Wall Street just priced agentic commerce into Shopify twice in one week.

𝗪𝗵𝗮𝘁 𝘁𝘄𝗼 𝗕𝘂𝘆 𝘂𝗽𝗴𝗿𝗮𝗱𝗲𝘀 𝗶𝗻 𝗼𝗻𝗲 𝘄𝗲𝗲𝗸 𝗮𝗰𝘁𝘂𝗮𝗹𝗹𝘆 𝘀𝗶𝗴𝗻𝗮𝗹

Analyst upgrades are forward-looking instruments. When BofA and Stifel both raise a platform on the same thesis in the same week, they are not describing what the platform does today. They are betting on what the ecosystem around the platform will demand and fund over the next 12 to 24 months. Agentic commerce – AI systems that discover, evaluate, and complete purchases autonomously on a consumer’s behalf – is being priced as that next demand cycle.

Visa accelerating that same timeline is a meaningful corroboration. Visa announced partnerships with European banks and merchants this week to deploy agentic commerce infrastructure at payment-rail level. When the payments network is building the transaction substrate for AI-driven purchasing, the commercial viability of agentic commerce is no longer speculative. The rails are being laid. The analyst community is pricing the traffic those rails will carry.

The relevant implication for merchants is not “Shopify’s stock is going up.” It is that the ecosystem around Shopify is about to experience real capital flow directed at making agentic commerce functional at the merchant level. That means tooling, infrastructure, and integrations – and it means merchants who have not yet made their catalogs legible to AI agents will feel the competitive pressure sooner than most of them expect.

𝗧𝗵𝗲 𝗱𝗶𝘀𝘁𝗮𝗻𝗰𝗲 𝗯𝗲𝘁𝘄𝗲𝗲𝗻 𝗽𝗹𝗮𝘁𝗳𝗼𝗿𝗺 𝗰𝗮𝗽𝗮𝗯𝗶𝗹𝗶𝘁𝘆 𝗮𝗻𝗱 𝗺𝗲𝗿𝗰𝗵𝗮𝗻𝘁 𝗿𝗲𝗮𝗹𝗶𝘁𝘆

Here is where the analyst narrative and the operational reality diverge, and where the real merchant story lives.

Shopify expanding its AI tooling across the global merchant platform is a real and ongoing development. But platform-level investment does not automatically translate to storefront-level capability for individual merchants. Shopify’s native storefront search is still largely keyword-based. Its agentic and conversational features – the NLP shopping agent, the optional visual search surface arriving in 2026 – raise the baseline expectation. They do not deliver an integrated discovery engine to every merchant’s own storefront by default.

This distinction matters enormously. When Wall Street prices agentic commerce into Shopify’s valuation, they are pricing the platform’s infrastructure: Shopify’s ability to route agentic traffic, process AI-driven transactions, and support developers building on top of its APIs. What they are not pricing is whether any individual DTC fashion brand on Shopify actually has a trimodal discovery engine – natural language, precise color, image similarity – running on their storefront. That gap between platform ceiling and merchant floor is where the real commercial opportunity sits right now.

Digital Commerce 360 and ReFiBuy launched the first formal AI Commerce Rankings framework this week, designed to benchmark retailer readiness for AI-driven shopping. The initiative is significant precisely because it acknowledges that merchant readiness is unevenly distributed. A platform upgrade benefits the merchants who are prepared to capture agentic traffic. It does not prepare them automatically.

𝗔𝗜 𝘁𝗿𝗮𝗳𝗳𝗶𝗰 𝗮𝗿𝗿𝗶𝘃𝗲𝘀 𝗽𝗿𝗶𝗺𝗲𝗱. 𝗗𝗼𝗲𝘀 𝘆𝗼𝘂𝗿 𝘀𝘁𝗼𝗿𝗲𝗳𝗿𝗼𝗻𝘁 𝗺𝗮𝘁𝗰𝗵 𝘁𝗵𝗲 𝗲𝘅𝗽𝗲𝗰𝘁𝗮𝘁𝗶𝗼𝗻?

Similarweb data cited this week by Modern Retail shows that AI-driven traffic to Amazon more than doubled over six months, reaching roughly 13.9 million visits in June. More striking is the indirect attribution figure: 28.53% of MacBook buyers across major retailers had a category-relevant AI chat session in the 30 minutes before purchasing. The direct traffic number understates AI’s role by a wide margin. Consumers are arriving at storefronts already shaped by a conversational research session with an LLM.

This is the behavioral shift that should be driving merchant infrastructure decisions right now. A shopper who has spent 20 minutes describing what they want to ChatGPT or Gemini arrives on a storefront with high intent and a very specific mental model of the product they are looking for. If the storefront can only respond with a keyword search bar, the mismatch is immediate. The shopper has already been operating in natural language. Dropping back to a keyword input is a regression in the experience, not a continuation of it.

This is not an argument that on-site search is being replaced by off-site AI surfaces. It is the opposite argument. On-site discovery needs to be upgraded to match the same describe-and-show paradigm the shopper just experienced off-site. The two fronts are complementary: off-site legibility gets you found; on-site trimodal discovery – natural language query handling, image similarity, precise color matching – is where the sale closes and where the margin stays with the merchant rather than with the platform or marketplace.

Journey Further’s analysis of the AI search era reinforces the data infrastructure side of this same coin. 𝗧𝗵𝗲𝗶𝗿 𝗳𝗶𝗻𝗱𝗶𝗻𝗴 𝘁𝗵𝗮𝘁 𝗹𝗼𝗻𝗴-𝘁𝗮𝗶𝗹 𝗾𝘂𝗲𝗿𝗶𝗲𝘀 𝗼𝗳 𝗳𝗶𝘃𝗲 𝗼𝗿 𝗺𝗼𝗿𝗲 𝘄𝗼𝗿𝗱𝘀 𝗴𝗿𝗲𝘄 𝟯𝟰 𝘁𝗼 𝟱𝟮 𝗽𝗲𝗿𝗰𝗲𝗻𝘁 𝘆𝗲𝗮𝗿 𝗼𝘃𝗲𝗿 𝘆𝗲𝗮𝗿 𝗶𝗻 𝟮𝟬𝟮𝟲 𝗿𝗲𝗳𝗹𝗲𝗰𝘁𝘀 𝗲𝘅𝗮𝗰𝘁𝗹𝘆 𝘁𝗵𝗲 𝗸𝗶𝗻𝗱 𝗼𝗳 𝗶𝗻𝘁𝗲𝗻𝘁-𝗿𝗶𝗰𝗵, 𝗱𝗲𝘀𝗰𝗿𝗶𝗽𝘁𝗶𝘃𝗲 𝘀𝗵𝗼𝗽𝗽𝗲𝗿 𝗹𝗮𝗻𝗴𝘂𝗮𝗴𝗲 𝘁𝗵𝗮𝘁 𝗮 𝗻𝗮𝘁𝘂𝗿𝗮𝗹 𝗹𝗮𝗻𝗴𝘂𝗮𝗴𝗲 𝗱𝗶𝘀𝗰𝗼𝘃𝗲𝗿𝘆 𝗲𝗻𝗴𝗶𝗻𝗲 𝗶𝘀 𝗯𝘂𝗶𝗹𝘁 𝘁𝗼 𝗵𝗮𝗻𝗱𝗹𝗲 – 𝗮𝗻𝗱 𝘁𝗵𝗮𝘁 𝗮 𝗸𝗲𝘆𝘄𝗼𝗿𝗱 𝗶𝗻𝗱𝗲𝘅 𝗶𝘀 𝘀𝘁𝗿𝘂𝗰𝘁𝘂𝗿𝗮𝗹𝗹𝘆 𝘂𝗻𝘀𝘂𝗶𝘁𝗲𝗱 𝗳𝗼𝗿. The same analysis flags that most LLMs cannot render JavaScript or dynamic content, which means dynamically loaded product descriptions are effectively invisible upstream. Catalog legibility is both an on-site and an off-site problem simultaneously.

𝗩𝗶𝘀𝘂𝗮𝗹 𝗰𝗼𝗺𝗺𝗲𝗿𝗰𝗲 𝗶𝘀 𝗻𝗼𝘁 𝗮 𝘀𝗲𝗽𝗮𝗿𝗮𝘁𝗲 𝘁𝗿𝗲𝗻𝗱. 𝗜𝘁 𝗶𝘀 𝘁𝗵𝗲 𝘀𝗮𝗺𝗲 𝘁𝗿𝗲𝗻𝗱

Meta introduced an AI-powered room visualization feature this week that lets shoppers see real products in their own spaces before purchasing, with the transaction completing on the brand’s own website. AI try-on was independently linked to higher ecommerce conversion rates in a separate report. And Marks and Spencer moved to deploy AI specifically for online product discovery on their own digital storefront.

These three signals read as separate stories on the surface. They are not. They are all expressions of the same underlying shift: shoppers increasingly expect to see, not just read, before they buy. The describe-and-show paradigm that is reshaping search queries off-site is the same paradigm driving visual commerce adoption on-site. A shopper who can tell an LLM “I want a midi dress in rose shade with a relaxed fit. Oh yeah, and make it sleeveless please.” and get relevant results off-site will expect to be able to do the equivalent – or upload an inspiration image – on the storefront they land on.

Meta deploying room visualization at social-platform scale raises the ambient expectation. It does not solve the problem for individual DTC merchants on their own storefronts. Marks and Spencer investing in AI discovery for their own site is the more instructive signal: even retailers with significant engineering resources are treating on-storefront AI discovery as a strategic investment worth making explicitly, not something the platform handles for them.

𝗧𝗵𝗲 𝗾𝘂𝗲𝘀𝘁𝗶𝗼𝗻 𝗳𝗼𝗿 𝗺𝗲𝗿𝗰𝗵𝗮𝗻𝘁𝘀 𝗯𝘂𝗶𝗹𝗱𝗶𝗻𝗴 𝗿𝗶𝗴𝗵𝘁 𝗻𝗼𝘄

Wall Street has made a directional call. Payments infrastructure is being rebuilt. Consumer behavior has already shifted. The benchmarking frameworks are live. The visual commerce evidence is accumulating.

The merchants who will capture the agentic traffic being priced into these platform valuations are the ones who close the gap between what the platform promises and what their own storefront actually delivers – catalog structured for LLM legibility, discovery that matches how intent-rich shoppers actually arrive, and visual surfaces that meet the expectation meta-scale tools have already set.

Here is the question I keep coming back to: if your storefront’s discovery experience has not materially changed in the last 18 months, which side of this readiness gap are you on – and what specifically would it take to move