The Invisible Funnel: Why Google’s Agentic Search Makes Competitor Ad Spying More Critical, Not Less

Think about what this means for the marketer who built their entire measurement strategy around observable behavior. Your attribution models depend on clicks you’ll never see. Your content strategy assumes a research phase that now happens inside a black box. Your competitive analysis relies on watching where users go — but increasingly, they don’t go anywhere. They stay inside Google’s AI interface, receive a curated recommendation shaped by signals you can’t inspect, and transact without ever visiting your site or your competitor’s.

This is the “dark funnel” — and it’s no longer a buzzword tossed around at conferences. It’s the default experience for over a billion monthly AI Mode users, a figure confirmed at Marketing Live alongside the announcement of new ad formats specifically engineered for this collapsed journey.

So what do you do when you can no longer watch the journey unfold? You obsess over the artifacts it leaves behind. The ads that surface inside AI Mode. The recommendations that appear in Highlighted Answers. The Direct Offers that Gemini bundles and serves at the moment of purchase intent. These outputs are the only visible residue of a decision-making process that now happens almost entirely out of sight. And if your competitors’ ads are showing up in those moments and yours aren’t, you won’t even know you lost — because there’s no click-trail to tell you the battle happened at all.

The Darwinian Ad Slot — Why AI Filtering Creates a Survival-of-the-Fittest for Creatives

If you could buy your way to the top of a Google results page with nothing but a fat budget and a mediocre ad, congratulations — you lived through the golden age of lazy paid search. That era is now facing extinction pressure. The new ad formats Google unveiled at Marketing Live 2026 don’t just occupy different screen real estate; they operate under entirely different rules of survival, rules where contextual fitness matters more than financial brute force.

Consider what’s actually happening inside AI Mode. Google introduced two Gemini-powered ad formats — Conversational Discovery and Highlighted Answers — that fundamentally rewrite the relationship between advertiser and algorithm. As WordStream detailed in its breakdown, Conversational Discovery ads appear as a natural part of the AI Mode response, written to answer the user’s query rather than interrupt it. Gemini doesn’t just match the ad to the query — it matches the ad to both the query and the AI-generated response surrounding it, so the creative feels native to a conversation already in progress. Ask for fragrances to make your house smell like a fancy spa, and Gemini builds ad copy that directly addresses that desire while surfacing the most relevant product features. The ad isn’t a banner shoved into a sidebar. It’s an answer wearing a sponsor badge.

Highlighted Answers take this a step further. When someone searches for recommendations — “best apps for learning Spanish before a trip” — Google now serves a curated list of suggestions where your ad must earn its spot among organic options, rather than sitting in a quarantined ad block above or below the results. You’re not bidding for a billboard; you’re auditioning for a place at a table set by an AI editor.

Then there are the formats designed for deeper engagement. AdExchanger reported on AI-generated shopping ad units tailored for high-consideration purchases like consumer tech or household appliances, complete with AI explainers that elaborate on why a product fits a specific user’s needs — and which users can prompt for more information without leaving the page. And Google’s Jerry Taylor singled out Business Agents for Leads as his personal favorite: a format that embeds Gemini directly into the ad unit so users can interact conversationally within the ad itself. The throughline across all of these formats is unmistakable: Gemini is acting as an editorial filter, not merely an auction engine.

This is the Darwinian shift. In the old model, a mediocre ad with a high enough bid could muscle its way onto page one. In AI Mode, the algorithm evaluates whether your creative genuinely belongs in the conversation — whether it answers the specific question a user asked in the specific context Gemini has constructed. An ad that doesn’t contextually fit simply doesn’t appear, regardless of your budget. The auction still matters, but it’s now the second gate. The first gate is relevance scoring by the most sophisticated content-matching system Google has ever deployed.

That dual filter — economic and editorial — transforms the competitive intelligence value of every ad that survives it. A traditional search ad told you what a competitor was willing to pay. An AI Mode ad tells you what a competitor is willing to pay and what Google’s most advanced language model has judged to be a contextually perfect answer to a real user’s intent. These are no longer just ads. They’re algorithmically validated hypotheses about what converts — and that makes them exponentially richer intelligence signals than anything the old search results page ever offered.

The New Intelligence Gold Mine — What Surviving Ads Actually Reveal

Competitor ad spying used to be a straightforward exercise: screenshot the headline, note the call-to-action, log the display URL, and move on. You were cataloging surface-level creative choices — a word swap here, a price point there. In the agentic era, the intelligence surface area has exploded so dramatically that marketers who still limit their analysis to copy and CTAs are leaving the most valuable data on the table.

Start with offer architecture. Google’s new Direct Offers beta allows merchants and brands to upload discounts, local coupons, and other incentives that Google can then match or combine on the fly to present the most compelling offer for a specific AI search user. That last phrase is critical: most compelling. Google’s AI is no longer passively displaying whatever promotion an advertiser uploaded. It’s actively evaluating, stacking, and assembling offer components based on the user’s intent, context, and likely purchase threshold. When you spot a competitor’s ad surfacing in AI Mode with a bundled 10% discount plus a free accessory — the grill manufacturer scenario Google’s VP Jerry Taylor highlighted at his press briefing — you’re not just seeing a promotion. You’re seeing the offer combination that Gemini’s algorithms determined was necessary to close the sale for that query class. That’s a window into dynamic incentive stacking strategy that no competitor research tool has ever offered before.

Layer on the loyalty dimension. Retailers integrated with Google’s Universal Checkout Platform can now export their loyalty points or exclusive member discounts directly into agentic ads. If a competitor’s ad is surfacing with loyalty-point incentives attached, that tells you they’ve built a UCP integration deep enough to let Google weaponize their retention program as an acquisition tool. That’s strategic intelligence about their commerce infrastructure, not just their marketing messaging.

Then there’s the trust and authority layer embedded in curated recommendation lists. When someone asks AI Mode for the “best apps for learning Spanish before a trip,” Google now serves what are effectively Highlighted Answers — curated lists where ads appear as recommended options rather than in a separate slot above or below the organic results. If your competitor earned a spot on that list, they passed a relevance and credibility threshold that Google’s AI deemed worthy of a genuine recommendation. Reverse-engineering which brands consistently appear in these curated lists across multiple query variations reveals the trust signals — review velocity, structured data quality, merchant rating thresholds — that earn placement.

Finally, there’s an entirely new category of intelligence: conversational UX patterns inside interactive ad agents. Google’s Business Agents for Leads embed Gemini AI directly into the ad unit, allowing users to prompt and interact without ever leaving the page. Studying how a competitor’s agent handles objections, qualifies leads, or redirects ambiguous prompts gives you a playbook for conversational commerce design that didn’t exist six months ago.

Add it all up and the competitive intelligence dataset available from a single surviving ad in AI Mode — offer structure, incentive stacking logic, loyalty integration depth, trust signals for curated placement, and conversational agent behavior — dwarfs anything a traditional SERP screenshot could provide. The irony is sharp: even as the funnel becomes invisible to most marketers, the ads that break through its filters are radiating more strategic information than ever. The only question is whether you’re equipped to read the signal.

The “Brand as AI Trust Signal” Multiplier

Here’s the layer most ad spy practitioners miss entirely: in an agentic environment, the creative is only half the story. You can reverse-engineer every headline, dissect every offer structure, and catalog every call-to-action your competitors deploy — and still lose, because Google’s AI isn’t just evaluating what the ad says. It’s evaluating the brand behind it.

Neil Patel framed this shift with a question that should unsettle every performance marketer still operating on purely tactical playbooks: does the AI trust your brand? That question isn’t rhetorical. It’s architectural. When AI systems collapse the traditional funnel — from “Search → Website → Research → Cart → Purchase” down to “Ask AI → Receive recommendation → Buy” — the mechanism that determines which brands surface in that compressed journey isn’t ad spend or keyword bidding alone. It’s trust, modeled computationally from signals that most advertisers have never thought to spy on.

Consider what this means for competitive intelligence. The traditional ad spy toolkit captures the visible layer: copy, landing page, offer, extensions. But in an agentic search environment, there’s an invisible layer underneath that determines whether a brand even gets the opportunity to appear. That layer is built from citation frequency across the web, review sentiment and volume, content authority, branded search demand, and the consistency of a brand’s informational footprint across platforms. As Patel argues, strong brands are cited more often, generate more searches, and earn more mentions, reviews, and links — and these are precisely the signals AI systems are trying to model when they decide which recommendations to surface.

This creates a two-dimensional competitive intelligence problem. The first dimension is tactical: what does the ad say, what does it offer, and how does it present itself? The second dimension is structural: what brand signals allow this advertiser to surface at all? Marketers who only operate on the first dimension are copying outputs without understanding inputs. They’re replicating the visible symptom while ignoring the underlying condition that made the symptom possible.

The structural dimension demands new tools entirely. Tracking how competitors appear across AI-generated responses — their citation patterns, mention velocity, and visibility trends — requires capabilities that go far beyond traditional SERP monitoring. This is the kind of intelligence gap that platforms like Semrush’s AI Visibility Toolkit are designed to address, offering marketers the ability to track how brands are referenced and recommended across AI surfaces rather than just conventional search results.

The strategic implications are profound. When Google positions Gemini as a core intelligence layer across Search, shopping, YouTube, and beyond, it means brand trust signals propagate across every surface simultaneously. A weak brand doesn’t just underperform in one channel — it becomes systematically invisible across an entire AI-mediated ecosystem. Conversely, a strong brand compounds its advantage everywhere at once, because the same trust architecture that earns an AI citation in one context reinforces discoverability in every other.

This is why the marketers who combine ad creative intelligence with AI visibility tracking — understanding not just what competitors are saying but why certain brands consistently break through — will hold an asymmetric advantage. Those who only copy headlines will keep losing to those who reverse-engineer the full trust architecture. The creative is the tip of the iceberg. The brand is everything underneath the waterline. And in agentic search, the waterline is rising fast.

The Platform War Makes Intelligence Even More Urgent

Google isn’t building its agentic advertising infrastructure in a vacuum. It’s racing against a coalition that wants to define the rules of AI-powered commerce before Mountain View can lock them in. Understanding this platform war isn’t optional context — it’s the reason competitive intelligence has shifted from a tactical advantage to a strategic imperative.

The clearest fault line emerged at Google Marketing Live 2026, where Google aggressively expanded its Universal Checkout Platform across every surface it controls. As AdExchanger reported, the UCP checkout integration will now appear within YouTube ads, Google Maps listings, and other Google properties — meaning a viewer can complete a hotel booking or food delivery order without ever leaving the conversation. Google is also sweetening the deal for merchants by letting brands that integrate with UCP export loyalty points and exclusive discounts directly into agentic ads. The goal is unmistakable: collapse the entire purchase journey into Google’s ecosystem and make leaving it feel like friction.

But Google isn’t the only player writing this playbook. The same AdExchanger piece noted that Google’s UCP is competing against the Agentic Commerce Protocol, backed by OpenAI, PayPal, Stripe, and others. That rival protocol represents a fundamentally different vision — one where agentic transactions aren’t tethered to any single search engine’s ad auction. If the Agentic Commerce Protocol gains merchant adoption, it could create an entirely separate surface where AI agents negotiate, compare, and purchase on behalf of consumers without ever touching a Google ad unit.