Make better decisions on the AI in retail market

AI is spreading fast behind the scenes in retail. But when it touches the price tag, shoppers and lawmakers push back.

We've found a lot of interesting signals. For example, in December 2025, Instacart ended its AI-driven price tests after Consumer Reports and two other groups found that some shoppers paid up to 23% more than others for the same items, from the same store, at the same time. And in July 2026, New Jersey signed into law a ban on using shoppers' personal data to set grocery prices, along with a one-year pause on new electronic shelf labels. Retailers can still use AI to price smarter, but they now have to show it's fair.

In this section, we look at where retailers are putting AI first, why pricing draws so much attention, and what that means for the vendors selling to them.

Retail runs on thin margins. A few points of waste, empty shelves or bad markdowns can wipe out a year's profit.

We compare six kinds of retail AI companies, from merchandising and pricing tools to demand forecasting, store cameras and shopping assistants. Their buyers include grocers, fashion chains, big-box stores and convenience retailers. For each, we rank the pains that matter most, like fresh food thrown away, stock sitting in the wrong store, and price changes that take weeks to reach the shelf.

This shows which problems retailers will pay AI to fix first, and which ones they still leave to their own planners.

Yearly software licenses, fees per store or per category, a share of the savings, and hardware plus a subscription for cameras and robots: retail AI vendors charge in many ways.

We compare who pays, how much, and what each kind of tool costs to run. A forecasting tool sold per store needs months of data cleanup before it pays off. Pricing vendors paid on extra margin can earn more, but they must prove the gain. Shelf cameras and robots carry hardware and installation costs in every store. And big chains often try to build the tool themselves once it works.

You'll see which models keep their margins as retailers squeeze vendors, and which ones depend on pilots that never reach every store.

Better models alone won't grow this market. Here, we show what actually gets a retailer to sign, and then to roll a tool out to every store.

Some drivers are plain to see: rising labor costs, food waste targets and shoppers who expect items to be in stock. Others feed each other. Better forecasts cut waste, which frees money for more tools. Shelf data makes the forecasts better. And once a chain trusts AI with one category, it tends to hand over the next one.

Before building or investing, you should know what makes a retailer trust an AI tool with its prices and orders, and take it beyond the pilot stores.

Blue Yonder and RELEX come up first when retailers talk about AI planning, but they cover only part of the market.

Many of the most interesting players focus on a single problem. Some only forecast fresh food for grocers. Others track shelves with cameras, set markdowns for fashion chains, clean up product data so site search works, or give store staff answers on their phones.

They're hard to compare because each one works on a different decision, and results depend on the retailer's own data. Most retailers also don't name their vendors, so a tool can run in thousands of stores without anyone outside noticing.

Below, you'll find the market grids and ecosystem maps we've built to show who does pricing, planning and store AI, and where new players are getting in.

Afresh, Simbe, Lily AI… a few retail AI startups have found work that retailers keep paying for.

Afresh helps grocers order fresh food so less of it ends up in the trash. Simbe scans shelves with robots and cameras to find gaps and wrong prices. Lily AI rewrites product data in the words shoppers actually search for. Each found a clear job, though none has yet shown it can grow into a full retail platform.

We'll look at how these teams picked one costly problem, proved savings with real store data, and grew from a few stores to whole chains.