Unfiltered Views on Ecommerce: The Evolution of Product Discovery and Checkout

Get The Print Version

Tired of scrolling? Download a PDF version for easier offline reading and sharing with coworkers.

A link to download the PDF will arrive in your inbox shortly.

Key highlights

• AI shopping is shifting product discovery from keyword searches toward conversations about what customers need.

AI shopping is shifting product discovery from keyword searches toward conversations about what customers need.

• Complete, accurate product data helps shoppers and AI tools understand when a product is the right fit.

Complete, accurate product data helps shoppers and AI tools understand when a product is the right fit.

• Discovery, comparison, and checkout can happen across new surfaces, making a connected path to purchase more important.

Discovery, comparison, and checkout can happen across new surfaces, making a connected path to purchase more important.

• Trust remains essential: shoppers need clear purchase decisions, and merchants need to manage orders and customer relationships.

Trust remains essential: shoppers need clear purchase decisions, and merchants need to manage orders and customer relationships.

• Flexible commerce systems help brands adapt as the places customers shop continue to change.

Flexible commerce systems help brands adapt as the places customers shop continue to change.

What happens when a customer can find, compare, and buy a product without ever opening your website?

That question sits at the center of Kristina Pototska’s session at DNZ’s upcoming Ongefilterd event. As Director of Product Management at Commerce, Kristina has watched ecommerce move through search, social, and marketplaces. Now, she’s looking at another change in how people discover products and complete purchases: AI assistants that can help shoppers move from a question to a recommendation, and increasingly, to checkout.

Her unfiltered view is that the traditional ecommerce journey is changing. Brands still need to deliver a useful shopping experience, but they can no longer assume every customer will begin with a search bar, land on a product page, and buy on their website.

Shoppers are describing problems, not searching for products.

Imagine someone looking for running shoes. Instead of searching for “running shoes” and scrolling through results, they ask an AI assistant, “What are the best running shoes that won’t hurt my knees?”

As Kristina puts it, “Search is moving from keywords to conversation.” The shopper has supplied a need and expects a relevant answer. For a product to be considered, the information available about it needs to help the assistant understand why it might fit that need.

That makes Kristina’s next point hard to ignore: “Your catalog is the new SEO.” A product record with a generic title, a short description, and a few basic attributes may omit the details a shopper needs to make a decision. Accurate specifications still matter. So do use cases, meaningful attributes, and answers to the questions people ask before they buy.

This is not a call to cram every possible question into a description. It is a reason to look at a catalog through a customer’s eyes. Does the product information explain what an item does, who it is for, and why someone might choose it? If those answers exist only in a sales associate’s head or a customer support conversation, they may be missing from the places where discovery now happens.

The path from discovery to purchase is getting shorter.

Ecommerce teams are used to thinking in stages: attract a shopper, bring them to the site, help them evaluate options, and guide them through checkout. Kristina describes what happens when more of that work takes place inside an AI experience. A shopper can explain what they want, compare recommendations, and, in some experiences, make a purchase without visiting the merchant’s website.

This does not mean the storefront disappears. It remains a place to tell a brand’s story and give customers a direct way to shop. But it may no longer be the only place where a buying decision happens. And when a shopper does reach the site after using an AI assistant, they may arrive with a clearer idea of what they want.

As Sharon Gee, Commerce’s SVP of Product for AI, said in a Commerce Live 2026 announcement, “merchants can reap the benefits of agentic commerce while maintaining control of their data, their customers and their operations.” The question for brands is how to participate in new buying experiences while staying connected to the order, the customer, and everything that follows the sale.

Frictionless buying still has to feel trustworthy.

If an AI assistant helps a shopper choose a product, the next step should not require them to start their search all over again. Kristina argues that ease of transaction will become a more important way for retailers to stand out. But a shorter journey only works if the customer understands what they are buying, what they will pay, and when they are approving the purchase.

Trust matters on the merchant side, too. Product and availability information must be accurate. The order needs to reach the right system. The business still has to fulfill it and support the customer afterward. Moving checkout into a new surface does not make those responsibilities go away.

That is why this shift reaches beyond a new interface. Merchants need product data, payments, order management, and customer service to work together even when discovery starts somewhere else.

Build for the channels you cannot predict yet.

Kristina’s points are candid about the pace of change. New shopping experiences are emerging quickly, and no brand can know exactly which ones its customers will adopt. The useful response is to make product information more complete and build a commerce foundation that can support different experiences as they arise.

There are practical examples of that flexibility in EMEA. Working with DNZ and BigCommerce, Dutch fashion brandStudio Anneloes connected its online store to a native app that uses product information and checkout from BigCommerce. Since replatforming, the brand reported a 24% increase in conversion rate and a 25% increase in items sold. Those are results from broader improvements to its shopping experience, but the underlying approach shows how one commerce foundation can support more than one way to buy.

Royal Swinkels needed to manage distinct brands across multiple websites. DNZ helped bring them into a BigCommerce Multi-Storefront setup where products can be managed centrally and made available on the right storefronts. Each brand can maintain its own customer experience without duplicating the work behind every site.

Neither story predicts what AI shopping will look like in five years. Both show the value of keeping product information and commerce operations connected as the places customers shop continue to multiply.

The final word

The unfiltered truth is that brands cannot control every new shopping surface. They can improve what those surfaces know about their products, make the path to purchase easier, and keep ownership of the experience that follows.

So what happens when a customer finds and buys from you without opening your website? As Kristina puts it:

“Your customer might never see your homepage, but they’ll still experience your brand. They’ll see how well you answered their question, how easy you made the purchase, and what happened after they placed the order.”

“Your customer might never see your homepage, but they’ll still experience your brand. They’ll see how well you answered their question, how easy you made the purchase, and what happened after they placed the order.”

Kristina will explore these ideas at DNZ’s Ongefilterd event in Amsterdam on November 19. Join the conversation about what comes next for product discovery and checkout.

Commerce built for momentum.

Get up to speed with new features that drive your business forward — from core capabilities to agentic commerce (and beyond).