The smart cart has become one of the most visible investments in physical retail. Market forecasts put the category on a steep growth curve. Analysts project it to multiply several times over before the end of the decade. The announcements follow the same script almost every time. A retailer commits to a large fleet. The press release leads with a faster, frictionless checkout. The story gets filed under the familiar heading of eliminating the dreaded line at the register. That framing is where most retailers go wrong before they deploy a single cart.
Read the market analysis past the headline and a different picture emerges. The reports that size this category do not model the return on checkout speed. They model recurring revenue from hardware, software subscriptions, and, above all, retail media. The screen on the cart, the one that shows the running total, is also an advertising surface. It sits in front of the shopper at the exact moment of decision. That surface, not the scanner, is what the financial models actually count.
The smart cart is a retail media platform that happens to do checkout, not a checkout device that happens to show ads. A retailer who evaluates the investment against register labor is measuring the wrong metric, and will almost always conclude the numbers do not work. A retailer who evaluates it against recurring, high-margin media revenue is looking at the model the market actually runs on. Getting that distinction right before the purchase order separates a deployment that pays off from one that disappoints.
What the Market Forecasts Actually Measure
When research firms project the smart cart category growing at the rates they do, the growth is not coming from checkout speed. It comes from a shift in the revenue model. The analyses segment the market by revenue type, and they consistently name three streams. The hardware itself, whether bought or leased. The software and service fees. And retail media. The first two are familiar. The third is reshaping the economics, because it is recurring, high-margin, and does not depend on selling more carts to grow.
The analysts are explicit about this. One question they flag for retailers to watch is whether retail media revenue can shorten the payback period on the hardware. That framing is telling. It means the hardware, on its own, has a payback period long enough to worry about. The thing that rescues the investment is the media revenue layered on top. The cart does not pay for itself on checkout efficiency. It pays for itself when brands pay to reach the shopper on its screen.
Why the Checkout Framing Keeps Failing Retailers
The Checkout ROI Has Never Fully Materialized
The promise that smart carts would pay for themselves by reducing checkout labor has run for years. The clean proof of it at scale has been slow to arrive. The reason is structural. A smart cart does not remove the staff a store needs. It shifts what they do. Shoppers still need help. Exceptions still happen. Age-restricted and loss-prone items still require intervention. The labor does not disappear the way the original pitch implied. A retailer who justified the fleet on register savings alone finds, a year in, that the savings are softer than the business case promised.
As I described in the smart cart and scan and go analysis, the checkout function is real and valuable, but it is not where the differentiated return lives. Treating it as the headline sets up the investment to be judged against a number it was never going to hit cleanly. The checkout is the feature that gets the shopper to use the cart. It is not the feature that pays for it.
The Media Model Changes Who Should Approve the Investment
When the return comes from retail media rather than labor savings, the evaluation moves to a different part of the organization. A checkout investment gets judged by store operations against a labor budget. A retail media investment gets judged by the commercial side against an advertising revenue opportunity. Those are different budgets, different decision-makers, and different success metrics. The retailers who struggle often ran the decision entirely through operations. There, the media upside never even entered the business case, because operations does not get credit for ad revenue.
What a Retailer Should Actually Evaluate Before Buying
The Media Inventory, Not the Scanner Speed
If the cart screen is the revenue engine, the first question is not how fast the cart scans. It is who controls and monetizes the screen. In many deployment models, the cart provider keeps the retail media rights and the data. That means the recurring revenue that justifies the whole category flows to the vendor, not the retailer. A retailer who signs that deal has bought the hardware cost and handed away the business model. Before the purchase order, the retailer needs three answers. Who owns the advertising inventory? How is the revenue split? Can the retailer run its own promotional content without going through the vendor’s platform?
The Data the Cart Generates and Where It Goes
Retail media is only as valuable as the shopper data behind it. The smart cart generates a uniquely rich stream. What the shopper scanned, in what order, how long they paused, what they put back. As I described in the agentic commerce inventory accuracy analysis, that behavioral data is a strategic asset, and whoever owns it captures the long-term value. A deployment that sends the behavioral data to the vendor’s systems builds the vendor’s media business, not the retailer’s. The data ownership terms matter as much as the hardware specifications.
Adoption, Because an Unused Cart Shows No Ads
The media model has a dependency the checkout model does not. An advertising surface is worth nothing if shoppers do not use the cart. That makes customer adoption the single most important operational variable, and most deployments underinvest in it. A retailer counting on media revenue has to treat adoption as a revenue driver, not a convenience metric. If only a fraction of shoppers use the carts, the media inventory the business case depended on never reaches the scale that makes it valuable. The adoption plan is not a soft launch detail. It is the foundation the entire return rests on.
What This Means for LatAm Retailers
For LatAm retailers evaluating smart carts, the retail media reframing matters even more, because the region’s retail media networks are earlier in their development. That cuts two ways. On one hand, the media monetization that justifies the hardware is harder to achieve where the advertising ecosystem and the brand demand for in-store media are still forming. A LatAm retailer who assumes the media revenue will appear automatically may find the demand side is not there yet to fill the inventory.
On the other hand, the retailer who understands the model early can build the retail media capability alongside the hardware. The smart cart only makes sense in a market where the retailer can actually sell the media inventory it creates. For most LatAm retailers, the right sequence is to develop the retail media relationships and the data infrastructure first. Then bring in the hardware when there is demand ready to monetize the screens. The cart without the media business behind it is an expensive checkout device, measured against labor savings that will not carry it.
The Question That Reframes the Whole Decision
Are You Buying a Checkout or a Media Network?
Every retailer evaluating smart carts should answer one question first. Are you buying a faster checkout, or an in-store media network that also handles checkout? The honest answer determines which budget funds it, which team owns it, which metrics judge it, and whether it returns the investment. The retailers who answer checkout will measure against labor savings and conclude the economics are marginal. The retailers who answer media network will structure the deal around the revenue stream the market is actually built on.
The market is deploying smart carts at record pace, and the coverage keeps leading with the checkout line. But the checkout was never the business model. The forecasts that justify the category are built on retail media: the recurring, high-margin revenue that comes from the screen, not the scanner. The retailers who understand that structure the investment correctly. They secure the media rights, they keep the shopper data, they treat adoption as the revenue driver it is, and they judge the deal against the model it actually runs on. The retailers who keep believing the pitch about eliminating the line will buy the hardware, miss the business model, and wonder why the numbers never worked. The cart was always a media platform. The checkout was just how it gets shoppers to push it.
Adriana Rivas is a retail technology executive and AI strategist. She is a recipient of the Gold Stevie® Award, named to the RTIH Top 100 Retail Technology Influencers 2026. She is the author of How to Implement Self-Service Without Failing: From Store Automation to Store Intelligence, now available.