AI search traffic is expected to overtake traditional search by 2028
We’re watching search as we know it evolve in real time. Large language models like ChatGPT and Claude are reducing friction in the search process. They’re changing how people find and act on information. Users ask a question, get an answer in seconds, no clicking through ten links to dig for insight. That kind of instant clarity is hard to compete with.
According to current projections, AI-driven search will surpass traditional search traffic by 2028. That’s not a maybe. ChatGPT, for example, grew its weekly active user base 8x from October 2023 to April 2025, now topping 800 million. Google’s not standing still either. It’s rolling out AI Overview and AI Mode, shifting from classic search results to giving users simplified, direct answers. Behavior’s already shifting. People want faster, more usable inputs.
If Google fully defaults to AI Mode, we could hit this tipping point even sooner.
Now here’s what matters: as AI search grows, traditional traffic will shrink. In this transition, total traffic might decline temporarily. That’s because AI systems often answer questions directly, reducing click-throughs. But quality visits won’t disappear, they’ll be concentrated. The window to claim your position in AI results is open, and the businesses that move early will capture a disproportionate share of visibility.
Executives need to start thinking beyond traditional SEO. Visibility in AI environments depends on how your content is learned and represented by these systems. That includes not only what’s on your website, but where else your brand is mentioned.
The average AI search visitor has a conversion value 4.4 times greater than that of a traditional search visitor
When someone visits your site after using an AI search tool, they’re different. They’ve already done their research. The AI just summarized value props, comparisons, and pros and cons. So by the time they land on your business, they’re close to taking action.
The data is clear: the average AI search visitor is 4.4 times more valuable than a traditional organic search visitor. And this is just the beginning. AI search is still early in adoption, but it’s already generating traffic that converts better, faster, and with less friction. What that tells us is simple, the quality of traffic matters more than quantity.
Executives should see this as leverage. You don’t need more visitors, you need smarter entry points. If a user comes in already understanding your offer, they need less hand-holding. That saves marketing costs and shortens sales cycles.
AI search works more like direct recommendation than exploration. When a model references your brand in context, say as the best option in a certain category, it carries the perceived weight of a trusted advisor. That creates trust before a user even hits your site.
The good news: the skills that earned you traditional SEO visibility still matter. Publishing content that’s helpful, crawlable, and backed by real expertise lays the groundwork for visibility in AI environments. From there, it’s about refining how your message gets picked up by these systems. That means formatting content for clarity, maintaining consistent messaging across channels, and avoiding fragmented brand narratives.
Content needs to be machine-readable and reputationally strong. Don’t just aim to rank, aim to be understood, summarized, and quoted correctly. That’s where the conversion efficiency lives.
Traditional high-ranking SEO positions are less influential for AI search citations
In AI-driven search, your position on a standard Google results page matters less than you think. The models powering ChatGPT and other language-based tools value relevance at a granular level, individual paragraphs, statements, or data points, not page-wide authority.
Our data shows something leaders need to take seriously: nearly 90% of ChatGPT citations come from content ranked 21st or lower in traditional search results. That’s a reversal of conventional wisdom. Even sites buried in Google’s back pages are now earning visibility in high-intent AI-generated responses. Perplexity and Google’s own AI systems show similar patterns.
Here’s what that means strategically: content that addresses highly specific questions or edge-case use cases carries new weight. AI systems don’t index and rank like a search engine. They extract and synthesize. So whether you rank in the top 3 or not, if your information is clear, comprehensive, and relevant to the user’s query, it’s in play.
This kind of shift gives businesses more access, if they’re focused. Don’t just chase broad keywords or high-volume phrases. Craft content that answers nuanced, often-overlooked queries directly. That includes product guides, user support documentation, and explanation-driven blog content.
Also, AI tools are designed to understand intent on a deeper level. They parse natural language better than traditional engines and can contextualize over time. That’s a competitive edge for businesses creating content with precision and clarity. Page speed and exact keyword matches don’t hold the same weight here.
Priority needs to shift from page-level SEO metrics to content utility at the micro level. It’s not about gaming rankings, it’s about being the most useful voice in the conversation.
Quora and reddit are the most frequently cited sources in google AI overviews
Looking at what LLMs cite most in Google AI Overviews, one thing becomes clear: platforms with real discussions and user-generated insights dominate. Quora tops the list. Reddit is right behind it. These sources win because they contain content people actually use to solve specific, often overlooked problems.
Google’s models pull from these platforms not because of brand equity, but because the content is structured around questions, answers, and tangible experiences. That’s exactly the kind of information AI systems are trained to elevate, direct, targeted, and problem-solving in nature. Reddit also benefits from a deeper partnership with Google, where its data is being formally used to improve search and AI outputs.
For leadership teams in marketing, PR, or growth, this is a signal to adjust. Visibility in AI models isn’t just about your owned properties. Mentions in community-based ecosystems like Quora and Reddit carry weight, and they show up in front of users who are ready to act.
This doesn’t mean your brand needs to dominate Reddit threads. But being present on high-authority, frequently cited domains is a strong AI visibility signal. Using digital PR initiatives to earn citations in these environments can directly impact how AI systems interpret and respond with your brand.
Executives should also focus efforts on securing mentions from trusted media outlets, industry forums, or vertical-specific authority sites. The models reward wide and relevant presence, not just domain-level authority.
This is a broader shift in influence mapping. People often trust experiences and dialogue more than polished web copy, and it turns out AI does, too.
Half of ChatGPT citations are directed toward business/service websites
When ChatGPT responds to queries, about 50% of the links it includes point to business or service websites. This is a critical signal. AI systems aren’t just quoting community forums or media. They’re relying heavily on websites like yours to support responses on products, solutions, and services.
This changes the expectations for brand websites. They’re no longer just destinations for search traffic, they’ve become assets in training and informing AI responses. If your site delivers useful, structured, and well-articulated content, it stands a strong chance of being cited directly in AI answers. That creates visibility at the point when users are forming a decision or already leaning toward action.
To take advantage of this, content needs to be prepared in a way that aligns with how machine learning systems consume and interpret text. Clear structure. Defined headings. Clean layout. Strong language semantics. Presentation matters.
Multimodal content, text combined with visuals, video, or diagrams, also adds value. LLMs are moving toward interpreting more than just text. Giving them more forms of useful data expands how your site can be surfaced and referenced. Executives should view content now not just as marketing, but as data points that feed an ecosystem of machine-driven interpretation.
There’s another layer here: natural language processing (NLP) is how LLMs understand content. This means writing needs to be precise, context-rich, and phrased in a way that matches how humans ask questions. Add internal cues like comparison tables, product lists, FAQs, and case studies. These aren’t just UX features, they’re machine-readable artifacts that increase your odds of representation.
Lastly, any technical obstacle that prevents AI from crawling your site, like heavy JavaScript rendering, can sink your visibility. Crawlability is non-negotiable. If your site can’t be processed cleanly, you’ll be invisible no matter how valuable your content is.
The path forward is simple, but requires commitment: publish content that is technically accessible, highly relevant, and designed to be understood by systems, not just people. LLMs are assigning visibility based on utility, companies that align their digital assets accordingly will benefit from long-term returns.
Main highlights
- AI search is set to overtake traditional search: Leaders should begin reallocating SEO resources toward AI-driven platforms as AI search traffic is projected to surpass traditional search by 2028, with user behavior already shifting rapidly due to tools like ChatGPT and Google AI Mode.
- AI traffic delivers higher conversion value: Executives should focus on optimizing for AI visibility, as visitors from AI search are 4.4x more valuable than traditional search users, arriving better informed and closer to purchase decisions.
- Traditional rankings matter less in AI search: Organizations should invest in creating content tailored to specific use cases and high-intent queries, as LLMs often cite pages ranked well below the top 20 in traditional search results.
- High-authority communities drive AI citations: Marketers should build brand presence on platforms like Quora and Reddit, which are top sources in Google AI Overviews, due to their rich, user-generated content and trustworthiness.
- Business websites are heavily cited in AI responses: Companies must make their sites LLM-friendly by publishing structured, NLP-optimized, and multimodal content, since 50% of ChatGPT citations point directly to business and service websites.
