The Secret Algorithms Driving Up Your Grocery Bill
Grocery delivery giant Instacart has been exposed for secretly charging different prices to different customers for the same items at the same stores, according to a new investigation released by Consumer Reports and the Groundwork Collaborative. The study found price differences of as much as 23 percent for identical products and basket totals. Costs varied by about seven percent among shoppers who ordered the exact same groceries at the exact same time.
With food prices up more than 25 percent since the pandemic and a majority of Americans saying the cost of groceries is a source of “major” stress, the findings raise fresh questions about what’s really driving affordability concerns. Is it actual structural problems with the economy, or just corporate greed using new technology like AI to fleece everyday consumers?
The Consumer Reports investigation was published just days after President Donald Trump signed an executive order directing the Department of Justice and Federal Trade Commission to form a new task force to investigate price-fixing and anti-competitive behavior across the food supply chain.
Researchers recruited 437 volunteers nationwide and divided them into synchronized shopping groups. After reviewing screenshots of their online grocery bills, they concluded that every volunteer was a participant in Instacart price experiments without knowing it.
Nearly three-quarters of products showed different prices across shoppers. In North Canton, Ohio, some Instacart customers were charged $2.99 for a jar of Skippy peanut butter, while others paid $3.59. At a Safeway in Seattle, prices for Oscar Mayer deli turkey ranged from $3.99 to $4.89.
The total cost of identical baskets varied between $114.34 and $123.93 in one Seattle test, a spread of roughly $10. Based on typical monthly spending for a family of four, price variations of that size could translate into roughly $1,200 a year in added grocery costs.
Instacart admitted to conducting pricing “tests” at a subset of partnering retailers, saying the experiments are randomized, short-term, and intended to identify which items matter most to consumers. The company denied using personal or behavioral data and said prices do not change in real time, although the study found wide price discrepancies depending on the shopper.
Consumer Reports acknowledged that price discrimination is not unusual in competitive markets. “Charging different amounts to different customers for the same products is not illegal or new,” they said. “U.S. consumers have grown accustomed to paying different prices for the same airline seats, event tickets, hotel rooms, rideshares, and certain other goods and services that are subject to rapidly changing shifts in supply and demand.”
But instead of offering visible discounts for identifiable groups, Instacart appeared to sort shoppers into hidden price buckets, charging different customers different amounts for the same items. The pricing tests probed willingness to pay person by person. With enough individual data, algorithms can approximate the maximum price each shopper will tolerate and charge it.
Unlike posted discounts or loyalty offers, however, shoppers had no way of knowing different prices were being charged. The lack of disclosure, not just the variation itself, is what troubles many consumers.
Naturally, people feel uneasy when these pricing tools are used on basic household staples. As the report noted, “consumers express deep misgivings about algorithmically driven changes in pricing when it comes to more essential goods like food.”
A representative survey of Instacart users found that “72 percent… did not want the company to charge different users different prices for any reason.”
While the Consumer Reports study focused on Instacart, the advent of artificial intelligence has created an alarming possibility for real-time price changes at stores themselves that could put a beating on shoppers’ wallets. Electronic shelf labels now rolling out at Walmart, Kroger, and Whole Foods allow retailers to update prices across entire stores in seconds from a back office.
Many grocers already use apps and payment data to tailor promotions, raising concerns that the convergence of these tools could enable daily in-store price changes.
Here’s the terrifying scenario that could soon become a reality: Let’s say you normally shop at a supermarket and buy canned peaches once a week. When you scan your rewards card, that supermarket knows what you bought and when you bought it. Using a combination of facial recognition technology and GPS tracking of your phone, they could know exactly when you’re going down the canned peaches aisle. An AI tool will then automatically raise the price of canned peaches, knowing you’re likely to pay the higher amount since it’s a staple in your weekly shopping.
Some states are already acting to prevent this possibility from coming to fruition. New York now requires disclosures when prices are set algorithmically using personal data, and legislators in California and Pennsylvania have proposed banning surveillance-style pricing for essential goods altogether.
The Federal Trade Commission said it was “disturbed” by reports that shoppers may be unknowingly placed into pricing tests and opened a probe into Instacart’s Eversight pricing tool earlier this week.
Price discrimination itself is not new. Sales, coupons, and off-peak discounts help businesses manage inventory and compete for value-conscious shoppers. But those tools are visible, and consumers can decide whether the price is acceptable.
The concern now being raised is over a shift toward pricing that consumers cannot observe or compare, weakening the fair game of comparison shopping.
Responding to this concern does not mean price controls or expansive new bureaucracies. It simply means ensuring that shoppers have all the information they are entitled to when making decisions about what to buy, and that stores are not unfairly weaponizing technology – and sensitive personal data – against their own customers.
Sarah Katherine Sisk is a proud Hillsdale College alumna and a master’s student in economics at George Mason University. You can follow her on X @SKSisk76.