Agentic Commerce

The Four Strategies for AI-Mediated Commerce

The Shift from B2C to B2A2C

The commerce landscape is undergoing a fundamental transformation. As AI agents become increasingly capable, they're not just tools that help customers—they're becoming functional customers themselves. This shift from Business-to-Consumer (B2C) to Business-to-Agent-to-Consumer (B2A2C) requires a complete rethinking of commerce strategy.

đź’ˇ Key Insight

In the emerging B2A2C model, brands must optimize for two distinct audiences: human customers who value experience and emotion, and AI agents who prioritize data, APIs, and verifiable claims.

The 2Ă—2 Strategy Matrix

Agentic commerce strategies can be mapped across two dimensions: who controls the agent (brand vs. customer) and where the agent operates (brand channels vs. third-party platforms).

Brand Channels
Third-Party Platforms
Brand-Owned Agent

Strategy A

Brand AI on brand website
e.g., Acme.com chat assistant

Strategy C

Brand AI on third-party
e.g., Acme GPT in ChatGPT store

Customer-Owned Agent

Strategy B

Customer AI on brand channels
e.g., Shopping bot visiting Acme.com

Strategy D

Personal AI assistants
e.g., Gemini shopping across brands

The Four Strategies Explained

Strategy A: Brand AI on Brand Channels

This is the most common and immediately viable approach. Brands deploy their own AI assistants on their websites, apps, and contact centers. The AI guides customers through discovery, comparison, and purchase—all within the brand's controlled environment.

Example: Alex visits Acme Laptop Inc.'s website looking for a business laptop. The Acme AI assistant asks about his needs, recommends the ProBook X15, shows reviews, offers a business discount, and completes the checkout—all without Alex leaving Acme.com.

âś… Strategic Priority: HIGH

This is where brands should invest heavily today. It offers full control over the customer experience, data ownership, and brand narrative.

Strategy B: Customer AI on Brand Channels

In this emerging scenario, customers deploy their own AI agents to shop on their behalf. These autonomous agents visit brand websites, negotiate prices, compare options across competitors, and make purchase decisions based on the customer's preferences.

Example: Alex instructs his personal AI agent to "find the best business laptop under $1,500." The agent connects to Acme, TechDirect, and BestBuy via UCP-compliant APIs, discovers available products, negotiates enterprise discounts, and returns the best options—all through standardized protocol calls.

⚠️ Strategic Priority: PREPARE

Brands must prepare their infrastructure for agent-to-agent commerce: structured data, APIs, and machine-readable content become critical.

Strategy C: Brand AI on Third-Party Platforms

Brands extend their reach by deploying AI agents on platforms like ChatGPT, Claude, or other AI marketplaces. This captures customers during their research phase but faces significant limitations.

Example: Alex asks ChatGPT for laptop recommendations. ChatGPT's "Acme GPT" plugin provides detailed product information, but cannot access inventory, complete transactions, or offer personalized pricing—forcing Alex to visit Acme.com anyway.

đź’ˇ Strategic Priority: LOW

Current platform limitations (no transactions, limited personalization) make this more of a lead generation channel than a commerce solution.

Strategy D: Personal AI Assistants

This represents the most disruptive scenario: customers use general-purpose AI assistants (like Google Gemini or Apple Intelligence) that can shop across all brands, maintain purchase history, and make decisions based on a comprehensive understanding of the customer's preferences.

Example: Alex asks Gemini to "find me a business laptop." Gemini connects to multiple retailers via UCP, compares products using Alex's stored preferences, negotiates the best deal, and completes the purchase—all within a single conversation. Alex never visits any brand website.

🛡️ Strategic Priority: DEFEND

This scenario threatens brand disintermediation. Brands must build direct customer relationships and unique value that AI agents cannot replicate.

Strategic Recommendations

Invest in Strategy A

Deploy sophisticated AI agents across all owned channels. Focus on creating seamless end-to-end experiences that delight customers and capture valuable first-party data.

Prepare for Strategy B

Build robust APIs and structured data layers that allow customer agents to interact efficiently with your commerce infrastructure. The brands that make it easy for agents to shop will win.

Experiment with Strategy C

Maintain presence on AI platforms for brand awareness, but don't over-invest given current limitations. Monitor platform evolution closely.

Defend Against Strategy D

Build direct customer relationships that transcend transactional interactions. Create unique value through services, community, and experiences that AI agents cannot replicate.

The Universal Commerce Protocol (UCP)

Central to the B2A2C future is the Universal Commerce Protocol (UCP)—an open standard co-developed by Google with Shopify, Target, Walmart, and endorsed by 20+ industry partners including Visa, Mastercard, Stripe, and Best Buy. UCP creates a common language for AI agents to communicate with merchant backends, enabling seamless commerce across any platform.

UCP defines the core capabilities that make commerce programmable for agents:

  • Product Discovery: Agents query machine-readable catalogs via standardized APIs
  • Checkout Negotiation: Agents understand cart requirements, discounts, and fulfillment options
  • Transaction Completion: Secure, tokenized payments with cryptographic proof of consent
  • Post-Purchase Workflows: Order tracking, returns, and customer support
  • Capability Extensions: Loyalty programs, subscriptions, and merchant-specific features

UCP is designed to work with existing protocols including Agent2Agent (A2A), Agent Payments Protocol (AP2), and Model Context Protocol (MCP). Merchants expose their capabilities through profiles that agents dynamically discover, reducing the need for custom integrations.

đź’ˇ The UCP Opportunity

Brands that adopt UCP gain access to AI-powered shopping surfaces like Google AI Mode and Gemini, where high-intent shoppers can discover and purchase without leaving the conversation. The protocol is open-source—merchants can implement via REST APIs, MCP, or A2A bindings.

Conclusion: The Agent-Ready Brand

The shift to agentic commerce isn't a distant future—it's happening now. Brands must simultaneously:

  1. Enhance human experiences on owned channels (Strategy A)
  2. Build agent-ready infrastructure for autonomous shopping (Strategy B)
  3. Maintain visibility on AI platforms (Strategy C)
  4. Protect differentiation against disintermediation (Strategy D)

The winners in this new landscape will be brands that master the dual optimization: delighting human customers while efficiently serving their AI agents.

🎮 Try the Interactive Demos →

Experience the Four Strategies

Interact with simulations of each agentic commerce approach

Brand AI on Brand Channels

Experience shopping with Acme Laptop Inc.'s AI assistant on their website

Acme Laptop Assistant

Online • Ready to help
Hi! đź‘‹ Welcome to Acme Laptop Inc. I'm here to help you find the perfect laptop for your needs. What are you looking for today?
I need a business laptop What's the price? Compare options Checkout

What's Happening Here

This is Strategy A in action—a brand-controlled AI on the brand's own website. The experience is:

  • Fully controlled: Acme manages every aspect of the conversation
  • Data-rich: All interactions feed into Acme's CRM and analytics
  • End-to-end: From discovery to checkout without leaving the site
  • Personalized: Can access your account history, loyalty status, and preferences

Customer AI on Brand Channels

Watch your autonomous shopping agent work across multiple retailers

🤖

ShopAgent Pro

Ready
Task Input
Task Queue
Click Execute to start the agent
Activity Log
Waiting for task...
Find best business laptop under $1500 Compare laptops with 32GB RAM Find deals on gaming laptops

What's Happening Here

This is Strategy B—your personal AI agent shopping autonomously across multiple brands:

  • Multi-vendor: Simultaneously searches Acme, TechDirect, BestBuy, etc.
  • Negotiates: Automatically requests enterprise discounts and price matching
  • UCP-powered: Connects to merchants via Universal Commerce Protocol APIs
  • Autonomous: Can complete purchases with your pre-authorization

Brand Challenge: How do you differentiate when a robot is comparing you purely on specs and price?

Brand AI on Third-Party Platform

Experience Acme's GPT in the ChatGPT store

Acme Laptop GPT
Previous chat...
Another chat...
A

Hi! I'm the Acme Laptop GPT. I can help you learn about Acme's laptop lineup, compare specifications, and provide recommendations based on your needs.

What would you like to know about Acme laptops?

Recommend a laptop I want to buy this Compare models

What's Happening Here

This is Strategy C—Acme's branded GPT in the ChatGPT store. Notice the limitations:

  • Information only: Can provide specs and recommendations, but cannot transact
  • No personalization: Doesn't know your purchase history or preferences
  • No real-time data: Can't check inventory, current prices, or promotions
  • Handoff required: You still need to visit Acme.com to actually buy

Strategic Value: Good for brand awareness and lead generation, but limited commerce capability.

Personal AI Assistant

Experience shopping through Google Gemini powered by the Universal Commerce Protocol

Hi! I can help you shop across any UCP-enabled retailer. I'll search for products, compare prices across merchants, negotiate deals, and complete purchases on your behalf—all through secure, standardized protocol calls.

What would you like to find today?

Find me a business laptop Buy the best one Yes, proceed

What's Happening Here

This is Strategy D—the most disruptive scenario. Your personal AI assistant:

  • Brand-agnostic: Searches ALL retailers without brand loyalty
  • UCP-powered: Connects to all merchants via Universal Commerce Protocol
  • Full transaction: Can complete purchases end-to-end
  • Cross-platform: Maintains context across all your shopping

Brand Threat: Complete disintermediation. The customer never visits your website, sees your branding, or experiences your differentiation.

Strategy Comparison Summary

Dimension Strategy A Strategy B Strategy C Strategy D
Agent Owner Brand Customer Brand Customer
Location Brand site Brand site Third-party Third-party
Brand Control Full Limited Partial None
Data Ownership Full Shared Platform Customer
Transaction Capability Full Full None Full
Current Viability Now Emerging Now (limited) Near-term
Strategic Priority INVEST PREPARE EXPERIMENT DEFEND