Why the trillion-dollar opportunity everyone’s discussing requires an investment nobody’s making
As a CEO in the AI commerce space, I spend a lot of time reading industry reports and talking to retail leaders. McKinsey & Company ‘s recent analysis projects $1 trillion in orchestrated revenue from agentic commerce in the US alone by 2030. Sequoia Capital Capital calls it “the next Amazon-scale opportunity.” The message is clear and consistent: AI agents will fundamentally reshape retail.
But after three years building conversational commerce solutions and countless conversations with C-suite leaders, I’ve identified a dangerous gap between these ambitious projections and current retail readiness.
The industry is describing a brilliant future while overlooking the essential prerequisite that makes that future possible.
The Early Movers Are Already Capturing Billions
Before we dismiss agentic commerce as a distant future concern, consider what Amazon revealed in their Q3 2025 earnings call:
Rufus, Amazon’s conversational AI shopping assistant, has generated $10 billion in incremental annualized sales.
Let that sink in. Not projected. Not estimated. $10 billion in actual incremental revenue from conversational commerce—and we’re still in the early stages of agentic transformation.
The data gets more compelling:
- Rufus monthly users up 140% year-over-year
- Interactions up 210% year-over-year
- Rufus customers 60% more likely to complete a purchase
- 250 million customers used Rufus in 2025
Perhaps most significant: Amazon’s CEO Andy Jassy stated the company is “exploring partnerships with third-party agents”—signaling Amazon’s readiness to open their platform to external agent platforms once they’ve mastered the fundamentals.
The inflection point is arriving. The world’s largest e-commerce player is betting billions on conversational and agentic commerce, seeing real results, and preparing to integrate with external agents.
The question for every other retailer: What’s your response to a competitor that just added $10 billion in incremental revenue through AI-mediated shopping?
The Multi-Trillion Dollar Assumption
Both McKinsey and Sequoia paint compelling visions of 2030: AI agents that anticipate needs, navigate options, negotiate deals, and execute transactions autonomously. They describe sophisticated concepts like predictive shipping, agent-to-agent negotiations, and consultative purchasing at scale.
Global projections reach $3-5 trillion in agent-orchestrated commerce. The transformation will be as significant as the shift from physical retail to e-commerce—except it will happen faster.
Amazon’s leadership understands this deeply. On their earnings call, Jassy explained his strategic vision: “I think agentic commerce is a huge long-term unlock because it solves the ‘I don’t know what I want yet, help me narrow’ problem—an area where physical retail plus a salesperson still has an edge.”
He went further, drawing a parallel to an earlier technological revolution: “[Agentic commerce] reminds me in some ways of the beginning of search engines many years ago being sources of discovery for commerce.”
They learned from the search revolution. They’re not getting caught unprepared for the agent revolution.
But here’s the critical assumption buried in both McKinsey and Sequoia reports: retailers will have “agent-ready” infrastructure when external agents arrive.
Amazon is preparing. The question is: are you?
What “Agent-Ready” Actually Requires (And Why Amazon’s Investment Matters)
When McKinsey and Sequoia describe agent-ready infrastructure, they’re talking about:
- Product catalogs with semantic and behavioral metadata (agents must understand why someone buys, not just what they’re buying)
- Robust APIs enabling autonomous discovery and transaction
- Agent authentication systems and trust frameworks
- Dynamic pricing and inventory-aware recommendation capabilities
- Identity management that works for AI, not just humans
These aren’t capabilities you suddenly unlock when agents arrive. They’re infrastructure you must build through operational experience.
Amazon has been building this through Rufus for years. They now have:
- Operational data from billions of AI interactions (210% YoY growth in interactions)
- Product data optimized for AI understanding (not just human browsing)
- APIs tested at scale for AI-mediated transactions
- Trust frameworks validated with real customers (60% higher conversion)
- Competitive intelligence on what works in AI-mediated shopping
On the earnings call, Jassy positioned Amazon’s advantage clearly: “The customer experience with third-party agents in agentic commerce is currently lacking in personalization, accurate delivery estimates, and pricing.”
Translation: “We’ve built what agents need. Our competitors haven’t.”
And here’s the strategic question every board should be asking: How do you build infrastructure for something you’ve never operated?
Amazon spent years learning through Rufus before they’d even consider integrating with external agents. They walked before they’re running.
The Missing Bridge: Conversational Commerce as Strategic Foundation
The path from traditional e-commerce to agentic commerce isn’t a single leap. It’s a necessary progression:
Traditional E-commerce → Conversational Commerce → Agentic Commerce
Amazon’s trajectory proves this:
- 2023-2024: Deploy Rufus, learn from billions of interactions
- 2025: Generate $10B incremental revenue, prove the model works
- 2026 and beyond: Integrate with third-party agents from a position of operational mastery
Conversational commerce—deploying intelligent AI assistants on your own properties—isn’t an alternative strategy to agentic readiness. It’s the only realistic path to get there.
Here’s the strategic logic:
1. Discovery Through Operation
You cannot identify infrastructure gaps through planning exercises. Amazon discovered through operating Rufus what product data, APIs, and customer experience elements actually matter for AI-mediated commerce.
When you deploy conversational AI on your site, you immediately discover:
- Product data structured for human browsing doesn’t work for AI problem-solving
- APIs built for page rendering can’t support intelligent orchestration
- You lack the semantic layer connecting customer problems to product solutions
- Your systems assume human patience and judgment that AI won’t provide
These discoveries only happen through live operation. At NewEcom.ai we consistently see retailers uncover 10-15 critical gaps within the first 90 days of deployment.
Amazon spent years accumulating this learning. Your competitors who start today will have 2-3 years of it when external agents become significant. When will you start?
2. Infrastructure Reusability
Every capability you build for your own conversational AI becomes reusable infrastructure for external agents:
- Semantic product data works for both your agents and external agents
- Problem-solving APIs can be accessed by any authenticated agent
- Trust mechanisms you develop scale across the ecosystem
- Agent authentication frameworks apply universally
Amazon explicitly noted they’re positioning for third-party agent partnerships. That integration will be seamless because they built the infrastructure through Rufus first.
You’re not building twice. You’re building once, properly.
3. Competitive Intelligence Through Experience
McKinsey dedicates significant analysis to trust and risk in agentic commerce, noting that “trust becomes foundational infrastructure.”
Amazon’s data proves why operational experience matters:
- 60% higher conversion when customers use Rufus
- 210% increase in interactions year-over-year (customers trust it more)
- 140% growth in monthly users (word spreads about what works)
- 250 million customers willing to engage with AI shopping assistance
But trust isn’t theoretical. You build it through operational experience:
- Learning how customers respond to AI recommendations
- Understanding which products AI prefers and why
- Identifying failure modes before they happen at scale
- Developing optimization strategies for AI-mediated shopping
This intelligence is competitively differentiating. Amazon has years of it. Retailers who start accumulating it today will have enormous advantages when external agents become significant.
Retailers who wait will face an unrecoverable gap.
The Competitive Math That Should Concern Every Board
Let’s make this concrete with Amazon’s actual results:
Amazon’s Rufus impact:
- $10B in incremental annualized sales
- Started development ~2-3 years ago
- Now generating 140% user growth YoY and 210% interaction growth YoY
- Approaching readiness to integrate with third-party agents
Your competitive position if you start today:
- 2025: Begin conversational commerce deployment
- 2026: 12 months of operational learning
- 2027: 24 months of optimization, approach Amazon’s current maturity
- 2028-2030: Ready when external agents scale
Your competitive position if you wait 12 months:
- 2026: Begin conversational commerce deployment
- 2027: 12 months of operational learning
- 2028: Still learning basics while Amazon + early movers have 3+ years experience
- 2028-2030: Permanently behind in agent-mediated commerce
The gap compounds. Amazon’s 210% increase in interactions means their learning velocity is accelerating. Every quarter they’re pulling further ahead of retailers who haven’t started.
The Strategic Timeline That Should Concern Every CEO
Here’s the math:
- Agentic commerce inflection: 2028-2030 (McKinsey projection)
- Conversational commerce implementation: 6-18 months
- Operational learning period: 12-24 months
- Agent-ready infrastructure build: Additional 12-18 months
Total preparation time: 30-60 months Time remaining: 36-60 months
Amazon’s timeline: They’re 24-36 months into this journey already
The window is tighter than it appears. Retailers who begin today will have 2-3 years of operational experience when external agents become market-significant.
Retailers who wait 12 months to “see how it develops” will face an unrecoverable disadvantage—with Amazon demonstrating exactly what that disadvantage looks like in quarterly earnings.
The Business Model Implications
Agentic commerce isn’t just a channel shift—it’s a fundamental restructuring of value capture. Sequoia correctly notes: “Agentic commerce requires a fundamental rethinking of how value is created, captured, and delivered.”
Amazon’s earnings call provided strategic insight into how they’re thinking about this:
“If you know what you want to buy, there are few experiences that are better than coming to Amazon. I think AI and agentic commerce are going to change the experience online where narrowing what you want when you don’t know is going to get better online than it even is in physical environments.”
Translation: Amazon believes agentic commerce expands the total addressable market—and they’re positioning to capture the lion’s share of that expansion.
Three strategic questions emerge for every retailer:
1. Should you build your own agents or optimize for external agents?
Amazon’s answer: Both. Build Rufus first, then partner selectively with third-party agents.
Build your own if:
- You operate in consultative categories (complex products requiring expertise)
- You have proprietary knowledge external agents can’t replicate
- You can create genuine concierge experiences
Optimize for external agents if:
- Your products are more commoditized
- External agents will have superior comparative data
- Your competitive advantage is operational, not experiential
This decision has trillion-dollar implications. But you cannot make it intelligently without operational experience in AI-mediated commerce.
Amazon spent years operating Rufus before even discussing third-party agent partnerships. That sequence matters.
2. How open should you be to external agents?
McKinsey projects a shift from “traditional vertical destinations toward a more integrated, horizontal-agent ecosystem.”
Amazon is preparing to be open but from a position of strength. Jassy noted that current third-party agent experiences lack personalization, proper delivery estimates, and accurate pricing.
Translation: “We’ll work with external agents, but we’ve built infrastructure they’ll prefer routing to.”
Your options range from:
- Closed: Block agent traffic entirely (high-risk, potentially fatal)
- Selective: Whitelist preferred agents
- Open-with-differentiation: Allow all agents but compete on superior experience
- Fully open: Treat agent traffic like human traffic
Unless you’re Amazon-scale, full closure likely accelerates your irrelevance. But full openness without preparation means external agents discover your weaknesses in real-time.
The right strategy depends on your operational readiness—which you build through conversational commerce.
3. What new revenue models should you pursue?
Traditional advertising revenue will decline as agents mediate discovery. Amazon’s Rufus success suggests new models:
- Incremental sales capture: Rufus generated $10B by helping customers who might have bounced
- Higher conversion efficiency: 60% more likely to purchase means dramatically lower CAC
- Expanded TAM: Amazon believes agentic “expands the amount of shopping done online”
Additional models emerging in the ecosystem:
- Multi-brand bundling and revenue sharing
- Real-time negotiation fees
- Premium agent skills and subscriptions
- Data insights and analytics (anonymized agent behavior)
- Conversational marketplace commissions
- Contextual sponsorships integrated into agent experiences
First-movers like Amazon are defining these models. Late-movers will accept whatever standards emerge.
The Uncomfortable Competitive Reality
Most retailers will follow a predictable pattern:
- Ignore this until external agents are actively routing traffic
- Deploy superficial chatbots that teach them nothing meaningful
- Panic when Amazon and early movers demonstrate superior agent performance
- Attempt to retrofit agent-readiness into fundamentally broken infrastructure
- Fail to capture the agentic commerce opportunity
Meanwhile, Amazon just added $10 billion in incremental revenue and is preparing to partner with third-party agents from a position of operational mastery.
Your strategic advantage lies in recognizing that conversational commerce is the necessary first step that 90% of competitors will skip or execute poorly.
This is not about following Amazon. It’s about understanding the fundamental pattern of technological transformation and positioning accordingly.
Strategic Guidance: The Three Critical Decision Points
Decision Point 1: Investment Timing (Now)
Question: Do we invest in conversational commerce now or wait 6-12 months?
Strategic analysis:
- Amazon is 24+ months into this journey and generating $10B incremental revenue
- Every month delayed is operational learning you won’t have when agents arrive
- Competitors are either moving now or will be soon
- The technology is mature enough for production deployment (Amazon proves this)
- ROI is immediate—conversational commerce pays for itself even before agents arrive
- The downside of early movement is manageable; the downside of late movement is catastrophic
What Amazon’s success tells us: The learning curve is real, the revenue is real, and the competitive advantage is real. They didn’t wait to “see how it develops”—they built it and are now capturing value while positioning for the next wave.
Recommended action: Authorize Phase 1 investment immediately
Decision Point 2: Build vs. Partner (Month 6-12)
Question: Do we build agent capabilities internally or partner strategically?
Strategic analysis:
- Building requires 18-36 months and significant specialized technical talent
- Partnering accelerates deployment but creates dependencies
- Hybrid approaches (partner for foundation, differentiate on top) often prove optimal
- This decision cannot be made intelligently without 6-12 months of operational experience
What Amazon’s approach tells us: Even Amazon, with unlimited resources, built Rufus over years before considering third-party partnerships. Speed matters, but foundation matters more. They invested in deep operational learning first.
Recommended action: Begin with strategic partners who enable optionality (like NewEcom.ai ), make build/buy decision after accumulating operational learning
Decision Point 3: Ecosystem Positioning (Month 12-18)
Question: What role do we play in the agentic commerce ecosystem?
Strategic analysis:
- Infrastructure provider vs. agent operator vs. both
- Degree of openness to external agents (selective vs. broad)
- Revenue model innovation vs. fast-follower approach
- These choices define competitive position for the next decade
What Amazon’s positioning tells us: Build your own agent capability first, get operationally strong, then selectively partner. Don’t open your platform until you understand what agents need and can deliver it at scale.
Recommended action: Make decision based on 12-18 months of operational data, not theoretical analysis
How NewEcom.AI Serves as Strategic Partner in This Transformation
The challenge with agentic commerce preparation is that it requires simultaneous expertise in three distinct domains:
- Retail operations (you can’t optimize product catalogs without deep merchandising knowledge)
- Advanced AI (conversational systems, agent protocols, semantic frameworks)
- Strategic foresight (understanding ecosystem evolution and competitive dynamics)
NewEcom.ai m provides integrated capability across all three:
Operational Experience
We’ve deployed conversational commerce for retailers across categories for three years. We’ve encountered—and solved—the infrastructure problems you’re about to face. We accelerate your learning curve by 12-18 months.
While Amazon spent years building Rufus internally with massive resources, mid-market and enterprise retailers can leverage our experience to compress that timeline dramatically—without sacrificing quality or learning.
Technical Leadership
We’re actively implementing emerging protocols (MCP, A2A, AP2, ACP), monitoring agent platform evolution, and participating in standards development. We bridge today’s conversational commerce with tomorrow’s agentic ecosystem.
We track what Amazon builds, what OpenAI launches, what Google deploys—and distill the patterns that matter for your specific business context.
Strategic Advisory
We help leadership teams navigate the business model implications: new monetization strategies, competitive positioning, ecosystem participation timing. We don’t just implement technology—we guide business transformation.
Amazon has an army of strategists figuring this out internally. You can have a focused team with cross-industry pattern recognition and proven frameworks.
The Board-Level Conversation
If I were presenting this to your board, here’s what I’d emphasize:
The Opportunity
$3-5 trillion global market by 2030. This is Amazon-scale transformation. Early movers will capture disproportionate value.
Amazon’s proof point: $10B incremental revenue in approximately Year 2-3 of their conversational commerce journey. That represents meaningful value from a single capability. Scale that insight across the broader ecosystem.
The Requirement
Agent-ready infrastructure that can only be built through operational experience. 18-36 months of preparation time required to achieve competitive positioning.
Amazon’s timeline: They’re 24-36 months into this journey already and only now exploring third-party agent partnerships.
The Risk
Waiting 6-12 months to “see how it develops” creates potentially unrecoverable disadvantage. Competitors building operational experience today will have 2-3 year lead that compounds over time.
Amazon’s advantage: They have billions of AI interaction data points. Their agents understand what works. Their infrastructure is battle-tested at scale. Can you catch up starting 12 months later?
The Investment
Phase 1: $200K-500K over 6 months (pays for itself through conversational commerce ROI—Amazon proves this at scale) Phase 2: $500K-2M over 12 months (builds agent-ready infrastructure and protocols) Phase 3: $1M-5M+ over 18 months (ecosystem participation and business model innovation)
Total investment: $2M-7M+ over 36 months Opportunity cost of inaction: Potential irrelevance in $3-5T market + permanent disadvantage vs. Amazon and other early movers
Amazon’s investment: Estimated hundreds of millions in Rufus development plus billions in AI infrastructure. Their return: $10B incremental annualized revenue and positioning to dominate the agentic ecosystem.
The Recommendation
Authorize Phase 1 immediately. Partner with NewEcom.ai to accelerate operational learning and de-risk execution. Make build/buy and ecosystem positioning decisions based on operational data, not theoretical analysis.
Most importantly: Don’t let Amazon’s 24-month head start become a 36-month or 48-month permanent competitive gap.
The Final Strategic Question
The reports from McKinsey and Sequoia describe an inevitable future. Amazon’s Q3 2025 earnings prove that future is already arriving—and generating billions in value for those who invested early.
The question isn’t whether agentic commerce will transform retail—it will.
The question is: will your company be positioned to capture value when that transformation accelerates, or will you be struggling to catch up while Amazon and other early movers with 3+ years of operational advantage define the rules of the new marketplace?
Amazon didn’t wait for perfect clarity. They built, they learned, they captured $10 billion in incremental revenue, and now they’re preparing to engage with the third-party agent ecosystem on their terms.
You can begin accumulating that operational advantage today.
Next Steps
If this strategic analysis resonates, I’d welcome a conversation about:
- Your current digital commerce capabilities vs. Amazon’s demonstrated agentic readiness
- Competitive dynamics in your category and timeline pressures you face
- Appropriate investment timeline and resource allocation for your organization
- How NewEcom.ai om can accelerate your preparation while de-risking execution
The retailers who win in 2030 won’t be those with the best strategy decks. They’ll be those with the deepest operational experience in AI-mediated commerce.
Amazon has that experience and is monetizing it today. When will you start building yours?
Francois Silvain CEO, NewEcom.ai
Ready to discuss your strategic positioning? → Book a 45-minute executive briefing → Connect on LinkedIn
At NewEcom.AI, we partner with retail leaders to navigate the evolution from traditional e-commerce to conversational commerce to agentic ecosystems. Our focus is operational excellence that creates strategic advantage—the same path Amazon took to $10B in incremental revenue. The difference: we help you compress the timeline and avoid the costly mistakes we’ve already solved.
P.S. – In the time it took you to read this article, Amazon’s Rufus generated approximately $1.1 million in incremental sales. The question isn’t whether to act—it’s whether you can afford to wait another day.
Sources:
Amazon.com Announces Third Quarter Results
Sequoia: The $1T Opportunity to Build the Next Amazon in Retail
#RetailStrategy #AgenticCommerce #AITransformation #EcommerceInnovation #DigitalTransformation #AICommerce #Mckinsey #Sequoia #NewEcom #Rufus #Amazon #Conversational
