Your Store’s Future Increasingly Will Depend on How Well You Talk to Robots

By Craig Webb, President, Webb Analytics

Scot Wingo spends hours every day nudging AI-powered search engines until they notice his clients. He’ll point out when a retailer offers a special discount good across its website. He pays attention to whether a product’s online description reads “blue/white” or “white/blue.” He finds information about the client that could make it more attractive to ChatBot or Google Gemini or other services, then helps arrange for that stuff to go online before he nudges again.

Correcting all these “hundreds of little pieces of product data not aligning and confusing AI” go into what Wingo calls “agentic commerce optimization.” For building product manufacturers and dealers, this work could become crucial to their success in 2026. That’s because the explosive growth in web searches powered by artificial intelligence mandates that construction supply companies satisfy AI robots’ formatting expectations or risk getting shut out of customer searches. IBM noted that one company saw a 50% decline in Google search results when the search engine launched AI-curated responses.

Wingo’s company, ReFiBuy, was among the roughly 1,000 tech operations from around the world that focused on AI during the National Retail Federation’s Big Show this month in New York. AI starred at last year’s trade show, too, but in 2025 the highlights were all about the back-of-house operational improvements that artificial intelligence made possible. That trend deepened this year, with the rise of AI systems that act as agents on your behalf—identifying trends that could lead to new products, for instance, and then creating a “synthetic focus group” to critique the proposed widget.

For manufacturers, that may be a help. But for lumberyards and other dealers, the overriding message from the 2026 Big Show was that AI search engines are transforming how consumers find, learn about, and decide on products and companies. Dealers that haven’t told the internet much about themselves will lose the race to businesses that do.

Agentic Commerce Optimization builds on Search Engine Optimization basics—such as making sure your website includes an address, phone number, and operating hours—to dig far deeper into your operations and product offerings. The goal is to load the website with every possible tidbit of information about your company and your SKUs: pictures, descriptions, history, prices, services, product numbers, special knowledge, material safety data sheets … the works. 

That might seem like overkill, but not to an AI search engine. It wants to Hoover up all that information .so that it can produce deeper, more nuanced answers to consumer questions. And those types of questions are coming. Research by IBM and Adobe both found dramatic increases in consumers’ use of ChatBot and Gemini and the like. People increasingly are describing their wants in complete paragraphs and then asking AI agents to recommend solutions. 

One way to provide these extra insights is to write answers to frequently asked questions, experts said. Connections to videos is another must-have, as is access to as many reviews of a product as possible, even if those reviews weren’t originally posted on the dealer’s website. Some AI searches are pulling up comments posted on Reddit, said Vivek Pandya, Adobe’s Director of Digital Insights.

While volume is important, so is formatting. Wingo says the various AI search services tend to read the same information differently, down to how to interpret various terms and descriptions. For example, Shantha Farris of IBM noted that people have different definitions around the word “availability.” Does that mean the product is in stock, or does it have to be ordered? Is it on a shelf? If it’s not on the shelf, can it be delivered? If so, long can it take to deliver the product and still have it qualify as “available.” 

Some systems also are sticklers for knowing product numbers, and want those numbers placed directly next to the product name, Wingo says. ReFiBuy studies these differences, then adjusts its communications “nudges” with each AI search engine to get them to recognize information it might have overlooked or ignored before.