Google has been beating the AI drum hard lately when it comes to its new Search Generative Experience (SGE). Naturally, this has put the SEO community on edge. While Google has admitted that AI Mode and AI Overviews wil inevitably cannibalize regular organic search, they’ve also said that there’s a trade-off: you’ll be getting fewer but better-quality clicks.

Supposedly, when AI cites your content, the users who follow through will be more informed, more engaged, and more likely to convert.

But theory doesn’t always survive contact with the real world.

In a newly released study, Ahrefs analyzed traffic across 81,947 websites between May and June 2025. Their goal? To compare how visitors from AI platforms—like ChatGPT, Perplexity, and Microsoft Copilot—behave versus traditional search engine users.

The results weren’t exactly what Google’s PR team would want you to see.

Instead of deeper engagement, AI traffic showed signs of being more volatile, less curious, and quicker to bounce. Even though there were glimmers of hope in the form of potentially higher conversion rates, the overall engagement data challenges the idea that fewer clicks = better clicks.

And that forces us, as marketers, to ask some tough questions:

  • Are we optimizing for the right type of traffic?
  • Is AI referral really the holy grail of content distribution?
  • Or are we trading depth and loyalty for a shallow, one-click economy?

Let’s break down what the numbers really say—and what they mean for your SEO and content strategy moving forward.

The Key Numbers: What Ahrefs Found

To understand how AI traffic compares to search traffic, Ahrefs studied behavior metrics across nearly 82,000 websites. The data covered users coming in from AI platforms like ChatGPT, Perplexity, and Copilot, and benchmarked them against users from traditional search engines.

Here’s what the numbers looked like:

  • Bounce Rate. Visitors from AI sources bounced 4.1 percent more than those from search engines and 5.4 percent more than the overall traffic average.
  • Pages per Visit. AI traffic averaged 4 pages per session. Search traffic averaged 5.2 pages, and the site-wide average stood at 5.5.
  • Session Duration. AI visitors spent an average of 86 seconds per session. That’s 8 seconds longer than both search and average traffic.
  • Pages per Minute. Despite spending slightly more time on site, AI users viewed fewer pages per minute, suggesting lower overall engagement.
  • Conversion Rate. According to Ahrefs, AI traffic may convert better, with some sites seeing around a 12 percent conversion rate from AI sources.

These figures paint a picture that contradicts the narrative that AI search will lead to higher quality visits. More time on site may seem like a good thing, but when paired with fewer page views and higher bounce rates, it raises questions about how much users are actually engaging.

What the Numbers Actually Mean

On the surface, some of the data might not look all that bad. AI visitors spend slightly more time on site, and there’s a hint of stronger conversion rates. But once you dig deeper, the engagement metrics tell a more complex and less flattering story.

A. Lower Engagement, Not Higher Intent

Google and AI platform advocates love to suggest that AI-driven users are more decisive and better informed. That might be true in some cases, but the data shows that these users are also more likely to bounce and less likely to explore.

That suggests a different reality. AI users are probably clicking on links to confirm something quickly, not to research or dig deeper. This is not the behavior of someone actively shopping, evaluating solutions, or trying to build a relationship with your brand.

B. Longer Sessions Aren’t Always a Win

Yes, AI traffic had slightly longer session times. But when users spend more time on a site while visiting fewer pages, it’s often a sign they’re stuck or passive—maybe reading a single long page or leaving a tab open.

It doesn’t automatically mean better engagement. It might even point to poor UX or a disconnect between the AI’s summary and the page content.

C. Conversion Without Context Is Dangerous

That 12 percent conversion rate looks great in isolation. But conversions mean nothing if they aren’t backed by long-term value. Are these users coming back? Are they buying again? Are they sticking around?

AI visitors might convert at the bottom of the funnel because they’re dropping in at exactly the right moment. But that doesn’t mean they’re more valuable overall. You need to look beyond the first conversion and track retention, average order value, and customer lifetime value before calling this traffic “better.”

What Search Marketers Should Do Next

AI traffic isn’t inherently bad, but it behaves differently. If you treat it like traditional search traffic, you’ll miss the mark. Here’s how smart marketers can respond.

A. Segment AI Traffic in Your Analytics

Stop bundling AI clicks in with your organic traffic. These visitors behave differently and need their own performance benchmarks. Use GA4 custom reports or UTM tagging to separate AI referrals from Google, Bing, or direct traffic.

Once segmented, monitor metrics like bounce rate, pages per visit, and conversion rate in context. That gives you a clearer picture of what’s actually working.

B. Tailor Landing Pages for Short, High-Intent Visits

AI users often land with a specific goal. They’re not there to browse—they’re there to get a straight answer or take quick action. Create pages that front-load value, tighten your headlines, and surface your calls to action early.

Think “landing page logic,” not long-form blog post structure.

C. Don’t Abandon Traditional SEO

AI traffic might be growing, but it’s not replacing the value of organic search. Google’s core search still drives discovery, long-tail visibility, and brand engagement. That kind of traffic builds long-term value and trust, not just one-off conversions.

Keep investing in content that builds topical authority, earns backlinks, and performs well in regular search.

D. Align KPIs With Traffic Behavior

If you’re measuring AI traffic by the same standards as search, you’re likely drawing the wrong conclusions. AI clicks might never give you deep engagement or multi-page sessions. Focus instead on micro-conversions, scroll depth, or single-action performance.

And for traditional search, continue to measure broader engagement like time on site, newsletter signups, or returning visits.

Don’t Fall for the Hype

AI-driven traffic is the newest shiny object in search marketing, and it’s tempting to believe the hype. More targeted clicks, better-informed users, higher conversion rates—it all sounds great on paper.

But the data tells a more complicated story. AI visitors bounce more, view fewer pages, and engage less overall. The clicks might be more decisive, but they aren’t necessarily more valuable.

As search marketers, it’s our job to go beyond surface metrics and build strategies based on real behavior. That means:

  • Separating AI traffic from traditional sources in your reporting
  • Designing experiences for fast-moving users who want direct answers
  • Continuing to invest in organic search that builds trust and long-term brand visibility

Most of all, it means remembering this: Traffic is only as good as what it does for your business. Engagement, loyalty, and conversions still matter. AI is changing the game, but the fundamentals haven’t changed.

Don’t chase trends. Understand them, adapt to them, and use them to your advantage.