Google AI Overviews: Sorting Knowledge from the Noise
![Google AI Overviews: Sorting Knowledge from the Noise](/sites/default/files/2024-06/TPT%20DIGITAL%20-%20NO%20WORDS.jpg)
![](/images/down-arrow-w.png)
Will Google’s Greek Chorus Approach to Search Deliver Answers or Anarchy?
My question, asked to a vanload of college buddies on our way to Spring Break 1997, was simple: Did I just miss our exit? The range of responses I got was typical of that group: thoughtful, if unfocused.
Craig wondered where I’d stashed the printed map. John opined—not for the first time—that we probably should have stopped for coffee an hour ago. Jared resumed preaching about how driving on the inside curve would save gas mileage. Ben started rerouting us verbally through a neighboring state. And Holly seized that moment to swap my Eagles mixtape for Tom Petty.
Never mind that I was only seeking a definitive yes or no.
It’s been two weeks since Google officially rolled out AI-generated responses in search results, and I’ve been reminded of that trip nearly 30 years ago—a whole lot of noise with no knowledge, yet, of where it’s all headed.
From Google’s perspective, search gets better and faster. Customers can enter increasingly more complicated, more conversational searches, and AI Overviews can summarize a range of information that anticipates further questions and how that data can best be put to use.
For example, typing the query “Why is my garage door not opening, and what are the ways to fix it?” returned an AI Overviews list of things to try and test while also helpfully reminding me how to pull the emergency release cord to manually open the door. Sources for the troubleshooting tips were arranged in a scrolling row beneath the AI response, and my search results also included local business listings and ads for garage door repair.
![Google AI Overviews](/sites/default/files/inline-images/GoogleAIOverviews.jpg)
Sharp-eyed Googlers will notice parts of the AI response are stitched together from the same articles and forums it cites. So, some of the best information on the internet can still be one click away. But therein lies the existential dread gripping those same publishers and the SEO world: as the AI Overviews box increasingly dominates the top of search results, it also pushes ads and the old status quo of ten top “blue links” further “below the fold” and out of customers’ sights. This is true of the shortcuts just beneath the search field as well. You’ll find links to results for shopping, forums, videos, and images displayed prominently. For web results—those same authoritative articles—you’ve got to first click on “More” at the far right.
This is a change agency experts have been bracing for since Google introduced Search Generative Experience in its Search Labs testing ground last year. Although some initial tests have found that just 14% of search queries automatically return AI Overviews results, and 58% of queries have no AI Overviews results enabled (down from AI Overviews showing 65% of the time during the SGE phase), these same prognosticators expect AI-fueled summaries to encompass more than 80% of keywords across most industries and verticals.
According to the Associated Press, the emergence of AI Overviews in Google, as well as the enhancement of competing search engines like Bing and OpenAI, could dent organic traffic by as much as 25%. Not to mention cutting into Google’s own $175 billion ads business.
So how can innovation-minded brands stay ahead of the AI wave in SEO?
- AI Overviews still relies on crawling the web to assemble reputable sources. As we noted in our blog about avoiding SEO mistakes, global companies need to stay on top of technical SEO to ensure their sites are crawlable and their content is findable.
- Experts studying Google’s algorithm updates, including uncovering clues to the search engine’s inner workings in a major leak, know that sites historically earning organic clicks and holding onto users with relevant content are more valued by Google and are likely sources to train large language models. Keep producing reputable content.
- Follow Google’s advice: invest in your brand reputation and improve your “E-E-A-T”—develop content that shows your experience, expertise, authoritativeness, and trustworthiness. In the AI Overviews era, expect this to mean getting your content out to other channels, including PR, social media, and web forums, and gaining that old SEO staple—reputable backlinks.
- Organize your content so it hews even more closely to the user journey Google is modeling with AI Overviews. Map out questions and their broader context so that answers on your website follow the same paths search results aim to illuminate. For example: someone searching for an auto insurance quote is likely to consider what a fair price would be, which companies are more reputable, how they package other coverage, and even what type of car they should buy to balance the cost of insurance. Search Engine Journal advises creating topic overview pages to position yourself as a source.
While publishers are going through growing pains, realize that Google is, too. Since its wider release in the US, examples abound of results AI has flubbed. This includes the so-called “Your Money, Your Life” queries, where Overviews advised adding glue to pasta sauce, bathing with toasters, and eating “at least one small rock per day.”
These results are cobbled from the same mix of reputable blogs and forums like Reddit and Quora that Google seeks to draw from. As it eyes a global release for AI Overviews, Google must evolve to deliver a product that is consistently helpful, intuitive, and factually sound. Or else, as the answers amplify, so will the noise.
Interested in exploring more global insights on digital marketing, AI and important marketing trends to enhance your business? Visit https://www.transperfectdigital.com/ to learn more or get in touch with us at TPDigital@TransPerfect.com