information (e.g. NAP) and has an easy conversion flow (also a good idea to remove ReCAPTCHA on your site, as the agent is stopped when booking/
buying a product).
Christopher Hofman Laursen, SEO strategist
How to Measure GEO Results
Measuring the impact of GEO is still developing, but the first tools are already helping track clicks from generative results:
- Semrush has introduced an AI Traffic Dashboard showing visits and conversions originating from Google AI Overviews, Copilot, ChatGPT, and Gemini.
- Ahrefs Brand Radar lets you track brand visibility in AI Overviews and citation frequency, reporting the number of highlighted keywords and new links appearing in LLM answers.
- Perplexity clicks can easily be captured in Google Analytics thanks to its automatic perplexity.ai referrer or with custom source filters.
- Chatbeat by Brand24 monitors brand visibility in LLMs, checks what AI says about the brand, how often, and flags misinformation.
- OpenAI increasingly tags outbound links with the utm_source=chatgpt parameter—making it possible to attribute chatbot visits in GA4 without manual segmentation.
What’s more? In June 2025, Google confirmed that clicks, impressions, and link positions in AI Overviews/
Traditional SEO Combined with AI
While optimization for LLMs appears to be the future of SEO, it’s important not to forget that universal site optimization rules remain the foundation of rankings. Keyword and user intent research, on-site SEO (including technical analysis and internal linking), and off-site SEO (with link-building at the forefront) are still indispensable. Not to mention crucial elements like SSL certification, strong Core Web Vitals, intuitive navigation, and responsiveness. A site with errors and weak authority has far lower chances of appearing in SERPs, and LLMs will also be reluctant to cite it if it doesn’t inspire trust.
What’s changing today is that we now have access to more and more advanced AI tools that speed up core SEO workflows. This saves specialists time on repetitive tasks and allows them to focus on strategy, creativity, and creating valuable content. However, while AI can handle massive analytical work and deliver recommendations, it’s the SEO expert who must decide how to use them. The most effective approach is therefore a combination of machine intelligence and human expertise. As Pontus Vippelius stresses:
If you're doing modern SEO, you're already prepared for a new era of SEO where answers are presented with generative AI. If you're stuck in SEO from 10 years ago, you have big steps to take.
Pontus Vippelius, Head of SEO at AdRelevance
The Importance of Backlinks Confirmed by Google
Just as SEO is not dying, link building is alive and well. In fact: very well. Google officially confirmed the significance of link building during the Search Central Live APAC 2025 conference (July 23–25). Cherry Prommawin, Search Quality Analyst at Google, stated clearly in her presentation:
Links are still an important part of the internet and used to discover new pages, and to determine site structure, and we use them for ranking (Search Engine Journal).
This official confirmation carries tremendous weight for the SEO industry, which has debated for years how relevant links remain for ranking in the age of AI and machine learning.
The strength of link building is also reflected in the data:
- Backlinko’s study of 12 million search results showed that pages ranking #1 have on average 3.8 times more backlinks than pages ranking #2–#10 (Backlinko).
- Internet Marketing Ninjas analyzed 1,113 pages ranking on the first page for 200 random keywords—85% of them had more than 1,000 links from unique domains (Editorial Link).
- In a survey by Aira, as many as 91% of experts stated that link acquisition is effective or highly effective in improving SERP rankings (Aira).
- Specialists continue to believe in the power of link building: in a survey of 500+ SEO experts worldwide, 73.2% said backlinks influence the likelihood of appearing in AI-generated search results, and 91.9% are convinced their competitors are buying backlinks (Editorial Link).
Link Building in 2025 – What to Focus On?
Google’s official confirmation of the major importance of link building as a ranking signal is indisputable. Links influence not only traditional rankings but also visibility in AI systems. Websites with strong, relevant backlinks are crawled, cached, and displayed more often in AI-generated content, which uses them to assess the trustworthiness of sources included in answers. The question remains: what should link building look like in the new era of artificial intelligence?
Accelerating Routine Tasks
AI accelerates many monotonous processes—and backlink profile building is no exception. Already, over 50% of link-building tasks such as research and site qualification are performed by AI tools. SEO specialists use solutions like BacklinkGPT, Pitchbox, Respona, Semrush Link Building Tool, and Ahrefs Brand Radar. According to Omniscient Digital, 70% of link builders use AI daily, and 88.6% of them treat ChatGPT as a “second brain” for quick analysis and drafting outreach emails (Omniscient).
Automation frees up specialists’ time for creative campaigns and relationship-building—areas where human negotiation skills and storytelling remain unrivalled. While algorithms can find hundreds of potential link sources within minutes, it’s humans who should ultimately decide which are truly worth the effort and how to build relationships that result in natural, contextual, and lasting links.
The Quality and Context of Backlinks
The era of mass link spamming is long behind us. Every SEO professional knows quality matters—especially after the Helpful Content Update (2022), which targeted content created purely “for SEO.” Today, the quality and context of backlinks are more critical than ever.
AI’s ability to generate massive amounts of text has led to an explosion of mediocre, generic, and duplicate content. Entire “farms” of AI-driven PBN-style sites are being built solely for mass link placement. This shrinks the pool of high-quality websites worth collaborating with. And it’s no secret that a single backlink from a trustworthy, authoritative website is far more valuable than dozens of low-quality ones.
An effective link-building strategy should focus on the substance and reputation of sources. In the AI era, the emphasis must be on acquiring dofollow links from websites with strong E-E-A-T (Experience, Expertise, Authoritativeness, Trustworthiness) and high domain authority. The more such links a site has, the clearer the signal—not only to Google but also to LLMs—that the site is credible and worth citing. Thematic alignment also matters: links from outside the relevant context risk being flagged as unnatural.
As AI-generated content floods the web, links are becoming an increasingly critical signal of authority, relevance, and trust for both Google and language models to distinguish which sources truly matter.
Robert Nyberg, Founder, Senior SEO Expert at New Customers AB
The Rise of Digital PR
Link building is becoming increasingly intertwined with brand-building and relationship-building. Beyond classic sponsored articles, naturally acquired links through relationships and outreach are gaining importance. This is where digital PR campaigns come in—strategically leveraging online platforms and communication channels to increase visibility and build brand reputation.
This link-building technique aligns well with LLM behavior, as models use such naturally earned links and citations in their responses. While the process is time-consuming—since links are obtained through relationship-building—the results are worth it. Digital PR combines creative, expert content with backlinks and mentions in high-DR, high-traffic media outlets, precisely the type of signals algorithms need to trust a brand.
One prominent trend is content-led digital PR—a strategy that begins with an engaging, proprietary “hero asset” (e.g., a data report, interactive map, calculator, infographic, or mini-app) created specifically to earn media coverage, links, and online buzz. Data-led campaigns are also popular, where the “hero” is a number: proprietary or public data transformed into a story, visualization, or tool solving a real user problem. Language models prefer clear facts, since they’re easy to verify and quick to extract as insights.
For link building, this means that high-quality content, supported by strong distribution, will naturally earn links. Websites must publish material useful and credible enough for others to want to link to. That’s the essence of digital PR—one of its most tangible and measurable outcomes.
Digital PR is currently one of the most effective tactics for a brand to gain significant, large-scale coverage in well-known publishers, to connect with specific relevant topics of its choice, and to build authority thanks to that coverage. It can also contribute to increasing brand awareness and creating a more natural backlink profile. Most importantly for me, however, is that Digital PR is a ‘weapon’ we can direct towards any SEO goal we want, in line with our strategy, and use it to boost rankings.
Alex Galinos, Group Head Of SEO at hoppa
And what do the numbers say? According to Motive PR:
- Interest in digital PR is growing, reflected in a 71% increase in global searches for the term “digital PR” over the past 5 years.
- Digital PR campaigns, from launch to peak reach, earn on average 42 unique domain links.
- 20.62% of backlinks acquired through digital PR come from domains with DR between 70 and 79.
Digital PR is gaining serious momentum, and well-executed campaigns can serve as the foundation for building strong website authority in the AI era.
Digital PR is redefining link building today. It’s not just about backlinks, but also about reputation signals coming from mentions and brand citations in credible sources. Search engine algorithms still count and recalculate links to some extent, but language models take a broader view. They don’t operate like traditional indexing algorithms; instead, they look for patterns of credibility. They analyze context, pick up information, and absorb authority signals even when there’s no link.
It’s precisely these signals that determine whether a brand will appear in AI Overviews and other generative systems as a trustworthy source. Well-executed Digital PR doesn’t replace classic link building tactics — it strengthens them, adding a layer of reputation and credibility that links alone simply cannot provide
Agata Gruszka-Kierczak, International SEO Manager at WhitePress®
“Nofollow” Links and Brand Mentions
Links marked with the “nofollow” attribute don’t pass SEO power (link juice) and are largely ignored by crawlers when it comes to evaluating the target page. That doesn’t mean they’re useless. On the contrary: since generative systems filter the web to highlight the most trustworthy sources based on credibility and frequency of mentions, reviews, or citations—every reputation signal matters, including nofollow links.
The same applies to brand mentions without any links. Why? Because today, validation matters more than visibility. Visibility means showing up in search results, while validation means being mentioned on forums and “Best of X” rankings, cited by ChatGPT, reviewed on Amazon or marketplaces, or discussed in Reddit threads. It’s about confirming a brand’s authority/
Brand mentions are the new backlinks in the age of AI search. While Google counts links, AI counts conversations
Britney Muller, Founder Of AI Marketing Accelerator
It’s also worth tracking brand mentions in real time, with tools like Brand24 or Mention. This allows for quick responses to link-building opportunities—sometimes a simple email to the site owner asking them to add a link is enough.
One caveat: language models may occasionally misattribute or misquote brand names.
Entity Profile Building
Language models construct answers by linking together recognized entities—companies, people, products, organizations. The fuller and more consistent the entity profile, the more likely algorithms are to identify the website as a trusted source. Links and mentions play a key role in establishing an entity’s popularity and topic relevance, but additional factors matter as well:
- Consistent core data (NAP) – uniform name, address, and phone number across Google Business Profile, Bing Places, Apple Maps, and relevant directories like Yellow Pages or BBB.
- Structured data (schema.org) – e.g., LocalBusiness, Organization, Article markup.
- AI-intensive platforms – channels like YouTube, Medium, LinkedIn, or Reddit that are heavily scanned by AI.
Combined with high-quality content, these create a clear, repeated signal to language models that the entity is authoritative, visible, and worthy of citation.
SEO Isn’t Dying—It’s Evolving
We can view the revolution in search as an obstacle and cling to old methods, or we can treat it as a turning point: looking at SEO holistically and shifting from Google-only optimization to building multichannel visibility. Today’s key is blending classical SEO techniques with new GEO practices. Algorithms still evaluate links, structure, and content, but increasingly reward consistent entity signals and citability across channels that language models intensively scan.
The faster we adapt to this new reality, the greater the chance we’ll gain an advantage before competitors react.
Keep this in mind when thinking about SEO and link building in the AI Era:
- Optimize everywhere: classic SERPs + LLMs + social search.
- Create AI-ready content: conversational tone, concise and direct answers, one H1 and logical H2/
H3 structure, TL;DR summary section, source data, FAQ section. - Build authority with premium links: industry domains, digital PR, marketplace partnerships; remember, nofollow links and mentions also count.
- Strengthen your entity: consistent NAP, schema markup, reviews, GBP/
Bing/ Apple profiles. - Automate the routine with AI, leave the rest to specialists: prospecting and scoring → bot; personalized pitching and strategy → human.
Content Creator
