Insights
What happened to Search as we knew it?
You've read (almost) every announcement from Google I/O and seen countless articles on how the changes could affect SEO, marketing, and publishing.
One side is convinced website traffic will become non-existent, while the other insists that, if we follow Google's guidance, everything will be fine. The answer is likely in the middle.
Google's updated AI optimization guide says: "From Google Search's perspective, optimizing for generative AI search is optimizing for the search experience, and thus still SEO."
The reality is murkier than that statement suggests.
Yes, there is significant overlap between SEO and GEO, but we have sufficient evidence to show that AI retrieval systems have a preference for specificity, entity salience, and structural clarity. That probably includes Google AI.
Here are my 6 takeaways after reading and thinking far too much…
1. Google doesn't want anyone to leave
Google's direction was obvious long before ChatGPT arrived.
For years, Google was moving toward an answer engine with features like Knowledge Panels, Featured snippets, and People Also Ask. Post ChatGPT, we got AI Overviews and AI Mode.
These were all steps toward keeping users inside Google’s environment. ChatGPT just accelerated the timeline because it trained millions of people to stop typing keywords and start asking questions naturally.
Google is trying to own both the discovery layer and the interaction layer which means the “intelligent search box” is really an operating system for intent.
Changes to the search box, AI Mode, agents, and conversational search are all part of the same strategy to keep users inside Google longer. And yes, that probably means fewer clicks to many websites.
Google may not care if publishers lose traffic as long as the search environment doesn't completely fall apart.
They still need the open web for training data, fresh information, and authority signals, but they no longer necessarily need to send equivalent traffic back in return.
That's the tension site owners and publishers are reacting to now, but the trend is not new.
2. Google is only one AI surface
Google’s advice only applies to Google Search AI features, not to platforms like ChatGPT, Claude or Perplexity.
These systems use different retrieval systems, training data, and browsing methods.
Brand discovery is now fragmented across the web.
People (and buyers) ask questions in LLMs, on social media, communities, review platforms, and industry forums. This means if you optimize content only for a brand's website you're optimizing for just one piece of modern discovery behavior.
Success now depends on whether AI systems can reliably interpret and represent a brand across the entire web. This seems directionally correct.
I'm not saying traditional search doesn't matter. Google hovers around 90.00% of the search-engine market share, so keep your SEO fundamentals intact!
The point is, you can't be entirely dependent on a single platform you don’t control.
3. You need a broader content distribution strategy
You don't need to be everywhere, and that's where many brands go wrong.
Not every channel matters equally for every brand. The question isn't "are we on LinkedIn?" but "are we contributing something citable on the platforms where our buyers are forming opinions?"
Visibility online is about strategic targeting. It means showing up tactically in the places that influence retrieval systems and buyer perception. For brands, that might mean:
- founders publishing strong opinions on LinkedIn
- customers leaving detailed reviews
- experts participating in Reddit discussions
- appearing on niche podcasts
- publishing original research that gets cited
- building YouTube content around real product use cases
- contributing insights to industry communities
The goal is not to create more content, but to grow a trusted presence in the places AI systems repeatedly source and retrieve from.
4. Digital PR as non-commodity content
You know I strongly believe in Digital PR as "the biggest organic channel of the next decade" (h/t to Rand Fishkin).
Why? Because positive mentions are the best way to attract attention, raise awareness, and build brand credibility online.
Original research picked up by a journalist, a bylined article in an industry publication, a founder quoted in a relevant article are the kinds of third-party citations that carry authority.
They're non-commodity by nature because they require a unique perspective, provable credentials, and/or a remarkable story.
You can't automate your way to genuine earned media, which is precisely why it holds up under Google's quality filters and why LLMs surface it.
5. SEO and marketing are now inseparable for brand building
You don't “SEO optimize” your way into AI visibility if nobody is talking about your brand outside your own domain.
A brand that markets well is doing GEO, whether they call it that or not. A brand that ignores its wider digital footprint is invisible to half the surfaces that matter.
Technical SEO, crawlability and great site architecture still matter, but the brands that consistently surface in AI systems will be those building a presence beyond their websites.
For too long, SEO operated in a silo, but this AI "takeover" forces it back into the broader discipline of marketing.
SEO will be the connective layer between content, Digital PR, brand, social, community, product marketing, and reputation management. In other words, SEO today is increasingly about organized reputation distribution.
6. Don't panic
We're right in the middle of the May Core Update, so don't take any drastic action now.
It's been suggested that this update was deliberately scheduled for the day after Google I/O. That may or may not be true, but it has implications.
If Google is routing more queries through AI Overviews (and as AI Mode sessions grow), the ranking signals that determine which content feeds those surfaces gain importance.
If websites were already losing traffic to AI Overviews, the core update may compound the situation. On the flip side, content that meets Google's guidelines for 'ranking' in traditional search may make it more citable in AI Overviews.
Time will tell.
Related News
1. As Eli Schwartz said on LinkedIn, "Do not give up on SEO, there will always be traffic from the ten blue links, it's less than it used to be, but giving up means you will now have zero."
2. How many of you breathed a sigh of relief when AI Mode didn't become the default in Search?
After TechCrunch reported that “the era of the ten blue links is officially over,” Google pushed back publicly. [That Techcrunch post has since been updated].
Check this from Google News on X:
"You will absolutely continue to see blue web links in search results, and our AI features include prominent links to the web directly within responses. AI Mode is not the default experience in Search. Our new search box helps you describe exactly what you’re looking for, but using it does not mean that you will only get AI features — you'll continue to get a range of results on Search."
3. Good news update: Google is bringing the preferred sources feature to its AI search experiences, specifically AI Mode and AI Overviews. Plus, Google is also adding a perspectives carousel and the highly cited labels to the results. You'll see the screenshots in Barry Schwartz's post.
4. DuckDuckGo surges: Google says users love their AI features, but it's worth comparing that to people's behavior.
Do you follow me on LinkedIn? I share regular tips and stories I don't have room for here. Come and join me.
