Your product feed used to serve one purpose: getting your products into Google Shopping listings. With the Universal Commerce Protocol (UCP), that same feed becomes the primary way AI agents evaluate and select your products. If your feed data is incomplete, inconsistent, or outdated, AI agents will skip you entirely—and you will never know it happened.
This guide gives you a practical, checklist-driven audit of your Merchant Center product feed. You will learn exactly what AI shopping agents look for, which new attributes Google introduced for conversational commerce, and what steps to take right now so your products are discoverable when agentic commerce scales.
Whether you manage fifty products or fifty thousand, this checklist applies. The merchants who prepare their feeds now will have a significant head start when UCP adoption accelerates throughout 2026.
Why Your Feed Matters More Than Ever
AI shopping agents do not browse websites. They do not scroll through category pages, read promotional banners, or click through product detail pages. They read structured data—specifically, the product data you submit through your Google Merchant Center feed.
In the agentic commerce model, your Merchant Center feed is your storefront. Product titles, descriptions, attributes, pricing, availability, images, and every supplemental data point you provide—that is what the AI uses to decide whether to recommend your product to a shopper. There is no fallback. The agent will not visit your website to fill in the gaps.
Google recognized this shift when it announced UCP alongside dozens of new Merchant Center attributes specifically designed for conversational commerce discovery. These attributes go beyond traditional Shopping requirements. They include structured Q&A data, product compatibility relationships, substitute products, and usage scenario descriptions—all intended to give AI agents the context they need to make intelligent recommendations.
The implication is clear: if your feed only meets the minimum requirements for Google Shopping ads, you are already behind. UCP readiness requires a more complete, more accurate, and more richly attributed product feed than most merchants currently maintain.
The Core Shift
In traditional Shopping, your feed gets your products listed. In agentic commerce, your feed is the only thing the AI sees. Feed quality is no longer about avoiding disapprovals—it determines whether AI agents even consider your products.
The UCP Readiness Checklist
This checklist is organized into four categories, ranked by priority. Start with the must-have items and work your way through to the UCP-specific requirements. Every item you address improves your chances of being selected by AI shopping agents.
A. Core Feed Quality (Must-Have)
These are the foundational elements every product in your feed needs. Without them, you will not perform well in traditional Shopping ads, let alone agentic commerce.
- Product titles that are descriptive, keyword-rich, and include brand + key attributes. A title like "Running Shoes" tells the AI nothing. "Nike Air Zoom Pegasus 41 Men's Running Shoes — Black/White, Size 10" gives the agent everything it needs to match the product to a shopper's request. Follow product title optimization best practices and include brand, product type, key differentiators, and variant attributes.
- Complete product descriptions that go beyond title repetition. Your description should explain what the product is, who it is for, and why it is worth buying. AI agents parse descriptions to understand product context and match them against nuanced shopper queries like "comfortable shoes for long-distance running on pavement."
- Correct GTINs and MPNs for every applicable product. Global Trade Item Numbers and Manufacturer Part Numbers are how AI agents verify product identity and compare products across merchants. Missing GTINs reduce your product's credibility in the eyes of the algorithm.
- Accurate pricing that matches your landing page. Pricing parity between your feed and your website is already a Google Shopping requirement. Under UCP, pricing accuracy becomes even more critical because the checkout happens inline—any discrepancy will cause a failed transaction and damage your trust score.
- Real-time inventory and availability sync. AI agents check stock status before recommending a product. If your feed says "in stock" but the product is actually sold out, the agent learns to deprioritize your catalog. Aim for real-time sync through the Content API or at minimum hourly updates.
- High-quality images with multiple angles and white backgrounds. While AI agents primarily process structured text data, product images still matter for the shopping experiences those agents create. Include your primary image plus additional angles. Follow Google's image requirements for resolution and formatting.
- Proper Google Product Category mapping. Correct Google Shopping category assignments help AI agents understand where your product fits in the broader product taxonomy. Incorrect or missing categories mean your product may never surface for relevant queries.
B. New Conversational Attributes (High Priority)
These are the new Merchant Center attributes Google announced alongside UCP. They are designed to give AI agents the contextual information needed for conversational product recommendations. Most merchants have not implemented these yet, which means early adopters have a competitive advantage.
- Product Q&A fields. Structure common questions and answers about your product as structured data. Think about the questions shoppers actually ask: "Is this waterproof?" "Does it work with X?" "What is the warranty?" AI agents use these Q&A pairs to answer shopper questions in real time without leaving the conversation.
- Compatible accessories and "works with" relationships. Define which products in your catalog work together. If you sell cameras, link compatible lenses, memory cards, and bags. AI agents use these relationships to make bundle recommendations and answer compatibility questions on the spot.
- Substitutes and alternatives. Identify products that serve as alternatives to each other. This helps AI agents offer options when a shopper's first choice is unavailable or when the agent wants to present a comparison. Merchants who provide this data give the agent more reasons to recommend products from their catalog.
- Usage scenarios and "best for" contexts. Describe the situations and use cases your product excels in. "Best for trail running in wet conditions" or "Ideal for small home offices under 200 sq ft." These scenario descriptions map directly to how shoppers describe their needs to AI agents—conversationally and contextually.
Early Adopter Advantage
Most merchants have not yet added conversational attributes to their feeds. Implementing them now positions your products ahead of competitors when AI agents start weighing these attributes in their recommendation algorithms.
C. Data Accuracy and Freshness
Stale data is one of the fastest ways to lose credibility with AI shopping agents. Unlike human shoppers who might tolerate a slight delay in stock updates, AI agents make binary decisions based on the data available at query time.
- Inventory sync frequency: aim for real-time or at least hourly. Daily feed updates were acceptable for traditional Shopping ads. For agentic commerce, daily is too slow. Every hour your inventory is out of sync is an hour where AI agents might recommend a product you cannot fulfill—or skip a product you have back in stock.
- Price accuracy: any mismatch equals disapproval in traditional Shopping and irrelevance in UCP. If your feed price says $49.99 but your site shows $54.99, Google Shopping will disapprove the product. In UCP, the mismatch breaks the inline checkout experience entirely. Use automated price monitoring to catch discrepancies before they impact your feed.
- Product availability: out-of-stock items waste AI agent time and reduce your trust score. Every time an AI agent recommends your product and the transaction fails due to stock issues, your reliability score decreases. The agent will learn to prioritize competitors who consistently deliver accurate availability data.
- Shipping details: clear, accurate, and competitive. AI agents consider shipping speed and cost when making recommendations. If a shopper asks for "next-day delivery options," the agent can only include your product if your feed has accurate shipping data. Include shipping costs, delivery timeframes, and free shipping thresholds.
D. UCP-Specific Requirements
Beyond feed quality and conversational attributes, UCP has several technical requirements you must meet to participate. According to the UCP specification, these are non-negotiable.
- Add the
native_commerceattribute to your product feed. This attribute signals that your products support UCP-enabled transactions. Without it, AI agents will not consider your products for in-conversation checkout, regardless of how good your other feed data is. - Maintain an active Google Merchant Center account in good standing. Your account must be verified, have no critical policy violations, and be in compliance with Merchant Center guidelines. Existing suspensions or account-level issues block UCP participation entirely.
- US-based operations (currently). As of early 2026, UCP is available to US-based merchants. International expansion is expected, but for now, your business must operate within the United States to participate.
- Payment service provider that accepts Google Pay tokens. UCP's inline checkout processes payments through Google Pay. Your payment processor must support Google Pay token-based transactions. Check with your PSP to confirm compatibility—most major processors (Stripe, Adyen, Braintree) support this.
- Products must be eligible for checkout. Each product you want available through UCP must have complete checkout eligibility—accurate pricing, valid shipping configurations, return policy information, and tax settings configured in Merchant Center.
Technical Prerequisite
The native_commerce attribute is the single most important UCP-specific requirement. Without it, none of your other feed optimizations matter for agentic commerce. Add it to your feed as a first step.
Common Feed Gaps That Kill UCP Visibility
Based on common Merchant Center feed errors and diagnostics data, these are the most frequent feed gaps that will prevent your products from performing in agentic commerce:
- Missing GTINs (affects 20%+ of feeds). This is the single most common feed issue. Many merchants leave GTINs blank, especially for private-label or niche products. For branded products where a GTIN exists, leaving it empty tells the AI agent your data is incomplete and less trustworthy than a competitor who provides it.
- Generic titles that do not differentiate products. Titles like "Blue T-Shirt" or "Wireless Headphones" are too vague for AI agents to match against specific shopper queries. Agents will always prefer products with descriptive, attribute-rich titles over generic ones.
- Stale pricing data from sync delays. If your feed updates once a day and you change prices mid-afternoon, there is a window where your feed data is wrong. For UCP, where checkout happens inline, stale pricing creates failed transactions and erodes trust.
- Missing or poor-quality images. Products with a single low-resolution image receive less favorable treatment than products with multiple high-quality images. AI-generated shopping experiences use images to build product cards and comparison views.
- No product category or incorrect category mapping. Products without a Google Product Category or with an incorrect category assignment are harder for AI agents to classify. If the agent cannot confidently categorize your product, it is less likely to surface it in relevant conversations.
- Zero supplemental attributes. If your feed contains only the bare minimum required attributes—title, description, price, image, availability—you are giving the AI agent nothing to work with for conversational recommendations. No Q&A data, no compatibility information, no usage scenarios means the agent has no context beyond the basics.
The Compounding Problem
Feed gaps do not exist in isolation. A product with a generic title and a missing GTIN and no supplemental attributes is exponentially less likely to be selected by an AI agent than a product with just one of those issues. Each gap compounds the visibility problem.
How to Audit Your Feed
Here is a step-by-step process to assess your current feed's UCP readiness and build a prioritized action plan:
- Export your current Merchant Center feed. Download a full export of your product data from Merchant Center. This gives you a complete snapshot to work with. If you use a feed management tool, export from there as well to compare what your tool sends versus what Merchant Center receives.
- Check attribute completeness. For every product, verify that required attributes are populated. Aim for 95%+ completion on required fields (title, description, price, availability, image link, GTIN/MPN, Google Product Category, brand). Track completion rates in a spreadsheet so you can measure improvement over time.
- Review disapproval reasons in Merchant Center diagnostics. Go to Merchant Center > Diagnostics and review all active product issues. Each disapproval or warning is a feed gap that needs fixing. Prioritize errors that affect your highest-revenue products first.
- Identify products with missing optional attributes. Export a report of which products lack optional but valuable attributes: additional images, product detail fields, product highlights, energy efficiency ratings, and other supplemental data. These "optional" attributes become far more important in the context of agentic commerce.
- Cross-reference with new UCP conversational attributes. Check which of your products have the new conversational attributes (Q&A, compatibility, substitutes, usage scenarios). For most merchants, this number will be close to zero—and that is your biggest opportunity.
- Prioritize high-value products first. You do not need to optimize your entire catalog at once. Start with the products that generate the most revenue and work outward from there. This 80/20 approach ensures your most important products are UCP-ready first.
Prioritize What Matters Most
Tools like SKU Analyzer help you identify which products in your catalog drive the most revenue and should be prioritized for feed enrichment. Starting with your top 20% of products—the ones generating 80% of revenue—is the most efficient approach to UCP readiness.
Frequently Asked Questions
What is the native_commerce attribute in Merchant Center?
The native_commerce attribute is a new product feed field that signals to Google your products are eligible for Universal Commerce Protocol (UCP) transactions. Adding this attribute tells AI shopping agents that your product supports seamless, in-conversation checkout without redirecting the customer to your website. It is a required attribute for any merchant who wants their products discoverable through agentic commerce channels.
Do I need a feed management tool for UCP?
Not necessarily, but a feed management tool makes UCP readiness significantly easier. UCP requires high data quality, real-time inventory sync, new conversational attributes, and pricing parity across your feed and landing pages. If you have a small catalog and can maintain these standards manually through Merchant Center, a separate tool is optional. For larger catalogs or multi-channel sellers, a feed tool like DataFeedWatch, Channable, or Feedonomics can automate much of the compliance work.
How often should I update my product feed for agentic commerce?
For agentic commerce, you should aim for real-time or at minimum hourly feed updates. AI shopping agents check inventory and pricing before recommending products, so stale data leads to failed transactions and reduced trust scores. Price changes, stock-outs, and new product additions should be reflected in your feed as quickly as possible. Google's Content API for Shopping supports real-time updates and is the recommended approach for UCP-eligible merchants.
Will my existing Google Shopping feed work with UCP?
Your existing Google Shopping feed provides a foundation, but it almost certainly needs enhancements for UCP. Traditional Shopping feeds focus on titles, descriptions, images, pricing, and availability. UCP additionally requires the native_commerce attribute, conversational attributes like product Q&A fields, compatibility and substitution data, usage scenario descriptions, and a payment provider that accepts Google Pay tokens. Think of your Shopping feed as about 60-70% of what UCP needs—the remaining 30-40% involves new attributes and stricter data quality standards.
Conclusion: Feed Quality Is Your Competitive Moat
Feed quality has always mattered for Shopping ads. Better data leads to better ad relevance, fewer disapprovals, and stronger campaign performance. That has been true for years and remains true today.
With UCP, the stakes are fundamentally higher. Your product feed becomes the single most important factor determining whether AI agents discover, evaluate, and recommend your products—and now that shopping ads appear directly inside AI Mode, feed quality also determines your paid ad visibility in this new surface. There is no fallback to your website, no chance for a well-designed product page to compensate for thin feed data. The feed is the experience.
The good news is that the steps to prepare are concrete and actionable. Start with core feed quality—complete titles, accurate GTINs, real-time pricing and inventory. Then layer on the new conversational attributes that give AI agents the context they need. Add the UCP-specific technical requirements. And prioritize your highest-revenue products so the products that matter most are ready first.
Merchants who treat their product feed as a strategic asset—not just a compliance requirement—will be the ones who thrive in agentic commerce. The checklist in this guide gives you a clear path to get there.
SKU Analyzer helps you identify which products drive the most revenue and which ones waste budget, so you can focus your feed optimization efforts where they have the greatest impact. When every product in your catalog cannot get equal attention, knowing which ones deserve it most is the difference between efficient preparation and wasted effort.