Picture yourself pushing a grocery cart down the aisle. Suddenly, a small screen on the handlebar lights up with an ad for discounted strawberries, right as you enter the produce section. Do you grab a carton? And if you do, does that little nudge change anything else in your basket?
These questions sit at the heart of a new field study examining “smart” shopping carts equipped with digital ad displays. Published in the Journal of Retailing, the research shows that location-aware ads on cart screens increase sales not only of the products being promoted, but also of other items in the same category that weren’t advertised at all.
A gap in what we know about in-store ads
Retailers have long used in-store advertising to shape buying behavior, and shoppers themselves tend to view these ads favorably. A 2024 Kantar survey even found that consumers ranked in-store ads above cinema ads and event sponsorships for trustworthiness and usefulness. Most research on in-store advertising, though, has focused on stationary displays like shelf signage and above-aisle screens, or on push notifications delivered through smartphone apps.
That leaves a blind spot: what happens when ads travel with the shopper, displayed on a screen mounted to the cart itself? Mathias C. Streicher of the University of Innsbruck, working with Oliver B. Büttner of the University of Duisburg-Essen and Zachary Estes of Bayes Business School in London, set out to investigate.
Smart carts differ from smartphones in several practical ways. A phone ad needs a downloaded app, active Bluetooth, and a shopper who notices the alert. A cart screen sits in plain view, is hands-free, and can be timed to specific store zones using small Bluetooth beacons placed around the shop. That combination allows retailers to show an ad for wine precisely when a shopper rolls into the wine aisle.
How the field study worked
The researchers ran their study over eight days in a mid-sized European grocery store, covering roughly 1,200 square meters. They installed Bluetooth beacons at the ceiling to divide the store into seven shopping zones, following the typical path from entrance to checkout.
Three types of carts were rotated among 1,236 shoppers as they entered. Some got an ordinary cart with no screen. Others got a cart with a screen showing ads in the center of the display. A third group got a cart whose ads appeared on the left or right side of the screen, along with an arrow pointing toward where the advertised product sat on the shelf.
As shoppers with digital carts passed each beacon, the screen triggered two ads for products located in that zone, each shown twice in short bursts. Across the seven zones, that came to a maximum of 14 ads per trip, promoting items like strawberries, sparkling wine, mozzarella, pesto, beer, frozen fish, and chocolate. Every ad featured a price discount.
Shoppers were not told they were part of a study. After checkout, a research assistant quietly collected a duplicate receipt from the cashier and marked it to indicate which type of cart had been used. The team then coded each receipt for three measures: purchase quantity (total items), purchase variety (unique items), and total spending.
What the receipts revealed
Shoppers using carts with digital ads bought more of the advertised products. On average, spending on promoted items jumped from about 92 cents in the no-display group to roughly 1.78 euros with a central display and 2.08 euros with a peripheral display. In percentage terms, that’s an increase of more than 100 percent on advertised items.
The bigger surprise showed up in non-advertised products. Shoppers with digital displays also bought more items that were never promoted on the screen. Spending on non-advertised items rose from about 45 euros in the control group to around 48 to 53 euros with the display carts. In absolute dollars, this spillover was far larger than the bump in advertised-product spending, because non-advertised items made up most of the basket to begin with.
Overall, shoppers without a cart screen spent an average of 46.15 euros. Those with a digital display spent an average of 52.95 euros, a 14.7 percent increase in total spending.
The researchers also looked at whether the spillover stayed within the advertised category or spread elsewhere. An ad for Lindt chocolate, for example, could in theory lift sales of other sweets (within-category) or of entirely different items like chips or soda (cross-category). The data showed spillover only within the advertised category. The Lindt ad lifted sales of other sweets, but not of unrelated products. Ads for food items did not boost spending on non-food categories like apparel or garden supplies.
One hypothesis that didn’t pan out: the peripheral ads with directional arrows did not outperform the central ads. Pointing shoppers to the left or right side of the aisle didn’t give an extra lift beyond simply showing the ad.
What it might mean for retailers and brands
For retailers, the appeal is straightforward. Shoppers with digital cart ads spent about 6.80 euros more per trip on average. That gross increase compares favorably with the roughly 8 percent sales lift that a recent large study found for stationary digital signage. Even accounting for the cost of installing screens and beacons, the payoff could come relatively quickly.
For brand managers, the picture is less clear-cut. Yes, an in-cart ad lifted sales of the promoted brand. But it also lifted sales of competing brands in the same category. For a dominant supplier acting as a “category captain,” that may still be worthwhile if it grows the category overall. For a smaller brand trying to win share from rivals, paying for an in-cart ad that also helps competitors may be harder to justify.
There’s also a consumer-welfare angle worth noting. Every ad in the study featured a price discount, yet shoppers ended up spending more overall, largely on non-discounted items. The authors suggest that consumer policy should make shoppers aware of these broader effects, and that retailers should consider letting customers turn displays off.
Some caveats apply. The study was a quasi-experiment, meaning carts were handed out pseudo-randomly rather than under strict experimental controls, so self-selection can’t be entirely ruled out. Data came from a single mid-sized grocery store during daytime hours, and the researchers had no information about individual shoppers beyond their receipts. Whether these patterns hold in discount chains, hypermarkets, or convenience stores remains an open question.
