The other day, Lily Ray (Vice President of SEO Strategy & Research at Amsive) noticed Google had begun surfacing its own storefront for some competitive “Your Money or Your Life” searches.
By Google’s own admission, these kinds of “E-E-A-T” sensitive queries should be reserved for only the most authoritative and knowledgeable health sites.
Instead, they’re being claimed by Google-owned advertorials, touting the benefits of its Pixel Watch.
I did some digging in Ahrefs’ Site Explorer, and noted that 12% of Google Store’s pages have claimed an AI Overview in the last year.
That is no small thing.
For comparison, here are some similarly aligned ecommerce sites, and their respective AI Overview share:
- Samsung: 3.5% AI Overview share
- Apple: 2% AI Overview share
- Amazon: 0.4% AI Overview share
Google’s Store is taking more than double the AI overview real estate of their closest competitors, combined.
- Google Store AI Overview share: 12%
- Google Store competitors’ AI Overview share: 5.9%
And the number of Google Store pages appearing in AI Overviews is on the up.
Google is also steering flight searches toward its own product, Google Flights, as spotted by Carl Hendy (Founder of Audits.com).
To get to the bottom of just how self-referentials Google’s SERPs are, I went back to Ahrefs Site Explorer data to explore the organic traffic growth of 57 Google properties over the last two years. Here’s what I found:
- 50 Google properties have seen positive growth
- Only 7 have seen negative growth
- On average, Google’s properties have grown by 168% organically
- Google’s median organic traffic growth came in at 94%
In Q3 2024, Google’s revenue from search advertising reached $49.39 billion, signaling a 12% YoY increase.
And that’s thanks, in large part, to AI Overviews.
Google not only runs ads above and below AI Overviews, it now weaves them in natively.
According to analyst Doug Anmuth from JP Morgan, “AI Overviews are monetizing at roughly the same rate as non-AI searches.”
And Google’s “AI mode”—essentially an AI chat interface—will likely throttle paid advertising.
In short, AI overviews are boosting user engagement and satisfaction, opening up new ad space opportunities, and earning Google a pretty penny.
Meanwhile, we have recently found that AI Overviews reduce SERP clickthrough rate by a staggering 40%, while a similar study from Seer Interactive reports CTR drops of up to 70%.
Right now, we’re distracted with SEO busywork.
Chasing top of the funnel traffic. Creating Google-pleasing content for queries on the very periphery of relevance to our business. Spending untold hours trying to find untapped keywords in an ever-shrinking pool of… untapped keywords. And answering questions that Google is now perfectly capable of answering itself, using our content.
While we’re busy doing Google’s bidding, Google is busy looking out for #1.
Something needs to change.
Now that traffic is harder to come by, we need to refocus on building our brand and generating demand.
But Google is not the place to create that demand.
“Google has become a place people go *after* they discover a need rather than a demand-creation or even demand-nudging platform”
“For most companies, SEO is a channel for scaling growth, not achieving product/market fit. It’s a way to add fuel to your fire, not start the fire.”
Instead, we should be growing our owned audiences and raising brand awareness—whether that looks like creating subscription content (e.g. newsletters or paywalled articles), investing in direct publishing (e.g. Substack articles), doubling down on podcasting, or developing creative brand content.
Whatever tack you take, brand building is key to visibility beyond Google.
Case in point: new research from Kevin Indig shows that brand search volume is the biggest predictor for visibility in ChatGPT.
“After matching many metrics with AI Chatbot visibility, I found one factor that stands out more than anything else: Brand search volume. The number of AI Chatbot mentions and brand search volume have a correlation of .334 - pretty good in this field. In other words, the popularity of a brand broadly decides how visible it is in AI Chatbots.”
Make sure you’re measuring your brand awareness—tracking brand search volume, share of voice, and mentions—to understand what you need to do to drive up that visibility.
Ultimately, the more people know your brand, the more places it will naturally appear, and the less dependent you’ll be on any single source of traffic—or any one tech giant and its fickle affections.
Talking of traffic sources, now’s the time to think about diversifying.
Brand mentions in AI predominantly come from third-party sources, according to a study by Funnel.
If we can prize our attention away from the unattainable Google, we’ll have the brain space to focus on other, maybe more rewarding channels, like PR.
Move on from vanity traffic
Rather than gifting Google your traffic and getting… not a lot in return, focus on strategies that will move the needle for you.
Google’s AI Overviews are 99.2% informational, drawing the majority of their responses from top-of-the-funnel articles.
But some of our latest research shows that navigational (-0.06%) and branded queries (-0.34%) are far less susceptible to CTR decline in Google, vs. informational content which has dropped ~10% on average over the last year.
Forget the box-ticking ultimate guides—the Google-demanded preamble that no one wants to write or read.
Focus on optimizing the stuff closer-to-home—your product and service pages, about us pages, brand-specific content—and the journeys new AI audiences will take on your site once they get there.
You don’t have to stop creating informational content, but be strategic about it.
Make sure it answers your users’ questions, builds authority around your brand’s most important topics and use cases, and wins you visibility in AI.
Gatekeep your best content
Last month Google announced that it’s pumping $75bn into its AI cloud capacity.
Aside from multi-million dollar licensing deals with a couple of its favored “big brands” (Reddit and Stack Overflow), Google is doing surprisingly little to incentivize unique content generation.
That’s because, right now, it doesn’t need to—it’s happily using our IP and site resources¹ ².
But someday, when every search result and training input is a rearrangement of the same stale ideas, Google will be pleading publishers with unique content to feed its AI.
If you have incredible content, now’s the time to gatekeep it—hold it back for your customers and your owned audience.
According to Jed White, you can block AI from training on your content, yet still be referenced in its answers.
Some AI crawlers collect training data, while others retrieve real-time content. You may want different policies for each.
That said, blocking AI training crawlers can be a bit of a minefield. In Jed’s words, “overly aggressive bot protection can cut you off entirely.”
And that’s if Google even listens to your directives in the first place.
Jes Scholz (Marketing Consultant at Jes Scholz Consulting) has already spotted Google going rogue—ignoring canonicals, breaking robots.txt, and indexing “noindex” pages—to train its AI.
The other, niggling issue is that Google is quite possibly already training on your AI-blocked content via AI Overviews.
In the experience of Anne Berlin, Lead Product Strategist & Senior Technical SEO at Lumar, newly published content still surfaces in Google’s AI Overviews, even if it has been blocked.
That’s because there is currently no way to opt out of AI Overviews.
“There is no opt-out for AIOs, meaning if you want to get organic traffic from Google, you need to allow it to crawl your site, potentially use your content to train its models and surface it in AI Overviews. Chegg recently filed a lawsuit against Google for this.”
In other words, if you want any traffic from Google, you effectively have to sign your content rights away.
In which case, it’s not enough to just add robots.txt to a page to keep it hidden. You need to remove it from Google altogether.
By all means make your site accessible to crawlers to maintain brand visibility in search and AI, but be selective about the content you let Google access freely.
For every new piece of content you create, start weighing up the benefits of making it openly accessible vs. locking it down. Will you secure enough brand awareness to justify giving your IP away for free? Or will you, like Chegg, be forfeiting your own revenue?
Holding content back will not only keep your site lean (so that AI crawlers can’t run up a huge hosting bill at your expense), it’ll help you carve out a distinct brand and keep your most valuable content genuinely useful to your actual readers.
I think we’ll be seeing a lot more Slack communities, Discords, WhatsApp groups and other gated communities popping up with exclusive, AI-protected content in the very near future.
In the meantime, we need to step up our auditing, be extra vigilant over our IP, and consider holding certain content back for self-preservation.
Final thoughts
Google has changed. It’s gone from being a reliable partner to prioritizing its own interests—from surfacing its own properties, to recycling our content to power its AI.
There is some hope. Google’s AI Mode is reportedly opening up more search real estate, with reports of ~500 word AI overviews containing up to ~60 links, and opportunities for 2nd or 3rd page results to rank.
But, still, our CTR is down and our AI traffic is modest, while Google develops more features to keep users squarely on-platform.
Getting mentioned in Google’s AI is crucial for brand awareness, that much is true, but ultimately there is less incentive for us to create broad, top-of-the-funnel content to appease the algorithm.
Instead we need to take a leaf out of Google’s book, and prioritize our own interests—building our brand beyond search, focusing on the meaningful traffic that engages and converts, and being strategic about what content we freely offer to AI.
The relationship has fundamentally changed. Google is looking out for #1—and now we need to do the same.
