Shopify’s Agentic Commerce Strategy Is Moving Shopping Into AI Conversations

Shopify is positioning itself for a version of ecommerce in which a shopper may begin with a conversation rather than a search bar or brand homepage. In this model, an AI agent can interpret a request, identify products, compare options against live details such as price and availability, and guide the buyer to—or potentially through—checkout.

The evidence Shopify has published suggests meaningful early momentum, not a settled new order: AI-referred orders grew nearly 13 times year over year in Q1 2026, and sessions referred by AI chatbots grew more than eight times. Shopify also reports that AI-referred shoppers had average order values 14% higher than organic-search shoppers. 79

From website funnel to AI-mediated discovery

Traditional ecommerce generally begins with a search engine, ad, social post, or visit to a merchant’s site. Agentic commerce changes the entry point. A shopper can describe an outcome—such as a particular style, budget, color, or delivery need—and an AI system can search a structured catalog, narrow options, and present products in the conversation.

Shopify’s Agentic Storefronts are intended to make eligible merchants discoverable and purchasable through AI channels including ChatGPT, Google AI Mode and Gemini, Microsoft Copilot, and Meta. Eligible stores can manage those AI sales channels from Shopify Admin. 19

Underneath that distribution is Shopify Catalog, which standardizes product information so AI systems can work with titles, descriptions, imagery, pricing, inventory, and shipping data rather than depend solely on potentially stale web crawling. 6 The practical implication is straightforward: a merchant’s product page is no longer the only place where product information must persuade. Its data must also be intelligible to an agent deciding what to recommend.

Why the Universal Commerce Protocol matters

Shopify and Google co-developed the Universal Commerce Protocol (UCP), an open standard meant to define how agents connect and transact with merchants. It covers the mechanics that make commerce actionable for software systems, including catalog discovery, checkout sessions, orders, and related capabilities. 315

Shopify’s developer tooling uses UCP-compliant MCP servers to support structured actions across the buying flow: searching a catalog, building a cart, creating a checkout, handing off a buyer, and tracking orders. 1

That does not mean every conversation will end in a fully autonomous purchase. But it reduces the need for each AI platform and merchant to create a separate bespoke connection. For merchants, the strategic goal is presence wherever customers ask shopping questions; for customers, it is less manual filtering across multiple sites.

Product data becomes a discovery asset

For an AI agent, incomplete or ambiguous catalog data limits what it can confidently recommend. Shopify defines agentic-ready product data as structured, machine-parsable, real-time information that an AI system can query, interpret, and act upon. 22

That makes the basics newly consequential:

• Accurate titles and descriptions that state what the item is and how it differs from alternatives.
• Complete attributes, variants, imagery, availability, prices, and shipping details.
• Product information that supports comparison, such as material, dimensions, fit, compatibility, or use case.
• Current inventory and commerce logic so recommendations do not create a dead end at checkout. 622

Aviator Nation offers an early example of the work involved. The apparel brand has been adding more detailed metafields, product information, and imagery to its catalog. It reported roughly 20 orders through ChatGPT over 30 days—modest volume, but enough to justify preparing for an AI-led shopping channel. 17

Structured data alone cannot protect a brand story

The catalog is necessary, but it is not the entire brand. Aviator Nation’s concern illustrates why: its team found that large language model descriptions could overemphasize visible collaborations, including MLB and festival associations, instead of communicating the broader identity the company wants customers to understand. 17

That creates a second task alongside catalog cleanup: maintain a rich, consistent public narrative through imagery, editorial content, PR, and community signals. Product data tells an agent what an item is. Brand storytelling helps explain why a buyer might care.

This distinction matters especially in premium apparel and other emotionally driven categories. Fit, tactile quality, personal expression, confidence, and brand identity are difficult to compress into a list of attributes. AI can reduce the friction of finding and comparing products, but it does not automatically reproduce the feeling of aspiration or connection that often drives a high-consideration purchase.

Shop’s discovery ambition: new brands, not only familiar ones

Shopify’s Shop app is also part of the strategy. The company has described its recommendation approach as an effort to introduce shoppers to a broader set of relevant merchants rather than simply repeat brands or items they have already viewed. Shopify says almost half of purchases in Shop are from a brand the user is buying from for the first time. 7

That is an important distinction from classic retargeting, which focuses on people who have already interacted with a brand or abandoned a cart. 43 If Shop can consistently connect high-intent shoppers with unfamiliar but relevant brands, it becomes a discovery channel rather than merely another venue for follow-up advertising.

For merchants, the opportunity is customer acquisition. The risk is that recommendations become overly dependent on opaque platform ranking systems. Strong product data and a clear brand presence are therefore useful not only for conversion but also for being represented accurately when the first encounter happens outside the merchant’s own site.

Anthropic is pursuing retailer-built agents

Shopify is not the only company seeking a role in agent-led commerce. In September 2026, Anthropic announced blueprints for retailers to build shopper-facing and merchant-facing agents using Claude. Reuters described them as guidelines for retail, travel, and ticketing companies; shopper agents can make preference-based suggestions and add products to carts. 27

The distinction is significant. Rather than operating solely as a consumer shopping destination, Anthropic’s approach gives retailers building blocks for agents in their own customer experiences and internal operations. Early participants cited in reporting include Shopify, Visa, Mastercard, and Accenture. 28

The emerging competitive landscape is therefore not a simple contest for one dominant shopping chatbot. It is a contest over the standards, catalog connections, payment flows, checkout handoffs, and merchant tools that let many agents transact.

What brands should do now

The most defensible response is practical rather than speculative:

1. Audit the catalog. Ensure product attributes, variants, imagery, availability, pricing, and shipping data are complete and current. 622
2. Write for comparison. Give agents and shoppers the concrete details needed to distinguish similar products, not only evocative marketing language. 22
3. Review AI-facing brand representation. Check how AI systems describe the brand and identify where public storytelling or product context is being flattened or distorted. Aviator Nation’s experience shows that this can matter even before AI becomes a major sales channel. 17
4. Measure AI referrals separately. Shopify can attribute orders from channels such as ChatGPT in its admin tools, allowing merchants to track whether traffic and orders justify deeper investment. 10
5. Keep the human experience central. Use AI to remove search and decision friction, while preserving the creative direction, service, community, and emotional cues that differentiate a brand.

Early signal, not retail’s final form

Shopify’s reported growth rates show that AI-referral commerce deserves attention, particularly because its early order values and conversion indicators compare favorably with organic search. 79 Still, these are Shopify-reported platform metrics and individual brand experiments, not proof that AI will displace search engines, brand sites, physical stores, or human merchandising at scale.

The more immediate change is that discoverability is becoming conversational and machine-mediated. Brands that make their product information reliable, their inventory current, and their identity clear will be better positioned whether a purchase begins on their own storefront, in a search result, or inside an AI conversation.