The example shown in Google’s AP2 framework (see: video link) offers a glimpse into the emerging world of agentic shopping.

Looking ahead, it’s clear that shopping will increasingly shift toward AI-powered shopping agents. For younger consumers in particular, or in scenarios where traditional online shopping falls short—such as complex bundled purchases (outfits, flight + hotel packages) or time-consuming tasks (research, price comparisons)—shopping agents will become the natural entry point for traffic.

This shift will fundamentally change how merchants present their products. Instead of designing for human consumers, they’ll need to design for shopping agents. Whether the standard becomes Google’s A2A, MCP, or another protocol, the reality is that merchants will face a new kind of “storefront redesign”—this time not in HTML, but in agentic interfaces. And it won’t just be a handful of merchants: it will be an industry-wide transformation, comparable to the transition from websites to apps in the last era of e-commerce.

This transformation brings enormous demand, which we see unfolding in three stages:

1. Merchant-driven agentic storefronts.

Merchants will need technical support to complete this transition. Some will build in-house IT teams, while others will look for solution providers. The opportunity here is to provide a generalizable SaaS service—tools that allow merchants to easily build agentic storefronts while integrating seamlessly with their existing backends (product/ordering systems, CRM, logistics, etc.). Think of this as a “Shopify for agentic shopping.”

2. Low-friction interim solutions.

Change won’t happen overnight. Most merchants will wait for the market to mature before committing fully. In the meantime, they’ll need lightweight solutions to capture agent-driven traffic with minimal investment. A platform provider could step in here—extracting product data directly from merchants’ existing websites/apps and handling agentic traffic on their behalf. For merchants, this is almost invisible. Today, BorderXLab’s CloudStore AI already provides such a model.

3. Aggregators of merchant data.

In an agentic shopping world, consumers will still want to compare across multiple merchants. While shopping agents can query multiple merchants, this is inefficient. A more effective approach is to aggregate merchant data in advance, index it, and optimize search/retrieval to boost conversion. This allows faster, smarter interactions between agents and platforms—again, similar to how BorderXLab’s CloudStore AI works today.

At BorderXLab, we believe agentic shopping is an inevitable trend, and we’re fully committed to embracing it. Our AI strategy follows four steps:

1. Build shopping agents to participate in the emerging ecosystem. Our current focus is on scenarios that traditional e-commerce cannot handle well—such as multi-item outfits, styling recommendations, and sizing advice. Our first shopping agent has been launched on BeyondStyle.us to help accelerate growth.
2. Scale the CloudStore AI product to support more shopping agents. We’re actively developing an MCP interface to improve integration.
3. Enhance CloudStore AI capabilities—integrating agentic payments, onboarding more merchants, and advancing search algorithms—evolving toward meeting both interim and aggregator needs.
4. Launch SaaS services for merchants at the right time, offering them the tools to build their own agentic storefronts.

The shift to agentic shopping is just beginning, and BorderXLab is ready to help shape this new era.