A new investigation has revealed that Instacart, one of the largest grocery delivery platforms in the United States, is quietly running algorithm driven pricing experiments that can dramatically change what customers pay for identical items. The findings come during what consumer advocates describe as the worst grocery affordability crisis in a generation and show that families using Instacart may be paying hundreds of dollars more each year without ever knowing why.
Researchers with Groundwork Collaborative, Consumer Reports, and More Perfect Union analyzed real time shopping data from more than 400 Instacart users across four cities. Their report, titled “Same Cart, Different Price,” documents a system of hidden artificial intelligence enabled price testing that can raise the cost of a single item by up to 23 percent depending on the shopper. The average pricing variation seen across baskets of identical groceries could cost a typical household of four about 1,200 dollars per year.
The groups found that Instacart’s price experiments were occurring across major grocery chains including Albertsons, Costco, Kroger, Safeway, Sprouts Farmers Market, and Target. At a time when food inflation is outpacing overall inflation and grocery prices have become one of the largest financial burdens for American households, the report concludes that algorithmic pricing is amplifying hardship rather than offering convenience.
The investigation relied on controlled shopping tests conducted at Safeway and Target locations in North Canton, Ohio, Saint Paul, Minnesota, Seattle, Washington, and Washington, D.C. Researchers examined 193 high quality participant submissions and later confirmed similar pricing patterns at several additional retailers. Participants logged into Instacart at the same time to buy identical items from the same stores, allowing investigators to capture price fluctuations that were independent of timing or store inventory.
Across all tests, nearly three quarters of the grocery items studied showed different prices to different shoppers. Some single items carried as many as five simultaneous price points. In one example from a Safeway in Washington, D.C., a dozen Lucerne eggs appeared at 3.99 dollars, 4.28 dollars, 4.59 dollars, 4.69 dollars, and 4.79 dollars for different customers shopping at the exact same moment. The price swings were not limited to specialty items. Staples such as cereal, pasta, and peanut butter showed an average difference of 13 percent between the highest and lowest prices seen, with some items reaching the full 23 percent spread documented in the report.
The variance extended to full baskets of groceries. At a Safeway in Seattle, shoppers purchasing the same set of items were quoted totals ranging from 114.34 dollars to 123.93 dollars. The study also found a 23 percent difference in prices for Skippy peanut butter, Oscar Mayer turkey, and Wheat Thins crackers at another Seattle location. At two Safeway stores in Washington, D.C., Signature Select Corn Flakes showed a 23 percent price difference depending on which Instacart shopper viewed them.
The report did not find evidence that Instacart was setting prices based directly on users’ zip codes or income levels. However, investigators emphasized that Instacart’s acquisition of Eversight, an AI pricing software company, gives the platform the ability to incorporate personal data in future pricing decisions. Instacart’s chief executive previously said the technology helps the company “to really figure out which categories of products our customers [are] more price sensitive on.”
Instacart acknowledges on its own website that it runs price tests. According to the company, “shoppers are not aware that they’re in an experiment,” and the pricing system can “create an orthogonal array of data points that changes the game.” Researchers also found cases in which Instacart kept the sale price of an item the same but raised the original price listed, artificially inflating the size of the advertised discount.
Instacart has argued that its pricing experiments are randomized and “negligible,” but the documented basket variations and estimated 1,200 dollars in annual household impact undermine that claim. Groundwork Collaborative executive director Lindsay Owens stated that “Instacart is quietly running pricing experiments on millions of shoppers during the worst grocery affordability crisis in a generation, and it’s costing households as much as 1200 dollars a year.” She added that “they have turned the simple act of buying groceries into a high-tech game of pricing roulette. When the same box of Wheat Thins can jump 23% in price because of an algorithm, that’s not innovation or convenience, it’s unfair. It’s time for Instacart to close the lab. Americans shopping for groceries aren’t guinea pigs and shouldn’t have to pay an Instacart tax.”
Consumer Reports echoed these concerns. Justin Brookman, director of tech policy at the organization, said the tactics “hurt families who are simply trying to purchase essential groceries.” He described the price testing as “particularly egregious” at a time when households are already struggling with high costs. “Companies must be transparent and upfront with people about pricing, so that they can make informed choices and keep more of their hard-earned money. We encourage the Federal Trade Commission and state attorneys general to investigate Instacart’s pricing tactics.”
The report appears as lawmakers at the state and federal level turn their attention to algorithmic pricing and the growing use of personal data in setting costs for consumers. States such as New York, Colorado, California, Georgia, Illinois, and Pennsylvania have introduced or advanced legislation targeting individualized or AI driven price manipulation. New York passed the first Algorithmic Pricing Disclosure Act, requiring companies using such tools to inform consumers that “This price was set by an algorithm using your personal data.”
At the federal level, Representative Greg Casar of Texas has introduced the Stop AI Price Gouging and Wage Fixing Act to prohibit companies from using personal data to set individualized prices.
The report argues that the Federal Trade Commission already has tools available under Section 5 of the FTC Act, which bans “unfair methods of competition,” including “price discrimination not justified by differences in cost or distribution.” The authors say Instacart’s pricing experiments appear to fit this definition. They also urge the FTC to consider rulemaking that would define AI enabled pricing strategies as an “unfair or deceptive practice,” establishing a clear consumer protection standard for digital marketplaces.
The authors write that “fair and honest markets are the bedrock of a healthy economy” and warn that companies like Instacart “are increasingly pursuing corporate pricing practices that unfairly decouple the price of a product from its true cost.” While public pressure may encourage some companies to change voluntarily, they conclude that “if they do not, policymakers should intervene and require them to change their practices.”
The findings show how algorithm driven pricing, once confined to airline tickets and ride hailing apps, has quietly entered the realm of essential goods. For families relying on Instacart amid rising grocery costs, the result may be a system that charges different people different prices without notice, transparency, or fairness.
COMMENTS