Adobe says AI-referred shoppers spend 53% more, how does a store get recommended?

It measures revenue per visit, not total sales. One visit from an AI assistant leaves more revenue, on average, than one visit from other sources. The figure says nothing about AI traffic’s share of total sales.

According to Adobe, in July 2025 a non-AI visit was worth 128% more per visit. By July 2026 the picture has flipped. Our short reading: assistants filter shopping intent better and send visitors who are closer to a decision.

The Perakende.org analysis gives 53% for May 2026, citing Reuters. Adobe’s August report gives the same rate for July 2026. So whenever you name a month, name the source next to it. The two periods should not be read as one data point.

Open the referral sources report in your analytics tool. Visits from chatbots show up as separate sources. Grouping them into one segment is the first step to seeing your own rate.

Murat Eren Taşçı, board member at MISIRLI Underwear & Socks, says AI is felt on the visual side of textiles and underwear and helps on costs. On its wider effect, he says “we don’t yet fully know which sectors it will affect, and how” (our translation). He argues for using stores, online and digital channels together.

The 62% annual growth. Even though it is a US figure, it shows a channel growing fast. For a Turkish retailer, the question is whether its products will be on the assistant’s list once this channel grows.

Brands that spell out size, material, use and returns on the product page gain. An assistant can turn that information straight into an answer. Boutique brands with tidy product data on their own sites can also earn a place next to bigger players.

A store with one-line descriptions, weak images and outdated prices struggles. The product sits on the shelf, but the assistant does not know it exists. A seller that competes only on marketplaces and only on price also narrows its chance to stand out.

Physical stores feel it indirectly. A shopper who decides with an assistant may still want to see the item in person. If your stock data is wrong and the item is not there, that visit can be the last; our piece on what a small store should digitize first looks at this side.

Give every product fields for brand, material, size, color, price and stock. Also mark these fields up on the page with schema markup. A missing field means an empty cell in the assistant’s comparison.

The ones shoppers ask an assistant. Questions like “who is it for, how do I wash it, which size should I take” belong inside the description. Short guide texts on category pages do the same job.

If the product an assistant recommends is out of stock, the shopper loses trust in your store. Feeding price and stock from a single source is what turns a recommendation into a sale.

In the same field experiments, the gains come mainly from higher conversion rates. Tools that cut friction in search, information and personalization stand out. Our reading: the value of AI lies in helping shoppers find what they want by a shorter route.

In a 2026 Vogue Business survey cited by Perakende.org, 98% of fashion and beauty shoppers do not regularly use AI chat tools. Only 24% trust AI-driven campaigns. The sample and country are not given. Still, it reads as a reason to build hybrid channels rather than hand all communication to AI.

The analysis cites an academic study reporting gains of more than 13% in clicks and conversion for AI-designed products. The scope of that study has not been verified. Treat the figure as a signal, not a planning number.

Start with the products at the top of your revenue ranking. Gather their attribute fields, images and descriptions in a single table. Fill the empty cells the same week.

In your analytics tool, put chatbot-referred visits into their own segment. Compare that segment’s conversion and average basket with your other sources. Your own number is a better guide than US data.

Open your seasonal product pages before the season starts and add question-and-answer sections. Our piece on back-to-school shopping shows which products seasonal demand spreads across.

Not directly. The data comes from US retail sites. For Türkiye, you need to measure AI-referred traffic separately in your own analytics tool.

If your product information is clear, complete and current, your chance of being recommended goes up. Filling in product attribute fields and adding question-and-answer content is the first step.

Not definitely. In field experiments the effect is zero in some workflows and 16.3% at best; the result depends on the tool you use.