The Era of Agentic Commerce: How Commerce is Redefining Product Discovery through AI Enrichment

In the rapidly evolving landscape of digital retail, the battle for consumer attention is no longer fought solely on search engine result pages (SERPs) or through traditional social media advertising. Instead, the frontier of ecommerce has shifted toward "agentic commerce"—an ecosystem where artificial intelligence agents act as intermediaries, performing research, comparing products, and ultimately making purchases on behalf of users.

Recognizing that the efficacy of these AI agents is entirely dependent on the quality of the underlying product data, Commerce—the parent company of BigCommerce, Feedonomics, and Makeswift—has unveiled two sophisticated AI-driven tools: Feedonomics Enrichment and BigCommerce Catalog Enrichment. These solutions are designed to bridge the gap between static product catalogs and the dynamic, query-driven world of generative AI.

The Foundation of AI Discovery: Why Data Quality Matters

The core premise behind Commerce’s new offering is simple yet profound: AI agents can only be as helpful as the data they are provided. If a product’s metadata is sparse, inconsistent, or lacks context, an AI agent—whether it be a virtual shopping assistant on a brand’s website or a general-purpose tool like ChatGPT or Claude—will fail to present that product effectively to the consumer.

"AI agents can only answer questions about your products as well as your data allows," says Sharon Gee, senior vice president of product for AI at Commerce. "Whether you’re a marketer driving AI-based Engine Optimization (AEO) or a product leader building shopping agents, better data is the foundation for better agentic experiences."

The launch of these enrichment tools marks a strategic pivot for Commerce, aiming to provide a data pipeline that ensures product information is optimized for performance across a fragmented digital landscape. From Google and OpenAI’s answer engines to niche B2B procurement portals, these tools aim to standardize and enrich product data at scale.

Chronology: The Shift from Traditional Search to Agentic Shopping

To understand the necessity of this innovation, one must look at the recent evolution of online commerce:

1. The Era of Keyword Search (2000s–2010s): Retailers focused on SEO, ensuring their product descriptions contained the right keywords to rank high on search engines like Google.
2. The Rise of Marketplaces (2010s–2020s): Merchants began dealing with the complexity of "one-to-many" distribution, syndicating product data across Amazon, eBay, and social commerce platforms.
3. The Agentic Turn (2023–Present): With the advent of Large Language Models (LLMs), the consumer experience has shifted toward conversational interfaces. Consumers now ask AI to "find me a pair of high-performance running shoes under $150 that are suitable for flat feet and come in a sustainable material."

The challenge for retailers is that these models do not browse websites in the way humans do; they ingest "knowledge" from structured data. If a retailer’s catalog is not configured to answer these nuanced questions, they are effectively invisible to the new generation of AI-driven shoppers. Commerce’s new tools were designed specifically to address this technical bottleneck.

Supporting Data: The Scale of the Commerce Ecosystem

Commerce’s influence in the market is substantial, providing the technological backbone for a significant portion of North American retail. According to the Digital Commerce 360 Top 2000 Database, retailers utilizing Commerce as their ecommerce platform generated more than $538 billion in online sales in 2025.

This massive volume of transaction data serves as a feedback loop for the company’s AI strategy. By analyzing the behavior of both B2B and B2C buyers, Commerce identified a recurring pain point: the sheer labor intensity of maintaining "clean" data across disparate channels. As Gee noted in a conversation with Digital Commerce 360, the businesses utilizing their tools are often deeply entrenched in enterprise retail and B2B sectors, where product catalogs can span tens of thousands of items, each with varying specifications, language requirements, and regional compliance standards.

The Mechanics of Enrichment: How the Tools Operate

The Feedonomics and BigCommerce Catalog Enrichment tools function as an automated layer of intelligence that sits atop a merchant’s existing database. Rather than requiring manual input, the tools perform the following operations:

• Semantic Enhancement: The tools generate rich, descriptive facts and snippets that move beyond simple keyword tags. This allows AI platforms to interpret not just what a product is, but its specific utility, target audience, and relative value.
• Question-and-Answer (Q&A) Generation: By predicting the types of questions a consumer might ask, the system automatically populates fields that answer these queries in a natural, conversational format.
• Cross-Channel Standardization: The tools eliminate the need for manual CSV exports or fragmented third-party plugins. By creating a single "source of truth," merchants ensure that the product details found on their own storefront are identical to those retrieved by an AI agent on Perplexity or Gemini.
• Automated Localization: For global brands, the ability to translate and adapt product catalogs across 15+ languages—and then syndicate those translations to diverse AI answer engines—is a massive operational undertaking. Commerce’s tools handle this "one-to-many" problem, ensuring that the brand’s messaging remains consistent, regardless of the language or the AI interface.

Implications for B2B and Hybrid Businesses

While much of the focus on AI in ecommerce has centered on consumer-facing "personal shoppers," the most significant productivity gains may actually be occurring in the B2B sector.

"It’s fairly trivial to sell things online anymore," Gee notes. "That’s a well-solved problem over the past two decades. There are many use cases that have not been solved well yet."

One such use case is the processing of complex, high-volume purchase orders. In traditional B2B environments, procurement officers often handle PDF-based purchase orders containing thousands of line items. Manually entering this data into an ERP system is a time-consuming, error-prone task that offers zero strategic value.

Commerce’s "Purchase Order Agent" addresses this directly. By allowing users to drag and drop a PDF into the system, the AI agent can parse the document, identify the items, check inventory availability, and automatically build a cart for approval. This shifts the role of the human employee from "data entry clerk" to "strategic oversight," allowing them to focus on relationship management and complex negotiations rather than rote administration.

The Future of the "Composable" Tech Stack

As the digital retail environment becomes increasingly complex, businesses are moving toward composable commerce—a modular approach where merchants select best-of-breed components to build their tech stack. Commerce’s strategy of offering both self-managed tools (BigCommerce) and managed-service models (Feedonomics) reflects this need for flexibility.

By positioning its enrichment tools as the "data pipeline" for the future of AI, Commerce is effectively betting that the next decade of ecommerce success will be defined by data accessibility. For retailers, the implication is clear: the cost of inaction is growing. As consumers and businesses increasingly rely on AI agents to navigate the marketplace, companies that fail to "teach" their product data to these agents will find themselves locked out of the next generation of commerce.

Conclusion

The release of Feedonomics Enrichment and BigCommerce Catalog Enrichment is more than a simple product update; it is an acknowledgement of a fundamental shift in how the world interacts with the internet. We are moving away from a "search-and-click" model toward an "ask-and-receive" model.

For brands operating on the Commerce platform, these tools offer a path to future-proof their operations. By automating the enrichment process and facilitating the seamless integration of their catalogs into the broader AI ecosystem, Commerce is enabling its merchants to remain relevant in an era where the most important shopper might not be a human at all, but an AI agent acting on their behalf.

As the technology continues to mature, the gap between those who have invested in "agent-ready" data and those who have not will likely become the primary differentiator in the hyper-competitive world of global ecommerce. For Commerce and its vast network of enterprise clients, the transition to agentic commerce has officially begun.