With the rise of generative AI, the top of the marketing funnel is getting more complex. The digital marketing funnel from search to consideration to purchase is fragmenting into exploration across more varied digital channels, making it more challenging to capture a consumer’s attention and lead them where we want them to go.
More and more consumers, who previously turned to Google search for product recommendations — whether for your brand, category or a specific item you sell — are now experimenting with AI tools like OpenAI’s ChatGPT, Google Gemini and others. Even if they opt for a traditional Google search, they may be served an AI Overview summary of the answer to their question at the top of the results page, above the links to ratings and reviews and above the paid search ads.
The Challenge and Opportunity of Google’s AI Overviews
That’s a challenge for marketers if it discourages potential customers from exploring more. Whether on your website or a third-party web page that includes a positive review of your product and links out to a place to buy, we want to lead the interested consumer down the path to a purchase. Except now that path may be blocked — about 75% of searches that trigger AI Overviews wind up as zero-click searches.
This is mostly a top-of-the-funnel phenomenon — 90% or more of the searches for which consumers are presented with an AI Overview have an informational intent as opposed to a transactional intent.
If the searcher enters a “best deals on” search for a product, Google will happily serve up paid and organic content appropriate for transactional intent, recognizing that the searcher is ready to buy and just wants a little more guidance. And consumers who have strong transactional intent are still more likely to turn to Google than ChatGPT for guidance.
Ensuring Your Brand is on AI’s Radar
Yet digital marketers recognize that success depends on effective marketing throughout the buyer’s journey, starting at the point where the consumer becomes aware of the existence of a product or service and wants to learn more.
In research on fast fashion brands, we found that given a query on “affordable fashion” — a term for which Shein tends to rank high in traditional search — AI chatbots like ChatGPT tended to skip over Shein and recommend more traditional retailers. While not a hard and fast rule, AI tools tend to go for the more established brand over the upstart.
If you’re a digital leader at Shein, your challenge is to influence AI models to recognize and prioritize your brand. If you are a consumer packaged goods (CPG) company that sells primarily through retailers rather than your own website, this may be a good time to reinvest in your brand website. Even if it’s not a direct sales driver, it strengthens your brand’s online presence, making it more visible to AI models that pull information from the web. Beyond that, you want your brand to be mentioned in other places, as many as possible, to get on the chatbot’s radar.
Winner Takes All: The Renewed Importance of Search Rankings
The first companies to feel the impact of AI-driven traffic declines have been publishers — businesses built on providing information. For example, when consumers were searching for Thanksgiving and Christmas holiday recipes, cooking magazines and recipe websites generally captured less traffic in 2024 than in 2023, very likely because questions like “how long to cook a turkey” were adequately answered by an AI Overview. This decreased the likelihood that the consumer would click through to read even a #1 ranked article on that topic thanks to AI-generated previews.
Traffic to most of the well-known cooking and recipe websites was down by double-digits in December 2024, including a 37% year-over-year drop for Epicurious.com. The exception was Allrecipes.com, where traffic was up 13.9%. When AI Overviews appeared for searches where Allrecipes.com had a top-ranked article, the site was frequently cited in a sidebar as the source of that information, then listed again in the search results, keeping it visible and driving clicks.
Even if Allrecipes.com lost some traffic because of AI Overviews, it was in much better shape than any website farther down on the page. Since the beginning of search marketing, we’ve understood the importance of ranking at the very top of the results list, but an AI Overview redoubles the value of ranking number one.
This is happening at a time when other consumer behaviors are also changing, for example, with younger people being more likely to search YouTube than Google for advice. Or they may go to a community site like Reddit seeking answers from people rather than algorithms. That’s another reason to seek favorable mentions of your brand all over the place, not just in traditional media.
The Influence of AI Overviews on Product and Branded Searches
While AI Overviews rarely show up on searches where the intent is clearly transactional, they do appear on searches aligned with the consideration phase of the buyer’s journey, when the purchase decision is not final and neither is the brand or product choice.
For example, while a search for a specific Milwaukee Tools power tool would put a series of buying opportunities right at the top of the page, a comparison query like “milwaukee tools versus dewalt” often triggers an AI Overview.
In this case, the AI Overview cites a Reddit discussion thread as the primary source of its summary, which praises Milwaukee Tools for quality and durability but gives DeWalt the edge on price. As a marketing leader for either of these brands, or for a home improvement store trying to boost sales of power tools, you might prefer that the consumer see one of the other reviews farther down on the search results page — maybe on a site that includes your affiliate link. The AI Overview makes it significantly less likely that shoppers will get there.
For perspective, in a rank tracking analysis using Home Depot as an example, AI Overviews were triggered for a tiny fraction of the Milwaukee Tools keywords the retailer ranks for — a dozen or so out of a sampling of more than 1,000. Even where they were detected, AI Overviews didn’t necessarily appear all the time. In the chart below, the diamonds represent the days when sampling of search traffic and results detected an AI Overview. For the Milwaukee versus DeWalt search, the overview was detected eight days out of 10. For other queries, AI Overviews appeared more sporadically. Tracking how often AI overviews are displayed for a given search.
This variation makes it even more critical to understand which of the searches that drive traffic to your website trigger AI Overviews, how often and whether they are important ones for your business.
AI Overviews: The Same, but Different
Google has signaled its plans to double down on its use of AI in 2025. AI Overviews are likely to become even more common, and that’s probably only one of many ways the technology will come to market across Google’s platform. We might see more generative AI features surface on YouTube, for example.
At the same time, OpenAI has introduced its own search engine as an integrated feature of ChatGPT, potentially driving more outbound links for those able to win the chatbot’s favor.
To a large extent, AI Overviews continue a process that’s been going on for years — Google building more summaries and instant answers into its SERPs (search engine results pages). The difference is that an AI Overview doesn’t just provide an answer; it also asserts a strong opinion on which sources are the most credible. At least AI Overviews are likely to be in sync with search, because they were introduced as an add-on to the familiar Google search experience. Whereas by incorporating search into ChatGPT, Open AI positions search as a secondary feature — where chat is the primary experience but you can get a link to click on if you really want one.
Whether a mass market of consumers will come to prefer the AI chat-centric way of getting answers remains to be seen. But if even a significant fraction of them do so, it will change what marketers must do to get consumers’ attention.
Focus More on Credibility and Visibility, Less on SEO
AI engines train on data from across the web, just as search engines do. But instead of indexing the web, AI engines seek to digest it, understand it and use it to answer questions. If your brand isn’t getting mentioned by AI engines, that’s a credibility and visibility problem much more than it is a technical SEO problem. Under the covers, the AI engine is likely tallying up all the places it has seen your brand mentioned favorably, so the challenge is to get your brand mentioned favorably in more places.
If AI engines are citing social media like Reddit as a source of truth, that implies marketers will have to do even more to ramp up influencer marketing programs. If their answers mirror what’s being said in the news or on ratings and review sites, make sure your public relations and other outreach efforts include the publications the bots seem to source from most faithfully.
Knowing that AI Overviews are more likely to appear on informational content also suggests the value of creating content associated with higher intent to purchase, for example, a “which product is best for me?” quiz that appeals to buyers already predisposed to buy your brand but unclear on which of your products is most appropriate.
Meanwhile, expect the buyer’s journey to continue to change as AI encourages further mutations across the digital ecosystem. In 2025, staying ahead of these changes will be challenging, but not impossible. You’re smarter than a bot, aren’t you? Aren’t you?