Neural TL;DR, Powered by Claude
AI Summary
The shift from Attention Economy to Intention Economy isn't a prediction—it's underway, with McKinsey projecting $3-5 trillion in global agentic commerce by 2030. Zero-click searches have surged to 69%, AI agent traffic to retail sites grew 4,700% year-over-year, and protocol standards are consolidating in real time. Most businesses are optimizing for a world that's disappearing while the infrastructure for what's next remains unbuilt. Neural Partners has spent the last eight months engineering for this transition—not theorizing about it. The question isn't whether agentic commerce is coming. The question is who's ready.
In Part 1, we laid out the uncomfortable truth: the marketing funnel is becoming obsolete as consumers discover they have options beyond traditional search. In Part 2, we traced the cascading economic impact—from Google to advertisers to publishers to the entire programmatic ecosystem.
Now for the part we've been building toward: What does the other side look like, and who's actually ready for it?
The Divergence Is Here
For two decades, the dominant paradigm has been the Attention Economy: platforms competing for human eyeballs, monetized through impressions and clicks. This model generated $1.17 trillion in global advertising spend in 2025 (WARC) and built trillion-dollar businesses on aggregating human attention.
We're now watching a transition to something fundamentally different—what we call the Intention Economy. Not an evolution. A structural reorganization of how value flows.
The unit of value shifts from "clicks" to "executions." Consumers delegate discovery, comparison, and transaction to AI agents. These agents don't browse—they query structured data, negotiate via API, and execute based on protocol rather than persuasion.
The trillion-dollar marketing industry spent years solving the wrong problem. While obsessing over cookie deprecation, the consumer itself was transforming. The real disruption isn't how we target humans. It's that humans are no longer the only audience.
The Numbers That Should Keep You Up at Night
This isn't speculation. The data is already in:
| Metric | Value | Source |
|---|---|---|
| Zero-click searches (news queries) | 56% → 69% since AI Overviews launched | Similarweb, May 2025 |
| AI Overviews presence in search | 6.49% → 13.14% of queries (Jan-Mar 2025) | Semrush |
| CTR drop when AI Overviews present | ~40% decline | Industry analysis |
| GenAI traffic to U.S. retail sites (YoY) | +4,700% | Adobe Digital Insights |
| AI agent traffic growth (Jan-Aug 2025) | +1,300% | HUMAN Security |
| Product-related AI agent interactions | 87% | HUMAN Security |
| U.S. organic traffic lost in 12 months | 600M visits | Similarweb |
These agents aren't casually browsing. 87% of AI agent page interactions are product-related. They're shopping, not surfing.
The Scale of What's Coming
McKinsey's November 2025 research describes agentic commerce as a "seismic shift"—and their projections explain why:
| Projection | Value | Source |
|---|---|---|
| U.S. B2C retail orchestrated by AI agents (2030) | $900B - $1 trillion | McKinsey |
| Global agentic commerce (2030) | $3 - $5 trillion | McKinsey |
| Organizations currently using AI | 88% | McKinsey Global Survey |
| Organizations experimenting with AI agents | 62% | McKinsey Global Survey |
| Consumers trusting GenAI search over traditional ads | 41% | BoF-McKinsey |
| Higher satisfaction with AI-assisted shopping | 85% | BoF-McKinsey |
As McKinsey put it: "This isn't just an evolution of e-commerce. It's a rethinking of shopping itself."
But here's what caught our attention: 62% of organizations are experimenting with AI agents, yet fewer than 10% are scaling them in any given business function. The gap between awareness and execution is massive.
That gap is where we live.
The Protocol Landscape: Rails Being Laid in Real Time
While most marketers debate attribution models, the technical infrastructure for agent-mediated commerce is consolidating fast:
Model Context Protocol (MCP) — Anthropic's standard for AI agents to share context across models and tools. This is how agents remember, reason, and maintain objectives across environments. It's also the foundation of our Neural Core technology infrastructure.
Agentic Commerce Protocol (ACP) — Launched September 29, 2025 by OpenAI and Stripe. Enables transactions from Etsy and Shopify merchants directly through ChatGPT conversations. Salesforce, Checkout.com, and Shopify announced support within weeks.
Agent-to-Agent Protocol (A2A) — Google's protocol for autonomous agents to coordinate and negotiate. Moved to Linux Foundation governance in June 2025. 50+ technology partners including Atlassian, Salesforce, SAP, ServiceNow.
Payment Network Protocols — October 2025: Visa and Mastercard both launched agentic payment frameworks. Visa is already piloting transactions where AI agents spend on behalf of users within preset budgets—partnering with Anthropic, Microsoft, OpenAI, Perplexity, and Stripe.
Six months ago this was fragmented experimentation. Now it's infrastructure.
The companies building on these protocols today will own the rails of the Intention Economy. Those waiting for clarity will find themselves invisible to the agents that increasingly mediate commerce.
What Cloudflare's CEO Is Losing Sleep Over
In December 2025, Matthew Prince—CEO of Cloudflare, which sits in front of 20-25% of the web—told WIRED:
"The business model of the internet is about to change dramatically. I don't know what it's going to change to, but it's what I'm spending almost every waking hour thinking about."
The context: Cloudflare has blocked 416 billion AI bot requests since July 1, 2025. Nearly 3 billion per day.
The agent layer isn't theoretical. It's already generating traffic at scale that dwarfs human browsing.
We're seeing it ourselves. Here's a 24-hour snapshot from our own domain (December 12, 2025):
Human traffic to neuralpartners.ai: 39 visits in 24 hours (Source: Cloudflare Web Analytics)
AI agent traffic to neuralpartners.ai: 292 requests in the same 24-hour period (Source: Cloudflare AI Crawl Control)
292 AI requests vs. 39 human visits. On our own site, agent traffic outnumbers human traffic by nearly 7.5x. This isn't a forecast—it's Tuesday.
The Infrastructure Void
Here's the problem nobody's solving: Current commerce infrastructure was built for human eyes.
HTML. Visual merchandising. Persuasive design. Emotional storytelling. Brand narratives. All of it assumes a human on the other side—someone who can be impressed, persuaded, delighted.
AI agents don't get impressed. They don't feel delight. They have evaluation functions.
When an agent queries your business, can it:
- Find your products through structured data?
- Access real-time inventory and pricing via API?
- Evaluate your specifications against competitors?
- Trust your data enough to recommend you?
For most businesses, the honest answer is no. They're invisible to agents—and increasingly invisible to the humans who delegate to them.
This is the void. And it's widening every day.
The Trust Problem Nobody's Talking About
Traditional marketing assumes human psychology as its canvas. Brand building, emotional storytelling, social proof—all designed for human cognitive biases.
AI agents don't have cognitive biases. They have evaluation functions.
We've been studying how LLMs evaluate brands, products, and services. The results are telling:
- Brand awareness means nothing to an algorithm comparing specifications
- Emotional advertising can't influence something without emotions
- Social proof becomes noise when agents detect fake reviews better than humans
- SEO as we know it is evolving—API calls don't serve ads
The entire apparatus of "brand equity" assumes human perception as the medium. Agents don't perceive. They process.
What do agents actually care about? That's a question we've spent considerable time answering—and the framework looks nothing like traditional marketing. Data fidelity matters more than creative excellence. Response speed matters more than brand recognition. Transparency matters more than persuasion.
The businesses that figure this out will have an unfair advantage. The ones that don't will wonder why their marketing stopped working.
Not Everything Gets Automated
Here's something the AI maximalists miss: not all commerce will be agent-mediated.
There's a hierarchy. Some purchases get fully delegated—routine replenishment, commodity products, anything purely transactional. Some purchases stay human—experiential products, aesthetic judgments, anything where physical interaction adds irreplaceable value.
Understanding where your business sits in that hierarchy changes everything about how you should be investing.
We've developed a framework for this. It's informed our entire approach to helping clients navigate the transition. But the short version is: if your product benefits from being seen, touched, or experienced in context, you're not fighting the agent transition—you're positioned for it. The physical showroom might become more valuable as agents handle the commodity layer. (This is exactly why we built Neural Showrooms.)
The businesses that understand this distinction will make very different strategic choices than the ones treating AI as a universal solvent.
What Neural Partners Has Been Building
Eight months ago, we started Neural Partners with a thesis: the advertising industry is about to be rewritten, and most players don't see it coming.
We weren't interested in building another agency that uses AI tools. We were interested in building infrastructure for what's next.
Here's what we can tell you:
We've built 50+ custom MCP tools that extend AI capabilities into complex analysis and execution. We've developed proprietary systems for helping brands become discoverable, transactable, and trustable by agents—not just humans. Our Digital Experience Infrastructure practice helps businesses build this foundation. We've created frameworks for navigating the transition that we're actively deploying with clients.
We're not theorizing about the Intention Economy. We're engineering for it.
Here's what we won't tell you:
The full operational playbook. The complete framework. The detailed architecture.
Not because we're being coy—because competitive advantage matters. We've invested significant time developing approaches that work. That IP is how we create value for partners.
What we will say: the solution isn't adding AI to your existing marketing stack. It's not optimizing your Facebook ads with machine learning. It's not building a chatbot.
It's recognizing that when your customer is increasingly code, everything needs to be rebuilt from first principles.
The Bimodal Reality
One thing we've learned: the future isn't purely automated or purely human. It's both—deliberately coupled.
You need infrastructure that agents can query AND experiences that humans value. You need machine-readable data AND brand narratives that resonate. You need API-first architecture AND white-glove service for complex decisions.
Most businesses are optimized for one mode. The ones that win will operate in both simultaneously.
We've built Neural Partners around this insight. Our technology layer (Neural Core) handles the agent-facing infrastructure. Our services layer (Neural Labs) handles the human-facing strategy and execution. Same data, two presentations. Same clients, two value propositions.
It's not either/or. It's both/and. And getting the integration right is harder than it looks.
The 18-Month Window
Here's our honest read on timing:
The protocol landscape is consolidating now. MCP, ACP, A2A—the standards are emerging. Payment networks are falling in line. The infrastructure is being built.
Businesses that start building agent-ready infrastructure in the next 18 months will have a meaningful head start. They'll learn what works, iterate on their approaches, and establish positions before the space gets crowded.
Businesses that wait for "clarity" will find themselves playing catch-up in a market where the early movers have already figured out what agents reward.
This isn't fear-mongering. It's pattern recognition. We've seen this movie before with mobile, with social, with programmatic. The businesses that moved early didn't just survive the transition—they defined it.
The Question We're Asking Every Prospective Partner
When we talk to businesses about working together, we start with a simple question:
If an AI agent were shopping for what you sell, would it find you?
Not "would a human find you through Google." Would an agent find you—through structured data queries, through API calls, through protocol-compliant interfaces?
For most businesses, the honest answer is no. And that answer gets more expensive every month.
What Happens Next
The Attention Economy isn't dead yet. There's still money to be made optimizing traditional channels. But the trajectory is clear: zero-click searches rising, agent traffic exploding, protocol standards consolidating, payment networks preparing for autonomous transactions.
The question isn't whether the Intention Economy is coming.
The question is whether you're building for it—or hoping it doesn't arrive before you're ready.
At Neural Partners, we've made our bet. Eight months of building. 50+ MCP tools deployed. Frameworks developed and tested. Clients already navigating the transition with our help.
We're not claiming to have all the answers. The landscape is evolving fast, and anyone who says they've figured it all out is selling something.
But we have figured out enough to be useful. Enough to help businesses see what's coming and start preparing. Enough to build infrastructure that works today and scales into what's next.
The void is real. The window is open. The question is what you do about it.
Ready to Build for What's Next?
The marketing funnel is dying. Traditional advertising is facing structural disruption. The economic foundation of the internet is shifting.
But opportunity exists for businesses willing to see clearly and move decisively.
We're having these conversations with forward-thinking brands right now. If you want to be part of them, reach out.
Sources & References
Market Size & Projections:
- WARC Global Advertising Report 2025 ($1.17T global ad spend)
- McKinsey & Company, "The Agentic Commerce Opportunity" (November 2025)
- McKinsey Global Survey on AI (November 2025)
- BoF-McKinsey State of Fashion 2026
Search & Traffic Data:
- Semrush AI Overviews Study (10M+ keywords, March 2025)
- Similarweb Zero-Click Analysis (May 2025)
- Adobe Digital Insights (GenAI traffic growth)
- HUMAN Security (AI agent traffic patterns)
Protocol & Infrastructure:
- OpenAI/Stripe Agentic Commerce Protocol (September 2025)
- Google A2A Protocol / Linux Foundation (June 2025)
- Visa Trusted Agent Protocol (October 2025)
- Cloudflare/WIRED Interview, Matthew Prince (December 2025)
This concludes our three-part series on The Agentic Consumer. For ongoing intelligence on AI-mediated commerce and the Intention Economy, subscribe to Neural Intelligence.
Subscribe to Neural Intelligence
Get ongoing insights on AI-mediated commerce and the Intention Economy delivered to your inbox.
Subscribe