Your Complete Guide to Agentic Commerce in 2026

What Is Agentic Commerce?

Agentic commerce refers to shopping powered by autonomous AI agents that can anticipate, plan and execute purchases on behalf of consumers. Unlike traditional chatbots that simply respond to prompts, these agents reason through complex decisions, remember preferences and take independent action to complete transactions.

The difference between a chatbot and an agentic shopping assistant is significant. When you ask ChatGPT for product recommendations, it provides information. When you ask an agent to buy you running shoes under £100 that match your gait type and preferred brands, it searches multiple retailers, compares options, checks your loyalty rewards, applies the best available discount and completes checkout. That entire journey happens autonomously.

According to Gartner, 33% of enterprise software applications will incorporate agentic AI by 2028, up from less than 1% in 2024. For ecommerce specifically, Morgan Stanley projects that agentic shoppers could represent $190 billion to $385 billion in US ecommerce spending by 2030, capturing 10% to 20% of market share.

McKinsey's research is even more striking. By the end of the decade, the US B2C retail market alone could see up to $1 trillion in orchestrated revenue from agentic commerce, with global projections reaching $3 trillion to $5 trillion.

These aren't distant predictions. Roughly 23% of Americans have already made purchases using AI in the past month. Shopping-related generative AI searches grew 4,700% between July 2024 and July 2025. The infrastructure is being built, the protocols are being standardised and consumer behaviour is shifting faster than most brands realise.

How Agentic Commerce Works

To understand agentic commerce, you need to understand both how AI agents function and how they interact with merchants and payment systems.

The Three Components of a Shopping Agent

Every effective shopping agent relies on three core capabilities.

Memory allows agents to retain your preferences, sizes, past purchases and brand affinities across sessions. When you tell an agent you prefer sustainable brands and typically wear a size 10, it remembers this for every future shopping task.

Reasoning enables agents to break down complex requests into structured, actionable steps. Rather than simply searching for keywords, an agent reasons through what you actually need, which criteria matter most and how to evaluate trade-offs between price, quality and convenience.

Tools give agents the ability to take action. Through APIs and integrations with ecommerce platforms, agents can search catalogues, check inventory, apply discount codes, manage loyalty accounts and complete checkout. Without tool access, an agent is just a very sophisticated recommendation engine.

The Three Interaction Models

Agentic commerce operates through three primary interaction models.

Agent to site describes agents interacting directly with merchant platforms. Your personal shopping agent visits a retailer's website, searches their catalogue, adds items to a cart and completes checkout. This works best when retailers have optimised their sites for agent interaction with structured data and robust APIs.

Agent to agent involves autonomous communication between AI systems. Your shopping agent might negotiate directly with a retailer's commerce agent to secure a bundle discount or confirm delivery windows. This machine-to-machine negotiation can happen in milliseconds and optimise outcomes that would take humans hours to achieve.

Brokered agent to site uses intermediary systems to facilitate multi-agent and multi-platform interactions. When booking a restaurant through an agent, for instance, the personal agent might communicate with a platform like OpenTable's agent, which then coordinates with individual restaurant systems to find availability and apply loyalty benefits.

As these models mature, the boundaries between platforms will blur. The traditional funnel of search, browse, compare and buy will collapse into a single, intent-driven interaction.

The Protocols Powering Agentic Commerce

For agentic commerce to scale, AI agents need a common language to communicate with merchants and payment providers. This is where protocols become crucial.

Universal Commerce Protocol (UCP)

The Universal Commerce Protocol, co-developed by Google and Shopify and launched at NRF 2026, establishes an open standard for AI agents to connect and transact with any merchant. UCP covers the entire shopping journey, from discovery and checkout to post-purchase support.

What makes UCP significant is its industry backing. Walmart, Target, Etsy, Wayfair, Mastercard, Visa, Stripe, PayPal, American Express and more than 20 other major retailers and payment providers have endorsed the protocol. This isn't a single company's proprietary system. It's an industry-wide standard designed for interoperability.

UCP is built in layers, similar to how TCP/IP underpins the internet. The shopping service defines core transaction primitives like checkout sessions and line items. Capabilities add functional areas such as checkout, orders and catalogue management. Extensions augment these with domain-specific features for complex scenarios like subscriptions, pre-orders or custom delivery windows.

For Shopify merchants, UCP is now available through Agentic Storefronts, managed directly from the Shopify Admin. This means brands can sell through Google AI Mode, the Gemini app and Microsoft Copilot without building custom integrations for each platform.

Agentic Commerce Protocol (ACP)

OpenAI's Agentic Commerce Protocol, developed in partnership with Stripe, powers shopping within ChatGPT. ACP enables agents to reason over structured state, invoke merchant tools at each step and keep customers informed in real time throughout the purchase process.

While UCP takes a broader view of the entire commerce lifecycle, ACP focuses specifically on standardising the transaction flow. Both are open source and the ecosystem will likely support both, with merchants implementing capabilities based on which AI platforms they want to reach.

Agent Payments Protocol (AP2)

Google's Agent Payments Protocol addresses the critical challenge of secure, verifiable payments in agentic transactions. AP2 uses cryptographically signed mandates that link intent, cart and payment across users, merchants and payment networks, creating an audit trail that ensures transparency and accountability.

Mastercard, PayPal, American Express, Adobe and Alibaba have backed AP2 as a secure, open standard for agent-led transactions. This is essential for building consumer trust. When an AI agent makes a purchase on your behalf, you need confidence that the transaction is authorised, traceable and reversible if something goes wrong.

Model Context Protocol (MCP) and Agent-to-Agent Protocol (A2A)

Anthropic's Model Context Protocol enables AI agents to share context, intent and data across different models and tools. Unlike static prompts, MCP allows persistent, structured communication, enabling agents to retain memory and objectives across environments.

Google's Agent-to-Agent Protocol enables autonomous agents to coordinate, negotiate and complete tasks directly with one another. A2A supports long-running tasks, dynamic capability discovery and multimodal collaboration, establishing the foundation for complex multi-agent ecosystems.

These protocols work together. A shopping agent might use MCP to maintain context about your preferences, A2A to negotiate with a retailer's agent, AP2 to authorise payment and UCP to complete the checkout. The standardisation of each layer makes the entire system more reliable and scalable.

Why Agentic Commerce Matters for Ecommerce Brands

The commercial implications of agentic commerce extend far beyond a new sales channel. This represents a fundamental restructuring of how products are discovered, how buying decisions are made and how customer relationships are maintained.

The Discovery Shift

Traffic from AI platforms to US ecommerce sites surged 4,700% year-over-year in 2025, according to Adobe. ChatGPT now drives referral traffic equal to roughly 20% of Walmart's total visits. When consumers use AI for product discovery, they bypass traditional search engines, social media and even marketplace search bars.

This changes everything about visibility. Traditional SEO optimises for keywords and metadata. Generative Engine Optimisation (GEO) optimises for solutions. The question isn't whether your product appears for a specific search term, but whether an AI agent recommends your product when asked to solve a customer's problem.

Salesforce research shows 39% of consumers are already using AI for product discovery, with over half of Gen Z adopting these tools. The brands that appear in AI recommendations today will build the relationships and loyalty that compound tomorrow.

The Conversion Opportunity

Early data suggests AI-generated product recommendations drive 4.4 times higher conversion rates than traditional search, according to McKinsey. When an agent pre-qualifies products based on detailed customer context, the remaining options are already filtered for relevance.

But there's a catch. Current conversion from AI-driven traffic lags significantly behind other channels. This gap exists because merchant infrastructure wasn't built for agents. Retailers with clean, structured data and responsive APIs capture the conversion potential. Those without it lose customers at the moment of transaction.

The Relationship Challenge

In an agentic world, your customer may no longer be a human with a browser. It's an autonomous agent acting on that customer's behalf. This creates both risk and opportunity.

The risk is disintermediation. If agents optimise purely for price and specifications, brand differentiation becomes harder. An agent comparing running shoes might not care about your carefully crafted brand story or aspirational marketing.

The opportunity is in the moments that require human involvement. Complex purchases, emotional decisions, high-consideration products. Agents will escalate to humans when the situation demands it. Brands that excel at these touchpoints can build loyalty that transcends algorithmic optimisation.

How to Prepare Your Ecommerce Store for Agentic Commerce

Preparing for agentic commerce requires action across data, infrastructure and strategy. Here's what matters most.

Structure Your Product Data

AI agents rely on clean, consistent, structured information to make accurate decisions. When product data is fragmented or inconsistent, agents struggle to interpret offerings, assess value or communicate reliably.

Every product variant should have a valid GTIN (UPC/EAN). Attributes need standardisation. If your colour is listed as "Midnight Sky" rather than a standard hex code or colour family, an agent may not match it correctly to a customer preference.

Beyond basic attributes, agents benefit from enriched fields including compatibility data, materials, specifications, sustainability information, sizing guidance and variant details. Schema.org markup and GS1 standards help machines understand exactly what a product is, not just what it's called.

Optimise for AI Discovery (GEO)

Generative Engine Optimisation is emerging as the new discipline alongside traditional SEO. The goal is ensuring your products and brand appear in AI-generated responses.

This means anticipating questions rather than keywords. Instead of optimising for "waterproof jacket", optimise for the scenarios an agent might reason through. What materials provide waterproofing? What features matter for hiking versus commuting? What trade-offs exist between breathability and water resistance?

Product descriptions should be solution-oriented and complete. FAQ content should address the questions an agent might ask on behalf of a customer. Technical specifications should be machine-readable, not buried in PDFs or images.

Ensure API Readiness

Agents need programmatic access to inventory, pricing, checkout and fulfilment information. If your ecommerce platform doesn't expose this data through robust APIs, agents cannot transact with you.

For Shopify merchants, the platform handles much of this complexity through Agentic Storefronts. But even Shopify brands should audit their data feeds, ensure pricing rules are consistent and verify that inventory syncs accurately across channels.

Custom platforms face more work. UCP provides clear specifications, but implementation requires investment in API development, testing and ongoing maintenance.

Prepare for Agentic Payments

Payment authorisation in agentic commerce works differently than traditional checkout. Agents need delegated authority to spend within defined limits, with clear consent frameworks and audit trails.

Mastercard's Agent Pay technology and Google's AP2 protocol are establishing the rails for secure agentic payments. But brands should also consider how loyalty programmes, stored payment credentials and fraud prevention systems will adapt.

The key principle is transparency. Every agent action should be logged and explainable. Customers should understand what their agent did, why it did it and how to reverse or modify decisions.

Trust, Security and the Customer Relationship

Trust is the foundation of agentic commerce. When customers delegate purchasing authority to AI agents, they need confidence that those agents act in their interests, that merchants honour transactions and that the entire system is secure.

Know Your Agent (KYA)

Just as financial services require Know Your Customer (KYC) verification, agentic commerce requires Know Your Agent processes. Merchants need to distinguish legitimate shopping agents from malicious bots attempting fraud or data scraping.

Companies like Trulioo and HUMAN are building agent verification capabilities. Prove Identity has launched Verified Agent, linking digital identity, intent and payment credentials. These systems help merchants welcome beneficial agent traffic while blocking threats.

Fraud Prevention

Agentic commerce introduces new fraud vectors. Malicious agents might exploit pricing errors, abuse promotional codes or attempt unauthorised transactions. Traditional fraud detection systems, designed around human behavioural patterns, may not recognise agent-based attacks.

Payment providers are adapting. The challenge is building fraud systems that approve legitimate agent transactions quickly while flagging suspicious patterns. This requires new data signals, new models and new approaches to risk scoring.

Maintaining the Human Connection

For all the efficiency of agentic commerce, some transactions demand human involvement. Regulatory requirements, merchant policies or situations an agent cannot handle will always require escalation.

UCP models this through checkout states. When a transaction needs human input, the protocol provides a standard way to specify what information is needed and how to collect it. A furniture retailer requiring customers to select a specific delivery window can communicate this requirement to any agent through UCP.

Brands that excel at human touchpoints will differentiate in an increasingly automated landscape. The agent handles routine purchases. The human relationship handles everything else.

What Comes Next: The Future of Agentic Commerce

The protocols and capabilities launching in 2026 are just the beginning. Several trends will shape how agentic commerce evolves.

Multi-Agent Systems

Today's agents largely work independently. Tomorrow's will collaborate. Gartner predicts one-third of agentic AI implementations will combine agents with different skills by 2027.

In commerce, this means your personal shopping agent might coordinate with inventory management agents, delivery logistics agents and customer service agents to orchestrate complex purchases. Multi-agent systems can handle scenarios no single agent could manage alone.

B2B Expansion

The $15 trillion in B2B spending that Gartner expects AI agents to intermediate by 2028 represents a massive opportunity. Agent-to-agent negotiation, automated procurement and intelligent supply chain management will transform business purchasing.

B2B merchants should prepare now. The same data quality and API readiness requirements apply, along with additional complexity around contract pricing, approval workflows and compliance requirements.

Vertical Specialisation

General-purpose shopping agents will give way to specialists. A fashion agent that understands fit, style and trends. A home improvement agent that knows compatibility, installation requirements and building codes. A groceries agent that manages household inventory and dietary preferences.

These vertical agents will demand deeper product data and more sophisticated integrations. Brands with rich, domain-specific content will be better positioned to appear in specialised recommendations.

The End of Channel-Based Marketing

Gartner predicts 60% of brands will use agentic AI for streamlined one-to-one interactions by 2028. When AI agents serve as persistent digital concierges spanning marketing, sales and support, the traditional channel-based approach to customer engagement becomes obsolete.

Brands must think about relationships, not campaigns. The agent remembers every interaction. The question is whether your brand earns a place in that memory.

Preparing for the Agentic Future

Agentic commerce represents the most significant shift in retail since the move online. The brands that prepare now, investing in structured data, API infrastructure and agent-optimised content, will capture value as the market matures.

The transition won't happen overnight. Most consumers will continue shopping traditionally for years to come. But the trajectory is clear. AI agents are becoming a primary interface between consumers and commerce. The question isn't whether to adapt, but how quickly.

At Charle, we've been tracking the evolution of agentic commerce since the first protocols emerged. Our search-first approach to Shopify development positions brands to succeed wherever customers discover and purchase products, including through AI agents. If you're thinking about how agentic commerce affects your ecommerce strategy, we'd welcome the conversation.