Adobe Commerce Catalog Optimization for AI Agent Discoverability

AI shopping agents are reshaping how consumers find and buy products. Learn how to optimize your Adobe Commerce catalog with structured data, enriched attributes, and protocol-ready APIs so AI agents can discover, compare, and transact your products.

Your Adobe Commerce catalog was built for human shoppers. Someone lands on a product page, reads the description, compares a few options, and clicks “Add to Cart.” That model still works — but it is no longer the only model that matters.

AI shopping agents are already redirecting revenue. During the 2025 holiday season, traffic to retail websites from generative AI tools grew sevenfold compared with the prior year, according to Adobe’s own data. Shoppers referred from AI platforms converted at higher rates, generated more revenue per visit, and spent more time on-site. Research from nShift in early 2026 found that 58 percent of consumers have now replaced traditional search with generative AI for product discovery.

The question is no longer whether AI agents will influence your sales. It is whether they can find your products at all.

What Changes When AI Agents Shop for Your Customers

Traditional ecommerce search relies on keywords, facets, and browsing behavior. A human shopper tolerates imperfect data because they can interpret marketing language, scan images, and call a sales rep if something is unclear.

AI agents work differently. When a consumer asks ChatGPT, Gemini, or a brand-owned conversational assistant to “find the best commercial-grade espresso machine under a certain budget with a built-in grinder,” the agent needs structured, machine-readable data to evaluate your product against competitors. It queries attributes — not paragraphs. It compares structured specs — not taglines. If your catalog returns ambiguous, unstructured, or incomplete data, the agent moves on to a competitor whose catalog is machine-ready.

This is the shift Adobe recognized when it committed its commerce platform to agentic standards (opens in new tab) in February 2026, announcing support for the Universal Commerce Protocol and the Agentic Commerce Protocol. Adobe’s message to merchants was clear: product catalogs, pricing, and inventory must become machine-readable and actionable by AI agents, while still supporting conventional storefront experiences.

The Structured Data Foundation

Agentic discoverability starts with structured data. If your Adobe Commerce store lacks comprehensive Schema.org Product markup, AI agents cannot reliably extract product information from your pages.

What to implement:

JSON-LD Product markup on every product page, including name, description, sku, brand, offers (with price, priceCurrency, availability, itemCondition), image, gtin or mpn, aggregateRating, and review. This is the baseline that Google, ChatGPT, and other AI platforms use to understand what you sell.

Attribute-level specificity. Generic descriptions like “high-quality stainless steel” tell an agent nothing useful for comparison. Structured attributes with explicit values — material composition, dimensions with standardized units, certifications by code (not marketing claim), compatibility references — allow agents to match your products against precise consumer requirements.

Variant data as discrete entities. If your configurable products collapse all variant information into a single parent record, agents cannot distinguish between sizes, colors, or configurations. Each variant needs its own structured data representation with unique identifiers and specific attribute values.

Adobe Commerce supports JSON-LD output natively through its storefront theme layer, and the Hyvä frontend makes it straightforward to extend structured data templates. For headless implementations using Adobe Commerce’s GraphQL API, structured data should be generated at the frontend application layer using the rich product data available through the Catalog Service.

Catalog Enrichment for Machine Readability

Structured markup gets your products seen. Catalog enrichment determines whether agents recommend them.

Attribute completeness matters more than description quality. An AI agent evaluating espresso machines does not care about your brand storytelling. It cares whether your catalog exposes boiler type, pump pressure in bars, water reservoir capacity in liters, grinder burr material, and voltage compatibility — as discrete, queryable fields. Every missing attribute is a reason for the agent to prefer a competitor whose data is more complete.

Standardize units and taxonomies. Weight in pounds on one product and kilograms on another creates comparison friction for agents. Adopt a single unit system per attribute category, and use industry-standard classification systems (UNSPSC for B2B, Google Product Category for B2C) to help agents map your products into their internal taxonomies.

Enrich with use-case and compatibility data. Agents increasingly field contextual queries: “Which of these works with my existing setup?” or “What pairs well with this product?” Structured compatibility attributes, “works with” relationships, and use-case tags make your products answerable to these queries.

Adobe Commerce’s attribute management system is well-suited for this. Create dedicated attribute sets with mandatory completeness requirements for each product type. Use Adobe Commerce’s Product Recommendations (opens in new tab) and Live Search (opens in new tab) services — both of which now leverage AI — to surface enriched catalog data through Adobe’s SaaS Catalog Service. This enriched, indexed data is the same foundation that will power agentic commerce integrations.

Adobe Commerce’s Agentic Commerce Readiness

Adobe’s February 2026 announcement was not just a press release. It was an infrastructure commitment. Here is what it means for your store:

Universal Commerce Protocol (UCP) support means Adobe Commerce will expose product catalogs, pricing, and purchasing rules in a standardized format that any compliant AI agent can consume. Think of it as making your store “speakable” to AI in a common language, regardless of which AI platform the consumer uses.

Agentic Commerce Protocol (ACP) support extends this to full transaction capability — enabling AI agents to manage cart operations, apply promotions, execute secure checkout, and integrate with order management and fulfillment systems, all programmatically.

GraphQL API readiness. Adobe Commerce’s existing GraphQL storefront API already provides the kind of structured, queryable product data that agentic protocols require. Merchants who have invested in headless or hybrid architectures — particularly those using Hyvä’s Checkout or React-based PWA storefronts — are architecturally closer to agentic readiness because their product data is already decoupled from presentation.

The practical implication: if your Adobe Commerce implementation relies heavily on CMS blocks, static content, and unstructured product pages rather than attribute-driven catalog data exposed through APIs, you have work to do before agentic protocols can serve your products effectively.

A Practical Readiness Checklist

Rather than treating agentic commerce as a future initiative, here are concrete steps to improve your catalog’s AI agent discoverability now:

1. Audit your structured data. Use Google’s Rich Results Test and Schema Markup Validator on a sample of product pages. Identify missing fields, especially gtin/mpn, brand, variant-level offers, and aggregateRating. Every gap is a missed signal for AI agents.

2. Map your attribute completeness. Export your catalog and measure what percentage of SKUs have complete, structured attributes for their product type. Set a target — 90 percent completeness across core attributes — and treat the gap as a data quality project, not a marketing project.

3. Implement JSON-LD at the theme level. If your Adobe Commerce theme does not generate comprehensive JSON-LD Product markup, this is your highest-impact technical fix. For Hyvä storefronts, extend the product detail template. For Luma-based themes, use a dedicated module or your theme’s head block.

4. Expose your catalog through GraphQL. If you are not yet using Adobe Commerce’s GraphQL API or Catalog Service, begin planning the integration. Headless data access is the foundation for both current AI platform integrations and the upcoming protocol standards.

5. Standardize and enrich product attributes. Audit your attribute sets for consistency (units, naming conventions, completeness requirements). Add structured compatibility, certification, and use-case attributes where they do not exist.

6. Monitor AI-referred traffic. In Adobe Analytics or Google Analytics 4, create segments for traffic referred from AI platforms (ChatGPT, Gemini, Perplexity, Copilot). Track conversion rates and revenue per visit from these sources separately. This data will become your primary metric for agentic commerce ROI.

For a broader look at preparing your operations for AI-driven commerce at any budget level, see our guide on AI readiness for ecommerce.

Why This Matters Now, Not Later

Adobe’s data from the 2025 holiday season is the clearest signal yet: AI-referred commerce traffic is not hypothetical — it is growing exponentially and converting better than traditional channels. The merchants who capture this traffic are the ones whose catalogs are machine-readable today.

Agentic commerce protocols are being finalized and adopted across the industry. Adobe has committed to supporting them. The merchants who invest in catalog enrichment and structured data now will be positioned when these protocols go live. Those who wait will find themselves in the same position as businesses that ignored mobile optimization in 2014 — technically functional, but increasingly invisible to the channels where customers actually shop.

If your Adobe Commerce catalog is not ready for AI agents, the revenue does not disappear. It goes to the competitor whose catalog is. If you need help assessing your catalog’s agentic readiness or building a structured data strategy for your Adobe Commerce store, Creatuity specializes in exactly this kind of Adobe Commerce optimization.

Frequently Asked Questions

What is agentic commerce and how does it affect Adobe Commerce merchants?

Agentic commerce is the model where AI agents — powered by large language models like those behind ChatGPT and Gemini — discover products, compare options, and complete purchases on behalf of consumers. For Adobe Commerce merchants, this means your product catalog needs to be machine-readable and API-accessible so these agents can find and transact your products. Adobe has committed to supporting the Universal Commerce Protocol and Agentic Commerce Protocol to enable this.

How do I make my Adobe Commerce product catalog machine-readable for AI agents?

Start with comprehensive JSON-LD Schema.org Product markup on every product page, including variant-level data. Then enrich your catalog attributes with standardized units, structured specifications, compatibility data, and industry classification codes. Ensure your Adobe Commerce GraphQL API and Catalog Service are active and exposing complete product data. Treat attribute completeness as a data quality metric, not a content marketing exercise.

What are the Universal Commerce Protocol and Agentic Commerce Protocol?

The Universal Commerce Protocol (UCP) standardizes how AI agents access product catalogs, pricing, and purchasing rules across different commerce platforms. The Agentic Commerce Protocol (ACP) extends this to full transaction capability — cart management, promotions, checkout, and order fulfillment. Adobe Commerce has announced support for both, meaning merchants on the platform will be able to participate in AI-led shopping experiences through standardized interfaces.

Does Adobe Commerce support AI agent shopping natively?

Adobe Commerce is building toward native agentic commerce support through its commitment to UCP and ACP standards, along with existing capabilities like the SaaS Catalog Service, Live Search, Product Recommendations, and a comprehensive GraphQL API. Merchants who have invested in structured catalog data and API-driven architectures are closest to readiness. Adobe has indicated additional agentic capabilities will roll out throughout 2026.

How can I measure whether AI agents are discovering my products?

Track AI-referred traffic in your analytics platform by creating segments for visits from known AI platforms (ChatGPT, Gemini, Perplexity, Microsoft Copilot). Compare conversion rates and revenue per visit against traditional search traffic. Additionally, test your product pages with AI tools directly — ask ChatGPT or Gemini to find products in your category and see whether your store appears in their recommendations. Monitor your structured data validation scores as a leading indicator of discoverability.