The Rise of Agentic Commerce: When Your Next Customer Is an AI Agent

For decades, commerce operated on a simple premise: businesses sell to people. Whether consumer or business buyer, another human sat on the other side of every transaction. Your entire organisation was built around this reality. Marketing teams crafted messages for human emotions. Design teams optimised experiences for human eyes. Sales teams built relationships with human decision-makers. That foundational assumption is changing. A quiet transformation is reshaping digital commerce as autonomous AI agents become a new class of customer, discovering products, evaluating options, and executing purchases on behalf of humans. This isn’t science fiction. Over 700 million people already use ChatGPT weekly. In September 2025, OpenAI and Stripe launched instant purchasing directly within chat conversations. Your next customer might not be human.

The Paradigm Shift: From Human Buyers to Agent Customers

Traditional B2C and B2B models relied on one shared truth: a human made the purchasing decision. In the Business-to-Agent (B2A) era, that assumption breaks. Intelligent systems are now the primary interface between your business and end users. These agents don’t browse, compare, or feel – they analyse, decide, and buy. Your website, your UX, your emotional branding – all were built for human logic.

Why This is happening now

Two forces make Agentic Commerce inevitable: smarter AI agents and adaptive commerce infrastructure. AI systems now understand complex intent, maintain context, and act autonomously across multi-step tasks. Hundreds of millions already rely on them weekly and those interactions are shifting from information to transactions. Old monolithic e-commerce stacks were built for humans. Modern, cloud-native architectures expose data and functionality programmatically — exactly what agents need. Standardized protocols like the Model Context Protocol (MCP) further accelerate this shift by giving agents a consistent, secure way to discover, connect to, and orchestrate these commerce services. Together, capability and infrastructure remove the final barriers. Agentic Commerce isn’t on the horizon; it’s here.

From User Experience to Agent Experience

Your organisation has invested heavily in user experience design. Fast-loading pages, intuitive navigation, beautiful visuals, frictionless checkout. This optimisation assumed human users clicking, scrolling, and tapping through digital interfaces. Agent customers don’t experience your carefully crafted user interface. They interact through data and programmatic connections. Success requires a fundamentally different design philosophy.

The Agent Experience Paradigm

Traditional user experience is reactive and treats each interaction in isolation. Users take explicit actions, and systems respond to those specific requests without retaining context. Each interaction starts fresh. This creates friction and places the entire burden of action on the user.

Agent Experience is proactive and contextual:

Making your Business discoverable to Agents

Discovery in an agent-mediated world looks fundamentally different. Instead of optimising for search engine rankings and website traffic, you need to ensure agents can find, understand, and trust your offerings.

This means:

Success is measured not by whether humans click through to your site, but whether agents cite your business, understand your offerings, and recommend you to users they serve.

From Emotional to Verifiable Trust — Redefining Credibility in the Age of Agents

Trust has always been central to commerce, but establishing trust with agent customers requires different mechanisms than building trust with humans. Humans build trust through brand reputation, personal experiences and emotional connections. Agents evaluate trust through verifiable data signals and security mechanisms that prove transactions are authorized, safe and auditable.

The technical foundation for agent trust includes:

Major technology and financial companies are building this trust infrastructure. Google, Mastercard, Visa, and PayPal have developed protocols that create tamper-evident records of what users authorise their agents to do. These records prove that when an agent makes a purchase, it’s doing so with legitimate user permission and within specified limits.

Security Without Friction

The challenge is building security that doesn’t create friction for users or agents. Traditional commerce assumed a human clicking a buy button on a trusted screen. Agentic commerce requires proving authorisation without interrupting the autonomous flow. Modern payment protocols address this by creating signed records at the moment users give instructions to their agents. The agent can then present these authorisation records when making purchases, proving legitimacy without requiring repeated human verification. This enables truly autonomous agent behavior while maintaining security and accountability.

Instant Purchasing Within AI Conversations

The shift from theory to practice is already underway. Forward-thinking companies are building the infrastructure and capabilities that make agentic commerce real. The most visible manifestation of Business-to-Agent model is direct purchasing within AI platforms. OpenAI and Stripe’s instant checkout in ChatGPT allows users to complete purchases without leaving the conversation. Over 700 million weekly ChatGPT users can now buy from participating merchants directly through chat.

This represents a fundamental change in the customer journey:

Other major platforms are following similar paths. Perplexity introduced instant purchasing within its search results. Google is testing agent-driven checkout experiences that connect directly to its product database. These moves signal a clear direction: AI platforms are evolving from information sources to complete transaction environments.

Making Existing Systems Agent-Ready

Companies don’t need to abandon existing commerce infrastructure to participate in agentic commerce. Technology providers are creating bridges that make traditional systems accessible to agents. Leading commerce platforms are implementing standards that expose product catalogs, inventory data, and transaction capabilities in formats agents can access and understand. In demonstrations, AI agents successfully navigate complete shopping journeys on these platforms, browsing products, applying promotions, and checking out entirely autonomously. This approach allows businesses to become agent-ready through incremental updates rather than expensive platform replacements. The investment required is significant but substantially lower than complete system overhauls.

The Walled Garden Strategy

As B2A Business Model becomes real, major technology companies are making strategic moves to control the ecosystem. Understanding these dynamics is critical for business leaders navigating this transformation. Technology giants are building controlled environments where they manage both the agent and the underlying commerce infrastructure. Amazon blocks outside agents from accessing its product catalogue while building its own shopping assistant. The company also launched capabilities allowing its agent to purchase from other retailers without users leaving the Amazon environment. This creates a challenge for businesses: access to customers increasingly depends on relationships with platform owners who control the agents. Getting your products in front of agent customers may require integration with platform-specific systems and compliance with platform rules.

From Blocking to Controlled Access

The initial response from many retailers was defensive, blocking unauthorized agent access entirely. This approach is evolving towards more sophisticated strategies that grant selective access to approved agents through secure connections. The next phase involves becoming the preferred commerce partner for major agent platforms. Instead of universal access or complete blocking, businesses will establish formal relationships with specific agent ecosystems, providing privileged access to product data and commerce capabilities. This creates new strategic questions like: Which agent platforms should be prioritised and how to maintain leverage when platforms control customer access? What alternatives exist if dominant agent platforms impose unfavourable terms?

Strategic Imperatives for Business Leaders – Rethink Your Technology Foundation

Monolithic commerce platforms designed for human website visitors struggle to support agent customers. The rigidity that once provided stability now creates strategic liability. Modern commerce requires flexibility. Modular, component-based systems allow you to add agent capabilities without replacing your entire infrastructure. This architectural approach makes the transition to agentic commerce economically viable and operationally manageable.

Modernising your technology foundation is not only a technical necessity – it also unlocks new ways to increase revenue. By using modular architectures and standards like the Model Context Protocol (MCP), your business can plug into agent platforms more easily, participate in conversational marketplaces, and offer premium agent-enabled services. In a Business-to-Agent world, architecture becomes a direct driver of growth, not just an IT concern.

Invest in Agent-Optimised Design

Your design teams need to expand beyond screen-based interfaces. While AI Engine Optimisation helps make your products and business discoverable to AI systems (e.g. ChatGPT), Agent Experience design focuses on how intelligent systems perceive your offerings, evaluate options, and build trust in recommendations as they interact and transact with your business on behalf of users.

Three principles guide effective Agent Experience:

These principles require rethinking information architecture, data structures, and how you communicate value propositions. The investment extends beyond technology to organisational capabilities and design methodologies.

Build an Agent-Compatible Ecosystem

No business operates in isolation. Business-to-Agent model requires participation in broader ecosystems of compatible systems and standards. Engage with technology partners building agent infrastructure. Adopt emerging protocol standards that allow different systems and agents to work together. Participate in industry conversations shaping how agentic commerce will function. This ecosystem approach creates network effects. As more businesses become agent-ready using compatible standards, the value of agent capabilities grows for everyone. Early participants help shape standards in ways that serve their strategic interests. 

Start Small, Think Big

The magnitude of change required can feel overwhelming. The practical approach is beginning with contained experiments that build capability and confidence

Identify high-value, low-risk opportunities to test agent capabilities:

These use cases deliver tangible value while developing your organisation’s understanding of agentic commerce. Use early experiments to identify technical challenges, organisational gaps, and capability requirements. Build confidence and competence incrementally before expanding to more complex, mission-critical customer journeys.

Outlook

Agentic commerce represents a fundamental transformation in commercial logic. For the first time in history, the customer making purchasing decisions may not be human. The shift is already underway with hundreds of millions of people interacting with AI agents weekly and purchasing directly within agent conversations becoming reality. By 2027, the majority of digital commerce engagements will begin in agent environments rather than on company websites. The companies that recognise this transformation early gain substantial advantages by establishing themselves as preferred providers in agent recommendation algorithms and building technical infrastructure that competitors will need years to replicate. Your response to Business-to-Agent model and the rise of Agentic Commerce may be the most consequential competitive decision your organisation makes this decade.

Socials