The landscape of e-commerce is undergoing a seismic shift, ushering in an era where transactions can occur entirely outside of traditional websites, rendering website visits optional, and potentially obsolete. This evolution, driven by the rapid advancements in artificial intelligence, presents a significant challenge to e-commerce and SEO teams still operating under outdated paradigms. At the heart of this transformation lies "agentic commerce," a concept where the entire customer journey, from initial discovery to final checkout, can be seamlessly managed by AI. The pace of this development is nothing short of breathtaking.
A Rapidly Evolving Ecosystem: Protocols and Pivots
The genesis of agentic commerce can be traced to significant announcements in late 2025 and early 2026. In September 2025, OpenAI and Stripe unveiled their Agentic Commerce Protocol (ACP). This groundbreaking initiative empowered users to not only research and select products within a conversational AI interface but also to complete single-item transactions directly through "Instant Checkout" within ChatGPT. This development marked a significant step towards a frictionless purchasing experience, blurring the lines between information retrieval and immediate commerce.
Just months later, in January 2026, Google responded with its own ambitious initiative: the Universal Commerce Protocol (UCP). Announced via a company blog post, UCP was presented as a new open standard designed to facilitate agentic commerce across the entire shopping lifecycle, encompassing discovery, purchasing, and even post-purchase support. This parallel development by two of the tech industry’s giants underscored the growing momentum and strategic importance of this new paradigm.
However, the agentic commerce space is not static. In a notable pivot, OpenAI announced in March 2026 that it was stepping back from its initial "Instant Checkout" feature within ACP. Citing a desire for "greater flexibility," OpenAI indicated a renewed focus on product discovery within ChatGPT, allowing brands to gain valuable visibility while giving retailers more autonomy in integrating their own checkout solutions. This strategic adjustment suggests a recognition of the complexities involved in direct transaction integration and a potential move towards a more collaborative ecosystem.
Simultaneously, Google continued to refine its UCP offering. That same March, the company released an update that significantly enhanced its capabilities, introducing new features for checkout processes and catalog management, including the crucial addition of shopping cart functionality. These ongoing developments highlight a dynamic and competitive race to define the future of AI-powered commerce, with key players adapting their strategies based on market feedback and technological progress. The overarching lesson from these rapid developments is clear: businesses must remain vigilant and adaptable to navigate this evolving terrain.
The End of Traffic-Centric E-commerce
For decades, the bedrock of e-commerce success has been website traffic. Retailers have invested heavily in SEO, content marketing, email campaigns, social media, and promotional offers, all with the primary goal of driving potential customers to their product pages. While conversion rates, personalization, and cross-selling have always been crucial, their effectiveness has been contingent upon a user landing on the site in the first place. This "traffic-first" mentality has shaped the digital retail landscape, dictating strategies and resource allocation.
Agentic commerce fundamentally disrupts this model. The ability for AI agents to facilitate the entire customer journey, from the initial spark of interest to the final purchase, means that a direct visit to a retailer’s website may no longer be a prerequisite. This shift necessitates a complete re-evaluation of traditional e-commerce strategies.
A compelling illustration of this new reality came from Target, which, on the same day as Google’s UCP announcement, published a press release detailing their vision for agentic shopping experiences. The hypothetical scenario described a customer interacting with AI, seeking "cute and affordable floral leggings, in a light color" for working out. The expectation is that the AI would present options and facilitate a purchase directly within the chat interface, bypassing the need to navigate Target’s website. This vision underscores the critical question for retailers: how can they ensure their products are part of the AI’s recommendation set?
Understanding the Mechanics of Agentic Commerce
Agentic commerce operates on distinct principles compared to traditional search engine optimization or even AI content citations. It is not about achieving a specific ranking that might still be visible to users who scroll down a results page. Instead, in agentic commerce, a brand is either included or excluded from the AI’s consideration set. Furthermore, the SEO efforts focused on increasing brand visibility through AI mentions or citations are unlikely to directly translate into success within agentic commerce protocols.
The core difference lies in data acquisition. Instead of crawling customer-facing website content, agentic commerce protocols like ACP and UCP primarily draw information from merchant feeds or product feeds and on-page schema markup. This fundamental shift transforms the challenge from one of content optimization to one of robust data management and integration. The AI agents act as conduits, relaying product data to consumers and transaction data back to retailers.
This reliance on structured data has two critical implications. Firstly, both protocols maintain the retailer as the merchant of record, meaning companies like OpenAI and Google are not acting as resellers in the vein of a traditional marketplace. Secondly, for agentic transactions to be successful, these protocols require detailed, accurate, and up-to-the-minute product information. This is a crucial point, as a significant number of retailers currently fall short in this regard.
Many retailers currently treat their product or merchant feeds as a secondary concern, primarily used for updating dynamic ad campaigns with basic product details such as name, image, and promotional offers. This is because traditional ad campaigns are still geared towards driving traffic back to the website where the transaction ultimately occurs. However, for AI agents to complete transactions on behalf of retailers, these feeds must be far more comprehensive.
Critical Schema Fields for Agentic Commerce Readiness
Google’s UCP documentation highlights specific data signals that are paramount for AI agents like Google Gemini when evaluating product recommendations. Beyond basic transactional metadata, three additional attributes are identified as crucial:
- Product Availability: Real-time stock levels are essential. Inaccurate availability information can lead to failed transactions and a negative customer experience.
- Shipping Information: Details regarding shipping costs, delivery times, and shipping zones are critical for customers making purchasing decisions.
- Return Policy: Clear and accessible information about return windows, conditions, and processes is a significant factor in consumer confidence.
Absence or incompleteness in any of these fields can result in a product being entirely excluded from an AI agent’s recommendations, not merely ranked lower. This exclusion is absolute. Beyond these core attributes, speed of data retrieval is also paramount. AI agents are designed for efficiency, and slow API responses can deter them from utilizing a retailer’s feed in the future. Similarly, stale or inaccurate data, such as an item being listed as in stock when it is not, can lead to failed transactions and damage the perceived reliability of a retailer’s feed. This can prompt AI agents to favor products from brands with more trustworthy data sources.
Research Uncovers Significant Retailer Preparedness Gaps
A recent audit examining the agentic commerce readiness of top retailers has revealed stark disparities. The study analyzed 207 high-traffic product detail pages (PDPs) from 29 leading retailers, identified by their organic search traffic volume. After accounting for inaccessible pages and non-PDP content, 141 PDPs were audited against a 10-point UCP-readiness rubric derived from Google’s UCP documentation.
The findings indicate a mixed picture. On the positive side, nearly all audited PDPs successfully provided basic transactional information such as product name, description, and price. Furthermore, all 141 PDPs appeared to have implemented the product schema recommended by Google since 2014.
However, the audit revealed significant shortcomings concerning the more recently introduced, yet critical, schema fields. Approximately 70% of the audited PDPs failed to include all three of the most important attributes (availability, shipping information, and return policy), which were integrated into Schema.org between 2020 and 2021. This suggests that many retailers have not updated their schema templates to reflect the evolving demands of agentic commerce.
Another notable deficiency was the absence of a Global Trade Item Number (GTIN) in 65% of the PDPs. While a missing GTIN may not entirely prevent an AI from using a retailer’s feed, it severely hampers the AI agent’s ability to compare offerings across different retailers. Without a GTIN, an AI agent cannot definitively identify Product X from one retailer as the same physical item as Product X from another. This limitation prevents crucial price comparisons, potentially excluding a retailer from offering the best price when a customer inquires about it.
Platform Agnosticism and Configuration Challenges
The research also highlighted that readiness for agentic commerce is largely a matter of data configuration rather than the underlying e-commerce platform. Only three brands achieved an average score of 8 or higher across their top PDPs: Uplift Desk (Custom platform), Carbon38 (Shopify), and Sigma Beauty (Shopify). These top performers did not require platform overhauls but rather reconfigured their existing Content Management Systems (CMS) to expose the necessary additional data fields. This indicates that the barrier to entry is less about technological infrastructure and more about operational processes and data management practices. The primary takeaway is that addressing incomplete feeds and schema is achievable with focused effort and should not be an insurmountable expense.
An alarming finding was that 15% of the initially audited URLs were completely inaccessible to the audit tools due to HTTP 403 Forbidden errors. This included PDPs from major brands like Adidas UK, UGG, Converse, and Christian Louboutin. These websites had implemented bot-detection measures that prevented data scraping. While blocking bots can be a strategic decision to protect against competitor price scraping or other forms of data exploitation, these same defenses can inadvertently block AI agents from accessing critical product data for agentic commerce. This raises the question of whether such exclusions are conscious business decisions or unforeseen consequences of overzealous bot protection.
Category SEO’s Diminishing Relevance in Agentic Commerce
The audit methodology prioritized high-traffic product detail pages (PDPs) and excluded category pages (Product Listing Pages or PLPs). This exclusion was deliberate, as the research indicated a significant shift in relevance. Historically, category pages have often driven more organic search traffic than individual PDPs, with some retailers seeing a traffic ratio of 10:1 in favor of category pages. Consequently, much SEO effort has been historically directed towards optimizing these pages.
However, agentic commerce protocols largely disregard category pages because they typically lack the necessary structured data (schema). AI agents primarily process information at the product level. Therefore, a retailer that ranks exceptionally well for "men’s wax jacket" in traditional search results may be entirely overlooked by AI agents seeking to facilitate a direct purchase. This necessitates a strategic reorientation of SEO efforts away from broad category optimization and towards granular product data enrichment.
Addressing the Data Plumbing: Operationalizing Agentic Commerce Readiness
The data required for agentic commerce—shipping delivery times, return windows, price validity dates, GTINs—likely already exists within a retailer’s various internal systems, such as Enterprise Resource Planning (ERP), Product Information Management (PIM), or inventory management platforms. The critical missing element is the "plumbing" to efficiently and accurately transfer this information into merchant feeds and on-page schema, and to maintain its currency.
To achieve this, operational changes are imperative:
-
Treat the Merchant Feed as Essential Infrastructure: Retailers must elevate their merchant feeds from a secondary function to a core infrastructure component. This involves auditing existing feeds for completeness, particularly focusing on the three key UCP selection signals in schema for top-selling products. Setting achievable targets for optimizing products for agentic commerce on a monthly or quarterly basis is crucial.
-
Enhance Inventory Data Accuracy and Timeliness: The reliability of AI agents is directly impacted by the accuracy and speed of data retrieval. Stale, incomplete, or inaccurate inventory data can lead to failed transactions, degrading the perceived reliability of a retailer’s feed. Configuring feeds for near real-time updates, ideally with sub-hour or even sub-minute granularity, is the safest approach.
-
Prepare for Loyalty Programs and Dynamic Pricing: While early agentic commerce offerings featured static pricing, recent developments, such as Google’s UCP update in March 2026, have introduced "Identity Linking." This capability allows AI agents to interact with a retailer’s website on behalf of a customer, enabling access to loyalty benefits, personalized offers, wishlists, and authenticated checkouts. Implementing these features, while requiring additional effort, unlocks significant potential for personalized and dynamic customer experiences.
SEO Requires Supply Chain Thinking
For decades, e-commerce SEO has been built on a foundation of PDPs, categories, and content. Agentic commerce introduces a fundamentally different approach, one that requires a "supply chain" mindset rather than a traditional optimization strategy. Unlike conversational AI that prioritizes insights and thought leadership, agentic commerce protocols are indifferent to content quality or category performance. Their sole focus is on clean, accurate, and complete product data.
This necessitates a shift towards acquiring necessary component data from various systems, applying quality control, and packaging it for efficient delivery. This process demands cross-functional ownership and clearly defined deliverables. While SEO teams typically manage brand visibility, the merchant team often oversees product and merchant feeds, and IT or e-commerce platform teams are responsible for system integration and data synchronization.
Crucially, no single team is likely to possess the authority or budget to drive these necessary changes independently. The Chief Marketing Officer (CMO) must champion this initiative at the board level, articulating the direct correlation between data infrastructure completeness and commercial outcomes in the AI-driven marketplace. The retailers who succeed in agentic commerce will not be those with the loudest brand voices or the highest-ranking pages, but those who make their products the easiest for an AI agent to purchase. This transition signifies a profound evolution, demanding a strategic realignment of priorities and a collaborative, data-centric approach to navigating the future of retail.
