In a Q&A with Shopify’s Deann Evans, InternetRetailing’s managing editor Amanda Vlietstra asked what ‘success’ in agentic commerce looks like for retailers right now, what this means for traditional ecommerce websites – and how retailers should be preparing for what’s still to come.
Q: Where are you seeing the biggest impact of AI in commerce right now?
A: What we’re seeing is that consumer adoption of agentic commerce is really happening. When we look at AI searches, traffic’s up nearly 13 times year over year, and there’s been 8x year-over-year growth for AI-driven traffic to Shopify stores. Orders from AI search are also growing at twice the rate of organic search.
So, from a consumer perspective, agentic commerce is happening now. We’re seeing a lot of growth for retailers. But to really benefit, retailers need to make sure AI agents can surface accurate pricing, accurate inventory, and complete product attributes. Consumer adoption is scaling, and brands need to be ready for it.
Q: How are retailers responding? Are you seeing a wide variety in how they’re using AI, and what does “successful” look like?
A: Let me give you an overview of some of the solutions in the Shopify platform and how they fit together. Shopify started almost 20 years ago helping entrepreneurs sell their first product online. Commerce is complex — we often describe it as an “edge case” — and agentic commerce adds a new layer of complexity.
For agentic commerce, AI needs three things: a way to find the product, a common language to talk to the merchant, and a place for the purchase to happen. We’ve built a suite of products around this. Shopify Catalogue is purpose-built for AI discovery. It standardises product data, infers categories, pulls in variants, and ensures pricing and inventory are current. Every merchant on Shopify is automatically part of the catalogue, so their data and inventory are up to date by default.
Agent Storefronts enables Shopify merchants to have their products appear and sell in places like ChatGPT, Copilot, and Google Search. In Shopify admin, they can effectively tick a box and say, “I want a storefront in this agentic surface.”
We’ve also launched our Agentic Plan that extends these capabilities to merchants not currently on Shopify. They can still work with the Shopify Catalogue to ensure their product data is accessible and accurate wherever AI-driven discovery is happening.
We’re also seeing that shoppers who arrive from AI search convert nearly 50% higher, and their average order values are about 14% higher. So, for retailers, success today is really about: Is my catalogue data ready? Am I discoverable? Is there a seamless way to check out when that AI-driven shopper arrives?
Q: If we move toward fully agentic shopping, where people might not visit traditional retail storefronts at all, how does that affect the role of the storefront and of Shopify?
A: I think about this in two ways. First, the consumer needs enough trust to hand over the autonomy of shopping. They need confidence that the agent will act on their behalf within the guardrails and parameters they set — for example, “You can spend X pounds on this purchase.”
Second, the consumer needs confidence that the transaction will execute reliably. Agents are already very good at discovery. The hard parts are payments, fraud, tax, returns, clearing reliably.
That’s where the Unified Commerce Protocol (UCP) comes in. We co-developed UCP with Google – it’s an open protocol designed to make sure returns are handled consistently, loyalty programmes work, payments work. Basically, all the rails of the transaction are standardised. The aim is to give retailers one open standard rather than forcing them to integrate into 17 different proprietary systems.
On the merchant side, a key principle for us is that the merchant stays the merchant of record, regardless of where the purchase happens. Discovery will shift, checkout methods will evolve, but the brand should still own the customer relationship.
Q: How does this work for first-party data and ownership of the customer journey if discovery and checkout happen inside, say, ChatGPT or Perplexity?
A: First, I don’t think the retail storefront is going anywhere. It’s still incredibly important for both brands and consumers.
Right now, a lot of what we’re seeing is that discovery happens in an AI interface and the agent draws on Shopify Catalogue. Then, when it’s time to purchase, the checkout is effectively a merchant-controlled window or SDK overlay.
So, as a consumer, you might be in your AI of choice, but when you complete the purchase, you’re actually using the merchant’s own checkout. That means it’s still their checkout, they still own the data, and it’s still their customer experience.
We see this as democratising discovery rather than privileging only big brands with huge budgets. Reach is becoming more influenced by relevance than by pure ad spend. If you have the right product, the right data, and you’re using the catalogue effectively, you can show up just as effectively as a large player.
For new and scaling merchants, this is a big opportunity to drive growth. And for Shopify, when our merchants win, we win.
Q: AI adoption is high among retailers, but some retail leaders say they’re not yet seeing measurable commercial benefits. Why is that, and what are you seeing at Shopify?
A: A lot of the value right now is also on the operational side. We have Sidekick, our AI agent that lives inside the Shopify platform. I spend a lot of time with brands using it in very different ways. For example, OHMU, a Danish furniture design studio with a viral product — the Teddy sofa — uses Sidekick for new SKU launches, accessories, seasonal shifts and marketing performance. They’re saving around 10 hours a month just from Sidekick, and reinvesting that time into higher-value work.
We also see merchants using Sidekick for strategic questions like: “If I want to increase sales by X%, which products should I focus on?” or “Which markets should I expand into internationally?”
So we are already seeing real-world, measurable efficiencies and faster execution, which ultimately support growth — even if, for some leaders, the revenue uplift isn’t fully attributed yet.
Q: If you were advising a retail CEO on the single most important capability to invest in over the next 12 months, what would it be?
A: Data. It comes back to the idea of catalogue. How does your product data show up? Is your inventory in real time? Do you have complete, structured product attributes?
Consumers are already using AI for discovery. If your data isn’t ready, you’re invisible in that world. So I’d say: invest in your catalogue, your data quality, and your readiness to plug into these agentic surfaces.
Q: Finally, if we were having this conversation in 2030, what would most surprise the retail industry about the impact of AI and agentic commerce?
A: I wish I had a crystal ball for 2030! What I can say is that where we are today, from an agentic perspective, is built on the work we’ve done over the last 20 years at Shopify. So my view is: do the work now. Make sure your data is right, your infrastructure is in place, and you’re building along with the evolution of consumer behaviour. Whatever 2030 looks like, the brands that invested early in their data, infrastructure, and unified commerce foundations will be the ones in the best place to thrive.
Stay informed
Our editor carefully curates two newsletters a week filled with up-to-date news, analysis and research. Click here to subscribe to the FREE newsletter sent straight to your inbox. Why not follow us on LinkedIn to receive the latest updates on our research and analysis?
