The quarter itself
Shopify reported second-quarter 2026 results on August 5. Revenue reached $3.58 billion, up 34 percent year over year and ahead of the average analyst estimate of $3.45 billion (Yahoo Finance, 2026). Gross merchandise volume climbed to $115.57 billion, up 32 percent, the fifth consecutive quarter of GMV growth above 30 percent (Yahoo Finance, 2026). International GMV grew 37 percent, faster than North America's 28 percent and Europe's 34 percent (Digital Commerce 360, 2026). Offline commerce grew at the same pace as the platform overall: point-of-sale GMV was up 32 percent, with Shopify crediting momentum among large, multi-location retailers such as Holt Renfrew (Digital Commerce 360, 2026).
Those are strong numbers, but strong quarters from Shopify are not new by now. What is new sits one layer down, in how the company is framing its AI products, and it points to a different fight than the one most retail coverage has been having this year.
The numbers inside the numbers
Two products anchored the AI portion of the call: Sidekick, an assistant merchants use to run their stores, and Catalog, the structured product-data layer that feeds Shopify's own AI search and increasingly feeds outside AI shopping tools too.
Daily active use of Sidekick was up 3.6 times year over year, handling close to 34 million conversations in the quarter (Yahoo Finance, 2026). Merchants used it to build roughly 36,000 custom apps in the quarter, up from about 12,000 the quarter before (Yahoo Finance, 2026). Catalog's numbers are the more interesting ones: AI-powered searches drawing on Catalog's structured data converted at roughly twice the rate of AI searches drawing on scraped page content, and 75 percent of AI-attributed orders came from outside the top 100 product categories, meaning small and mid-sized merchants gained the most (Investing.com, 2026). Overall AI-driven traffic and orders to Shopify stores roughly tripled year over year (Investing.com, 2026).
Read those together and a pattern appears. Shopify's AI investment is not aimed primarily at shoppers. It is aimed at merchants, and at making merchant catalogs legible to AI systems that are not Shopify's own.
A protocol most retail coverage skipped
That framing lines up with something that happened earlier this year and got less attention than it deserved. In January, at the National Retail Federation's annual conference, Google introduced the Universal Commerce Protocol, an open specification for letting AI agents browse, compare and buy across retailers without each retailer building custom integrations for every AI system (PYMNTS, 2026). Shopify, Etsy, Wayfair, Target and Walmart were named as founding merchant partners, alongside a coalition of more than 20 payment networks and processors including Mastercard, Visa, Stripe and American Express (PYMNTS, 2026).
The protocol is plumbing, not a product a shopper ever sees. But plumbing is exactly what determines who gets paid when an AI agent, whichever one wins that separate contest, decides to buy something.
Why the data layer outlasts the chat window
Most of the public conversation about AI and retail this year has centered on the shopper-facing question: will people trust a chatbot to complete a purchase. That is a real question, and it has already produced some uneven results across the industry. It is also the wrong layer to watch if the goal is figuring out which companies come out ahead.
Whether a given AI shopping assistant survives, gets rebuilt, or gets replaced by a competitor's version, it still needs product data to work from: price, availability, images, variants, accurate descriptions. A platform that owns that data cleanly, machine-readably and current wins by supplying whichever agent ends up on top, not by betting on one. That is the argument Shopify's Catalog numbers support: a search built on structured data converts at twice the rate of one built on scraped pages, and the 75 percent figure suggests the advantage compounds for exactly the merchants least equipped to fix their own data on their own.
Whether Adobe Commerce, Salesforce Commerce Cloud, BigCommerce and the enterprise point-of-sale vendors respond with catalog products of similar depth is the real competitive question for the rest of the platform market, not whether any one of them ships a customer-facing chat window this year.
It is also, not coincidentally, the same data infrastructure retailers need for reasons that have nothing to do with chatbots. A merchant that can tell an AI agent exactly what is in stock at which location is a merchant whose supply chain and store network are visible to itself in the first place. The AI use case and the operational use case draw on the same investment.
What to watch
Shopify's third-quarter guidance calls for revenue growth in the low thirties percent, ahead of the 26.3 percent analysts had modeled (Yahoo Finance, 2026). If that guidance holds, expect more retailers to publish their own structured commerce feeds before the holiday quarter, not because a chatbot demanded it, but because the merchants who already invested in clean data are the ones showing up first when an AI agent goes looking for something to buy.