Google's AI Overviews reach 2 billion monthly users and cut organic clicks by 68% in 2026. For SaaS companies, ecommerce brands, and the agencies that serve them, all have felt the big shift that visibility now depends on being cited inside an AI-generated answer, not just ranked on a results page and most brands don't know how to do that.
What is AI search, and why is this the most important time to join in?
AI search refers to any question asked that's answered by the top AI models like Google AI Overviews, Claude, ChatGPT, Perplexity, Gemini, etc.. with a generated answer instead of the link list we've been used to for years. Users are no longer browsing 2-5 pages and picking one. Instead AI system is scanning the internet, making assumptions, and deciding the answer in seconds.
This isn't an emerging trend anymore. Google confirmed AI Overviews had reached 2 billion monthly users across 200+ countries by mid-2025, up from 1.5 billion just two months.
The shift isn't confined to Google. ChatGPT hit 800 million weekly active users by October 2025, Perplexity had 780 million queries in a single month (May 2025), Google's Gemini reached 450 million monthly active users the same quarter. All these show just how fast searching has changed and will continue to change. That's why the most important time to start getting your brand recognized by AI is right now.
Why the top brands are losing traffic but increasing revenue?
Most brands are stuck tracking metrics that worked for regular search. But AI search is changing what you have to look at.
Pew's controlled study of 68,879 Google searches found that when an AI summary appeared, users clicked through to a traditional website only 8% of the time compared to 15% when no summary appeared. A near 50% relative drop in click-through behavior. Only 1% of users clicked a link inside the AI summary itself. Users ended their entire browsing session on the results page 26% of the time when a summary appeared, versus 16% without one meaning more searches now end without a single website visit at all.
But these users are still making decisions and purchasing just not through the channels you've been focused on.
Are these AI-referred visitors worth anything?
Yes and they're more valuable they typical visitors.
Tracking over a trillion visits to US retail sites, Adobe found that during the 2025 holiday season, traffic from generative AI tools rose 693% year-over-year and those AI-referred shoppers converted 31% better than shoppers from any other channel, spent 45% more time on-site, and generated 254% more revenue per visit.
And this is the smallest AI search traffic will ever be again. Meaning if you get in right now it will only increase in value as time goes on.
Why does this matter more for SaaS, ecommerce, and agencies specifically?
Each of these three business types is exposed to the same shift, but the mechanism and cost of ignoring it differs.
Business type | Where AI search hits hardest | What the data shows |
|---|---|---|
SaaS / B2B software | Vendor comparison and evaluation research, where buyers now ask AI tools to shortlist options before ever visiting a site | Gartner found 45% of B2B buyers used generative AI during a recent purchase, drawing from an average of 7 information sources; Gartner separately projects traditional search volume will drop 25% by 2026 as buyers shift to AI agents |
Ecommerce / DTC | Product discovery and pre-purchase research, where AI-referred shoppers convert dramatically higher than average | Adobe saw AI referral traffic up 693% Omnisend found 59% of US consumers already use GenAI tools for shopping, and 1 in 4 say ChatGPT beats Google for product research |
Agencies | Future clients are asking AI models to do research on agencies before they even get on a call with them. | Conductor found 97% of digital leaders report AEO/GEO positively impacted their funnel in 2025, and 94% plan to increase AEO investment in 2026 |
For SaaS brands, the risk is being excluded from the shortlist entirely, if an AI tool never mentions your product while comparing options, you miss the user entirely in their discovery phase.
For ecommerce brands, if AI doesn't think you match the users needs you won't even be mentioned at all meaning potential customers won't know you exist.
For agencies the risk is credibility, clients will increasingly ask AI about you and if the answer isn't what they want they leave without you getting the chance to talk.
What actually makes a brand get cited by AI search engines?
This is the question everyone asks, and it's where most advice gets it backward. Most teams are focused on what worked for SEO when the game has changed completely.
Ahrefs analyzed 75,000 brands and found that off-site brand mentions correlate with AI Overview visibility roughly three times more strongly than backlinks do. More then 26% of brands don't have an AI Overview at all despite having SEO properly set up.
The peer-reviewed KDD 2024 "Generative Engine Optimization" study, run across roughly 10,000 queries by researchers from Princeton, Georgia Tech, and the Allen Institute for AI, found that adding cited statistics, quotations from credible sources, and factual density to content increased visibility in AI answers by over 40%.
And importantly the same study found that keyword stuffing (the old SEO reflex) actually did nothing at all.
And here's the myth worth killing outright: the fix isn't just an llms.txt file or special AI markup to be cited by Google's AI features.
Google's own Search Relations team has said on the record it does not support llms.txt and has no plans to, comparing it to the long-abandoned keywords meta tag. Google's official 2026 guidance states plainly that such files "will neither harm nor help" visibility, because Google Search ignores them entirely.
What actually moves the needle, based on evidence:
Earned mentions over backlinks. Digital PR, founder-led commentary, and third-party coverage build the brand-mention signal found most correlated with citation.
Fact-dense, sourced content. Original statistics, named citations, and quotable specifics, not keyword density, are what the KDD study measured as effective.
Presence on high-citation platforms. Places like Wikipedia, YouTube, and Reddit are among the most-cited sources in AI summaries; being discussed there matters more than most technical SEO checklists suggest.
Clean, crawlable structure not special AI files, but the same fundamentals that have always mattered: fast, accessible pages with clear semantic HTML.
How are marketing teams and agencies responding right now?
Teams are allocating more budget and focus on these AI answers moving from traditional SEO. Conductor's survey of 250+ digital marketing leaders found that 56% made significant AEO/GEO investments in 2025, and on average, teams are now allocating roughly 12% of digital marketing budgets to AI visibility work.
In a separate industry poll reported by Digiday, 55% of marketers said they already have dedicated budget earmarked specifically for GEO.
That reallocation is happening before most brands have full measurement in place which is itself a warning sign. A 2026 Forbes Business Council survey of 503 CMOs at companies with $500M+ in revenue found that 87% now have a board- or CEO-level mandate to develop an AI search strategy, but only 21% have operational practices actually in place to execute one.
Most brands are stuck right now knowing they have to build their infrastructure up here but have no idea where to start.
What should a brand actually do about this?
Here's the sequence that matches how the data says AI visibility is actually earned:
Baseline your current AI visibility. Before changing anything, find out whether your brand is currently mentioned or cited by ChatGPT, Google AI Overviews, and Perplexity for the questions your buyers actually ask. Don't assume you're already showing up.
Shift a defined budget slice toward earned mentions, not just content volume. Plan to allocated time, research, and money towards a real plan that will get your brand ranking and not just running through the same motions.
Rebuild content around fact density, not keyword density. Every page should carry original statistics, named sources, and quotable, specific claims.
Track AI-referred sessions separately in analytics. Given Adobe and Semrush both found AI-referred visitors convert meaningfully better than average traffic, treating this traffic as a rounding error inside "organic" undercounts its value.
Skip the llms.txt trend for Google specifically, and put that effort into genuinely crawlable, well-structured content instead — Google has said directly that the file does nothing for its systems.
Work with an agency, companies like Sorca can do all of this for you and get your brand ranking much quicker.
Is this just a Google problem, or does it affect every platform the same way?
That's what makes AEO/GEO so difficult. Each model has it's own differences and way of working that most teams can't keep up with optimizing for each.
Google has been explicit that AI Overviews and AI Mode run on its Search index. Focusing on crawlability and content-quality fundamentals that have driven organic rankings for years still apply.
Platforms like ChatGPT, Perplexity, and Copilot work differently. They draw on their own retrieval systems, often leaning more heavily on third-party sources like Wikipedia, Reddit, YouTube, review sites, and news coverage rather than crawling a brand's own site directly.
The practical takeaway: Don't expect one method to work everywhere, if you want to see success in AI search you have to make sure you're showing up where all the AI's are looking. The easiest way to do that is by working with a team that has the systems to get you in all the right spots like Sorca!
