Front Door to Search

Google's biggest Search update in 25 years isn't AI Mode. It's the new Intelligent Search box that reshapes the customer journey before users hit enter. We unpack why this is a structural shift for marketing leaders, who's most at risk, and the executive questions every brand should be asking now.

Transcript

SPEAKER_01 0:00

Imagine for a second you need um a new pair of running shoes. Right. And instead of giving you a list of links to read through, Google asks you to upload a quick video of yourself jogging, just so it can analyze your gait, uh, check your pronation, and then custom build a list of options based on your unique stride.

SPEAKER_00 0:19

Which is pretty wild to think about.

SPEAKER_01 0:21

It is. And today we are talking about why the search engine, as you have always known it, just essentially died. And well, what is actually taking its place? Welcome to the deep dive.

SPEAKER_00 0:32

Yeah, the whole premise of what a search engine is supposed to do is undergoing a massive, I mean a truly massive structural shift. Right. We are moving away from this index of websites into an era where the platform itself operates as an active decision layer. Like it guides, it shapes, and it actually executes your entire online journey.

SPEAKER_01 0:49

Aaron Powell And to explore this fundamental shift, we are unpacking the massive announcements from Google's I.O. 2026. We've got the official Google briefings detailing their major search and shopping updates. But we're also balancing that corporate messaging with some incredibly sharp strategic analysis from the COFOMO newsletter.

SPEAKER_00 1:10

Aaron Powell Which is essential reading, honestly.

SPEAKER_01 1:12

Wow, absolutely. For the uninitiated, CO FOMO is curated by Aleda Solis. It goes out to over 45,000 SEO professionals. It is pretty much the gold standard for understanding the mechanics of what search engines are actually doing to the web. So, okay, let's unpack this.

SPEAKER_00 1:29

Aaron Powell We really have to start our analysis precisely where you, the user, start yours. I mean the Google search box.

SPEAKER_01 1:35

Aaron Powell The classic blank box.

SPEAKER_00 1:37

Aaron Ross Powell Exactly. And Google is framing this as the biggest upgrade to the search doorway in over 25 years.

SPEAKER_01 1:42

Aaron Powell You know, when tech companies claim something is the biggest upgrade in a generation, my skepticism usually flares up big time.

SPEAKER_00 1:48

Aaron Powell Oh, same. It usually just means like they added a drop shadow to a button or something.

SPEAKER_01 1:53

Right. But looking at the documentation here, this is an entirely different beast. The box itself is fundamentally changed. It dynamically expands and it accepts uh multimodal inputs.

SPEAKER_00 2:04

Yeah, that's the key.

SPEAKER_01 2:05

You aren't just typing text anymore. You can drop in a PDF, upload an image, drag in a video clip, and even feed it your currently open Chrome tabs.

SPEAKER_00 2:15

It just absorbs context from whatever medium you throw at it. And perhaps the most crucial change is that the AI actively suggests better ways to formulate your question before you even hit the enter key. Wow. It anticipates your intent based on that whole blend of materials you've provided. As the Light of Solis noted in the newsletter, they didn't just flip search into some AI mode, they completely rebuilt the search doorway.

SPEAKER_01 2:41

So it's kind of like it's like going to a restaurant where instead of handing you a laminated menu to pick from, the chef comes out to your table.

SPEAKER_00 2:48

Okay, I like this.

SPEAKER_01 2:49

He asks how you're feeling, looks at the friends you're eating with, notes the weather outside, and just starts co-designing the meal with you on the spot. You aren't picking from pre-existing options. The options are being generated because of you.

SPEAKER_00 2:59

What's fascinating here is the implication for digital behavior. Because for two decades, digital marketing and user journeys were entirely about what happens after the query.

SPEAKER_01 3:10

Right. You search for a thing, then the journey starts.

SPEAKER_00 3:13

Exactly. You type a word, hit enter, and millions of businesses fight in an option over the results page.

SPEAKER_01 3:18

Yeah.

SPEAKER_00 3:19

But now, the machine is shaping the query itself.

SPEAKER_01 3:22

It's getting ahead of you.

SPEAKER_00 3:23

Yes. It is guiding you to ask longer, more nuanced questions, adding context you might not have even thought of. The old model of isolated keyword search behavior is just over.

SPEAKER_01 3:36

But if the chef is co-designing the meal with me, how does the system actually remember what we are building? Because historically, every single time you hit search, you're essentially starting from a blank slate.

SPEAKER_00 3:47

You mean that constant back and forth?

SPEAKER_01 3:49

Yeah. You search, look at 10 blue links, hit back, tweak the words, and search again. It's very disjointed. Trevor Burrus, Jr.

SPEAKER_00 3:54

Right. Well, that disjointed experience is totally replaced by context preservation.

SPEAKER_01 3:58

Okay.

SPEAKER_00 3:59

You can now move seamlessly from an initial AI overview straight into a conversational AI mode follow-up. And you do it without losing the thread of the interaction at all. It's an ongoing, guided environment.

SPEAKER_01 4:11

Aaron Powell Meaning if I start by dropping in a picture of my living room and asking for rug ideas, and then like three questions deep into the chat, I randomly ask, hey, what if I have a golden retriever? The engine remembers the photo of the room, remembers the mid-century modern style we just agreed on, and instantly filters the options for pet-friendly durability.

SPEAKER_00 4:30

Aaron Powell Precisely. That is the exact mechanism at play. And this actually brings up a critical takeaway from the COFOMO newsletter regarding this continuous conversation.

SPEAKER_01 4:40

What's that?

SPEAKER_00 4:40

The analysis highlights something called the framing effect. In any continuous dialogue, whoever sets the initial frame holds a disproportionate amount of power.

SPEAKER_01 4:49

Aaron Powell Oh, because they establish the baseline assumptions for the rest of the chat.

SPEAKER_00 4:53

Exactly.

SPEAKER_01 4:54

But wait, if Google is the one suggesting the questions, framing the initial comparisons, and keeping the whole interaction inside their own conversational interface, isn't that terrifying for brands?

SPEAKER_00 5:05

Oh, it's a nightmare for them.

SPEAKER_01 5:06

As the newsletter points out, if search used to be a doorway to the web, it sounds like it's now just a doorway to Google.

SPEAKER_00 5:12

That pushback is the exact existential dread rippling through marketing leadership right now.

SPEAKER_01 5:18

Yeah.

SPEAKER_00 5:19

The consumer funnel is getting shorter and messier simultaneously. Right. A brand might enter your journey already filtered through Google's assumptions about its product attributes or, you know, how it compares to competitors.

SPEAKER_01 5:31

So a company could completely win over a buyer without ever earning a single click to their actual website. Yeah. Simply because Google's AI read their data and presented it perfectly inside the chat.

SPEAKER_00 5:43

Yes, exactly.

SPEAKER_01 5:44

Or conversely, they could lose the sale before the customer even knows they exist because the AI filtered them out early on.

SPEAKER_00 5:51

That's the reality. Yeah. The user is kept in this continuous contextual bubble. Yeah. And to make sure you never actually need to leave that bubble to get things done, Google is deploying virtual agents to do the heavy lifting right inside the search results.

SPEAKER_01 6:04

Okay, let's talk about these agents.

SPEAKER_00 6:05

They're powered by Gemini 3.5 Flash, which is now the default model globally.

SPEAKER_01 6:10

Yeah, the documentation breaks these down into a few categories, starting with information agents. The examples they gave really highlight how these function like tireless interns operating 247 in the background. Say you're apartment hunting.

SPEAKER_00 6:23

A traditionally miserable experience.

SPEAKER_01 6:25

Truly. But instead of checking a real estate site every morning, you just dump your exact hyperspecific requirements into the search box. Your max budget, the need for south-facing windows, in-unit laundry, and being within a 10-minute walk of a specific subway line.

SPEAKER_00 6:41

And instead of just returning a list of current apartments and stopping there, the information agent continuously scans the web day after day. It just keeps running. Right. It synthesizes updates from various property management sites, local blogs, standard listings, and notifies you the instant a match hits the market.

SPEAKER_01 6:59

Or for the sneaker heads out there, you tell the agent to monitor the entire internet for any whispers of a new sneaker collaboration dropping from your favorite pro athlete.

SPEAKER_00 7:07

Oh, yeah, that was a great example.

SPEAKER_01 7:09

It finds the information, but it also synthesizes rumors from Reddit, news sites, and social posts into one single briefing.

SPEAKER_00 7:16

But you know, information retrieval is one thing. The architecture takes a massive leap forward with agetic booking.

SPEAKER_01 7:22

Yes, this is where it gets crazy.

SPEAKER_00 7:23

Here, the AI transitions from retrieving data to taking action on your behalf in the physical world. Trevor Burrus, Jr.

SPEAKER_01 7:29

The karaoke example they used at IO is wild. You give the AI specific criteria, like you need a private karaoke room for six people this coming Friday night, and it must serve food past midnight. Right. The agent pulls real-time pricing, checks availability across multiple venues, and for select categories like home repair, beauty salons, or pet care, it actually makes phone calls to the businesses using a synthesized voice to secure the reservation.

SPEAKER_00 7:57

Aaron Powell The search engine acts as your proxy. It literally negotiates the logistics so you don't have to.

SPEAKER_01 8:01

Making phone calls is impressive, for sure. But we have to talk about Google anti-gravity.

SPEAKER_00 8:06

Yes, we do.

SPEAKER_01 8:07

Because this is where the concept of a search engine bends into something entirely new: agentic coding. Search is now building custom generative UI on the fly.

SPEAKER_00 8:16

Aaron Powell It generates utility rather than just text.

SPEAKER_01 8:19

Right. To explain what that means, say you ask a complex question about the orbital mechanics of the solar system.

SPEAKER_00 8:25

Okay.

SPEAKER_01 8:26

Instead of giving you a Wikipedia summary, search codes a fully interactive 3D visual simulation for you to play with right there on the results page.

SPEAKER_00 8:34

It's incredible.

SPEAKER_01 8:35

Or the fitness tracker example. You want to start running. So search doesn't just give you a PDF schedule, it codes a bespoke mini-app just for you. It embeds live maps of your neighborhood, pulls in the local weather forecast for the week, and adjusts your running route daily based on whether it's raining.

SPEAKER_00 8:53

We are basically watching the underlying architecture of the web shift from human-readable websites to machine readable data.

SPEAKER_01 9:00

Aaron Powell Say more about that.

SPEAKER_00 9:01

Well, if the AI is generating the interface, the actual app you interact with, the original website only exists to feed raw data to the agent.

SPEAKER_01 9:10

Aaron Powell Okay, here's where it gets really interesting. If the agent is doing all the research, coding my custom fitness tracker and calling the karaoke bar to book my room, how do I actually pay for anything?

SPEAKER_00 9:21

Right. The money question.

SPEAKER_01 9:22

If I never see the checkout screen of these individual businesses, does the whole e-commerce pipeline completely break?

SPEAKER_00 9:29

Aaron Powell That logical friction is exactly why Google is rolling out the Universal Cart.

SPEAKER_01 9:34

Uh-huh.

SPEAKER_00 9:35

Okay. This is built on top of what they call the shopping graph, which is this actively updated catalog of over 60 billion product listings.

SPEAKER_01 9:43

Aaron Powell 60 billion listings. That is the training data for the agent to know what actually exists in the world.

SPEAKER_00 9:50

Yes. And the universal cart sits across almost the entire Google ecosystem. You can access it in search inside the Gemini app while watching YouTube and even directly inside Gmail.

SPEAKER_01 10:01

Aaron Powell So the practical application is I could be watching a YouTube review of a new digital camera and add it to my cart straight from the video player. Exactly. Then later I'm checking my Gmail. I see a promotional newsletter for a compatible camera lens. I add that to the same cart, and it's all synchronized in one place.

SPEAKER_00 10:17

Aaron Powell Yes, but the crucial mechanism here is that it actively reasons in the background.

SPEAKER_01 10:23

What do you mean by reasons?

SPEAKER_00 10:24

It is not a dumb digital basket. The official announcement highlighted this scenario where a user is building a custom PC. Right. Which is a notoriously tricky process with a lot of varying specifications.

SPEAKER_01 10:39

Oh yeah, lots of parts.

SPEAKER_00 10:41

You add a motherboard from one retailer, say New Egg, and a processor from Best Buy. The cart will proactively flag if those two parts have incompatible socket types. And it will actually suggest alternative components that fit together before you ever click buy.

SPEAKER_01 10:57

Oh wow. It's acting like a tech support specialist anticipating your mistakes.

SPEAKER_00 11:01

Totally.

SPEAKER_01 11:01

And the documentation mentioned it integrates deeply with Google Wallet, too, meaning it automatically surfaces your hidden loyalty points or merchant-specific coupon codes without you having to hunt the web for a promo code.

SPEAKER_00 11:14

But underneath all of this frictionless consumer magic, there is some incredibly heavy plumbing required.

SPEAKER_01 11:19

Right.

SPEAKER_00 11:20

You have to make transactions work seamlessly across thousands of different retailers without breaking their individual systems.

SPEAKER_01 11:25

Let's pop the hood on that because getting Walmart, Sephora, and a random Shopify store to all play nice in one Google cart sounds impossible.

SPEAKER_00 11:33

It's complex, definitely. The architecture relies on two major new protocols. First is UCP, the Universal Commerce Protocol.

SPEAKER_01 11:41

Okay.

SPEAKER_00 11:42

This is the bridge that allows you to check out using Google Pay across major brands like Nike, Target, Wayfair, and even Shopify merchants like Fenty or Steve Madden.

SPEAKER_01 11:51

Aaron Powell So no matter which of those stores I'm buying from, I completely bypass their clunky individual checkout processes. Exactly. The brand stays the merchant of record. They still get the sale, but Google controls the friction or the lack thereof in the actual transaction.

SPEAKER_00 12:09

Aaron Powell That's UCP. Now the second piece of technology is AP2 or the agent payments protocol. AP2. Right. This is strictly designed to scale secure agentic purchases. Because when you have an AI acting on your behalf of the background, making decisions and buying things, you need an ironclad system of trust.

SPEAKER_01 12:25

Aaron Powell For sure. Because if an AI is buying my groceries or securing karaoke rooms, I need a rock solid guarantee it's not going to glitch and drain my checking account on 400 pairs of limited edition sneakers.

SPEAKER_00 12:36

Exactly. The mechanism AP2 uses to prevent that involves a tamper-proof digital mandate.

SPEAKER_01 12:41

How does that work?

SPEAKER_00 12:42

You set strict guardrails for the agent. You specify the exact brands, the exact product categories, and a hard cap on how much it's allowed to spend. Under the hood, the protocol generates a single-use digital token.

SPEAKER_01 12:55

So this AP2 technology, it's basically like giving a corporate credit card to a hypercompetent personal assistant. But you've cryptographically hard-coded exactly how much they can spend.

SPEAKER_00 13:07

That analogy perfectly captures the mechanism, yes.

SPEAKER_01 13:10

Once they hit that $50 limit at that specific merchant, the token just self-destructs. They literally cannot overcharge you, and they are forced to leave a permanent digital paper trail of receipts.

SPEAKER_00 13:22

Spot on.

SPEAKER_01 13:29

Okay.

SPEAKER_00 13:30

So if you ever need to return an item, you and the merchant are looking at the exact same indisputable cryptographic record of what the agent actually did.

SPEAKER_01 13:37

Aaron Powell We've talked a lot about how amazing this looks for you as a consumer. Having a cart that does your tech support and an AI intern managing your receipts sounds like utopia.

SPEAKER_00 13:46

It does.

SPEAKER_01 13:46

But if you are one of our listeners who might be running a small business or manages a brand or just relies on search traffic to keep your company alive, this should honestly make you sweat.

SPEAKER_00 13:59

Aaron Powell It really should. The CFOMO newsletter frames this as a massive wake-up call, explicitly warning that certain types of businesses are incredibly vulnerable to being wiped out by this new architecture.

SPEAKER_01 14:10

Okay, who is on the chopping block, according to the analysis?

SPEAKER_00 14:13

Aaron Powell High consideration B2B brands are basically at the top of the list.

SPEAKER_01 14:17

Aaron Powell Wait, if high consideration B2B brands are vulnerable, wouldn't that also spell disaster for generic low-effort e-commerce sites? You know, the ones that just drop ship products and scrape Amazon descriptions?

SPEAKER_00 14:29

Oh, absolutely. Both ends of the spectrum are uniquely vulnerable, but for entirely different mechanical reasons.

SPEAKER_01 14:35

Break that down for me.

SPEAKER_00 14:36

Okay, for the B2B brands, high consideration means a buyer typically does weeks of research, right? Comparing complex software features, that kind of thing. Exactly. But in the new model, the AI agent reads all those white papers in a millisecond, synthesizes the pros and cons, and gives the answer directly to the buyer inside the search interface.

SPEAKER_01 15:04

Wow.

SPEAKER_00 15:05

The human buyer never visits the B2B website.

SPEAKER_01 15:07

Aaron Powell They lose their chance to capture the lead because the AI stripped away the need to even visit the site. That is brutal. And what about the e-commerce side? Why are they vulnerable?

SPEAKER_00 15:18

Aaron Powell The e-commerce vulnerability comes down to thin product data.

SPEAKER_01 15:22

Aaron Powell Thin data.

SPEAKER_00 15:23

Yeah. If you sell a couch and your website just has a picture, a price, and the title Blue Couch, the Universal Cartz reasoning engine will simply ignore you.

SPEAKER_01 15:32

Oh, I see.

SPEAKER_00 15:32

It will bypass you entirely in favor of a competitor who provided rich, structured data. Things like the fabric type, exact dimensions, weight limits, fire safety ratings. The machine just cannot reason with thin data.

SPEAKER_01 15:44

Aaron Powell That makes total sense.

SPEAKER_00 15:45

Aaron Powell And finally, the the analysis warns that agencies still selling traditional SEO, you know, optimizing purely for keywords and ten blue links. They're optimizing for a version of the internet that is evaporating.

SPEAKER_01 15:57

Aaron Powell Because the old model rewarded brands that captured demand after it was expressed. A user types a keyword, you catch them with a good link. Right. But the new model rewards brands that are legible to the AI systems that are shaping the demand in real time.

SPEAKER_00 16:11

Aaron Powell This is driving a massive shift in leadership priorities from SEO to GEO. Trevor Burrus, Jr.

SPEAKER_01 16:16

Generative engine optimization.

SPEAKER_00 16:18

Exactly. You aren't optimizing for a single keyword string anymore. You have to optimize your digital presence for scenarios, for product attributes, and for how an AI compares you against your competitors.

SPEAKER_01 16:30

So what does this all mean for the listener who might be running a business? The newsletter asks a really tough, fundamental executive question. If Google controls the initial question, controls the conversational journey, and controls the transaction path with the universal cart, what exactly is your brand's independent relationship with demand?

SPEAKER_00 16:48

Aaron Powell It really is the defining business question of this new era.

SPEAKER_01 16:52

Yeah.

SPEAKER_00 16:52

Because in a world of infinite synthesized information and agentic search, a brand's competitive moat can no longer simply be its search engine rankings.

SPEAKER_01 17:00

Right, that's gone.

SPEAKER_00 17:01

Rankings are incredibly fragile when the user interface can change on a whim to some anti-gravity generative UI.

SPEAKER_01 17:09

If the moat isn't search ranking, how do you defend your business?

SPEAKER_00 17:13

Well, the analysis points to three critical pillars. First is pure brand strength. People have to explicitly desire you, asking the agent for your specific company rather than just asking for a generic solution.

SPEAKER_01 17:24

Okay. Brand strength. What's number two?

SPEAKER_00 17:27

Second is pristine data quality. In the industry, we call this entity clarity.

SPEAKER_01 17:32

Aaron Powell Meaning the blue couch example, providing such comprehensive, perfectly formatted data about the fabric and dimensions that the machine has absolute clarity on what you actually sell.

SPEAKER_00 17:43

Aaron Powell Right. Complete attribute coverage. So the AI knows exactly who your product is for and why it is superior.

SPEAKER_01 17:49

Got it. And the third pillar.

SPEAKER_00 17:51

Direct audience access. Brands must cultivate channels they actually own, like robust email lists or dedicated apps where they are not reliant on a third-party AI mediating their discovery and dictating their relationship with their own customers.

SPEAKER_01 18:03

Aaron Powell Well, we have really gone on a journey today.

SPEAKER_00 18:06

Aaron Powell That's an understatement.

SPEAKER_01 18:07

Aaron Powell We started by looking at how the 25-year-old paradigm of the blank search box and the 10 blue links is being entirely replaced by a dynamic, multimodal decision layer. We explored the mechanics of 247 background information agents, how anti-gravity creates custom software on the fly, and how the Universal Cart uses protocols like UCP and AP2 to quietly solve our shopping problems while bypassing traditional checkouts.

SPEAKER_00 18:34

Aaron Powell has a lot to take in.

SPEAKER_01 18:36

It is. It is fundamentally rewriting the rules of the Internet for both everyday users and the businesses trying to reach them.

SPEAKER_00 18:42

It is a profound architectural shift in how humanity interacts with information. But you know, as as we wrap up, there's one specific lingering detail buried in the announcements that we really should leave the listener to process.

SPEAKER_01 18:54

Aaron Powell Oh, let's hear it.

SPEAKER_00 18:55

Aaron Powell We spent a lot of time today talking about agents acting on your behalf, right? And AI shopping for your PC parts, booking your karaoke night, and continuously tracking your fitness. Yeah. But remember that Google's personal intelligence is now securely connecting to your Gmail, your Google photos, and soon your calendar rolling out to nearly 200 countries. So if an AI agent has real-time, continuous access to your private emails, your stored memories, your daily schedule, and your universal wallet, at what point does the search engine stop answering your questions and start anticipating your life choices before you even realize you need to make them?

SPEAKER_01 19:31

That is a terrifying and fascinating place to leave it. Thank you for joining us on this deep dive. Keep questioning the information ecosystem around you, and we'll catch you next time.