Lindsay Owens on Price Experiments, Consumer Data and the Fight for Fair Prices

Theo Von introduces Lindsay Owens, author of Gouged and head of the Groundwork Collaborative. Owens says the Washington think tank researches corporate practices to inform policies that protect consumers.

Owens began studying inflation in 2021. She says that rising costs were real, but investor calls suggested some businesses also saw an opportunity to raise prices beyond those costs when customers expected increases.

The research began with a question about what inflation would mean for household budgets and expectations. Owens listened to companies explain their pricing to investors and came away concerned that public explanations did not tell the whole story.

Asked what price gouging means, Owens describes its legal use around excessive prices in an emergency, such as expensive water after a hurricane. She also uses the phrase more broadly for prices people perceive as unfair; the broader sense is her own framing.

Von wonders how large organizations with interests across different businesses can influence what consumers see and pay. Owens then separates that concern from the narrower question of how a dynamic price works.

Dynamic pricing, Owens explains, changes with conditions such as demand. Concert tickets offer a clear example: the last seats can go to people willing to pay most instead of the people who reached the front of a line.

Von asks whether that is necessarily unfair. Owens acknowledges the economic case for selling scarce tickets to the highest bidder, but argues that queues and lotteries give people another kind of access. Which method feels fair depends on the goal.

Owens says there are several ways to allocate a limited good. A lottery preserves a chance for people who cannot bid the most, while an auction prioritizes the buyers willing and able to spend more.

Owens worries about extending frequent price changes to groceries, where she says ordinary items are not scarce in the same way. She discusses digital shelf labels as infrastructure that could make future changes easier.

The hosts read Walmart’s explanation that the labels replace paper tags and that its existing pricing process remains in place. Owens says Walmart denies using the labels for dynamic pricing; her weather-based examples describe a possible future use, not an established practice.

When Von asks whether there is proof of weather-based pricing at Walmart, Owens says it is difficult to know exactly what the company does and repeats that it has denied using the labels for that purpose.

Owens points to a Walmart patent describing scan-and-go shopping and a possible change in the price of a complementary item after a customer selects another. A tuna and mayonnaise example appears in the discussion. A patent is evidence of a proposed technology, not proof that it was deployed.

In the scan-and-go example, a shopper adds an item on a phone rather than waiting until checkout. Owens says the patent describes changing a related item’s price, but corrects Von when he asks whether Walmart has admitted doing so.

The conversation moves to Live Nation and Ticketmaster. Owens argues that when one business controls venues and ticketing, consumers and artists have fewer alternatives, weakening the competitive pressure that could constrain fees and prices.

The competition argument is straightforward in Owens’s telling: when a customer can take the same business elsewhere, a seller has a reason to keep its price competitive. She argues that a lack of alternatives weakens that pressure.

The show describes a jury finding against the companies and a proposed breakup. The speakers note that the states had asked for a divestiture, but that the requested remedy was not an order already issued at the time of the discussion.

Owens draws on her experience working on Capitol Hill to describe lobbying. In her account, companies can seek delays, suggest changes or press for exceptions that weaken consumer-protection legislation before it becomes law.

Owens says lobbying can also involve apparently minor wording changes and loopholes. She recounts hearing from lobbyists after announcing her book, illustrating what she sees as sustained corporate attention to policy.

She wants Congress to use hearings and subpoena powers to ask companies how their pricing systems work. Public scrutiny, she argues, could also help lawmakers decide which rules are needed.

Von asks whether lawmakers lack technical understanding, face pressure from lobbyists, or both. Owens answers that both contribute in her view, making oversight of new AI and pricing systems harder.

Owens also favors restrictions on members of Congress trading stocks. She sees potential conflicts when lawmakers regulate companies while having personal financial interests, and argues that elected officials should be accountable to constituents.

Returning from sponsor messages, Von asks about experiments. Owens describes a project her organization conducted with Consumer Reports and More Perfect Union, recruiting 400 volunteer shoppers to compare a set of 23 grocery items.

The volunteers selected the same store and pickup location and photographed their carts. Owens says prices differed across shoppers for 75 percent of the items, with individual differences as high as 23 percent.

The exercise used ordinary groceries such as eggs, cereal, deli meat and snacks. By holding the store and pickup location constant and collecting screenshots, the researchers tried to compare what shoppers actually saw.

Owens says Instacart described the differences as A/B tests and later wound down its Eversight pricing tool. The observation of differing prices does not by itself establish that customers were targeted according to individual characteristics.

Owens says Instacart had acquired Eversight, a company working on pricing and promotions. According to her account, after publication and public criticism Instacart said shoppers at the same store and time would see the same prices.

She describes another comparison involving the same StubHub ticket listing: the listed ticket price was the same, but fees differed by as much as $60 among shoppers. She says her team also observed differences in delivery and service fees in food-ordering apps.

For food delivery, Owens says colleagues compared the same item delivered to the same location. She regards differing service and delivery fees as another reason consumers need better visibility into the final price.

Owens discusses a DoorDash patent describing an agitation score using app interaction signals. She says the application was withdrawn after she raised it in a Senate hearing. A patent application does not show that the proposed scoring was actually used.

The proposed DoorDash method included signals such as scrolling and device information, according to Owens’s description. Her interpretation is that such signals could affect promotions; she does not present evidence of the patented method being deployed.

For Owens, public reaction can matter. She recalls a proposed temperature-sensitive Coca-Cola vending machine as an example of an idea that met resistance when consumers saw how prices might change.

Owens says companies sometimes retreat after customers object publicly. Her broader point is that a practice can be technically possible while still being rejected as unacceptable by customers.

The discussion turns to the history of fixed prices. Owens describes Quaker merchants and early department stores as examples of a move away from bargaining toward posted prices that customers could see and compare.

The discussion also touches on department-store merchant John Wanamaker and the combination of posted prices with returnable goods. Owens presents that history as an earlier effort to make commerce more predictable.

Von describes liking a local produce stand where the price stays the same. Owens argues that a business might win trust and customers by making predictable pricing part of its offer.

Owens offers the Savannah Bananas as a different example: she says the team uses a fixed posted ticket price and a lottery when demand exceeds supply. Not everyone gets a seat, but the tickets do not all go to the highest bidders.

They contrast the lottery example with the 2026 World Cup. Owens criticizes FIFA’s use of dynamic pricing as a choice that favored revenue over access for less wealthy fans.

Owens describes patents associated with Walmart that mention data or technologies such as biometric signals and vehicle information. She treats these as reasons to ask questions about possible future uses; the patents alone do not demonstrate that those uses are operating.

One application Owens describes as an “appeasement engine” concerns how a business might respond to a dissatisfied customer. She acknowledges uncertainty about the data included and how any such idea would be implemented.

The guest argues that loyalty accounts and shopping records can give retailers a detailed picture of what a customer buys and how often. Von asks how much can be inferred from a car or shopping pattern; Owens describes hypothetical possibilities discussed in patent material.

Another patent example concerns images of a vehicle before and after shopping, with a possible inference about weight from its suspension. Owens and Von question why such an inference would be useful; the discussion does not establish its real-world deployment.

Owens also describes a hypothetical use of heart-rate or other wearable data to infer interest in a product. She treats the possibility as a privacy concern, while the company’s denial later in the conversation remains relevant to what is actually happening.

Owens compares digital tools with physical retail design, describing store layouts intended to keep shoppers browsing. She notes that products aimed at children may be placed at a child’s eye level, where a parent will be asked to buy them.

Owens recounts corresponding with Walmart after her organization examined its patents. She says the company denied individualized pricing and said it had no plans to use the patents that way; she nevertheless wants clearer commitments and scrutiny.

In the letter read on the show, Walmart says it does not and will not use an individual’s personal information, shopping history or willingness to pay to set an individualized price. Owens disputes whether its broader communications are sufficiently clear.

Owens suggests that Walmart withdraw pending applications or relinquish patents it says it does not intend to use. She also calls for independent review of pricing systems, presenting these as her proposals to reduce uncertainty.

The discussion turns to Walmart’s Sparky shopping assistant. Owens asks whether an assistant could steer shoppers toward more expensive products and stresses the difficulty of seeing how recommendations are generated.

Von and Owens discuss an airline customer-service post suggesting that a traveler clear cookies or use an incognito window. Owens says the airline later called the post an error and denied the implication; the exchange does not establish how the airline sets fares.

The airline example involves a traveler seeking a funeral fare after the price rose. The customer-service suggestion prompted speculation, but Owens notes that the company subsequently said the post was mistaken.

Owens argues that loyalty programs offer discounts while collecting extensive shopping data. She sees customer data as an additional business asset and worries that consumers do not always understand the trade they make for a coupon.

The speakers discuss data businesses connected to grocery retailers. Owens argues that repeat purchases can reveal a great deal over time; the figures mentioned on the show should be understood in the context of the companies’ broader business categories.

Asked what shoppers can do, Owens says friends can compare rideshare quotes or airline fares across devices or accounts. She presents this as a practical experiment, not a guarantee that any difference has a particular cause.

She recommends comparing the same retailer’s app, website and physical store, as well as nearby locations. The advice comes with a cost: the time required to check every channel is itself a burden on shoppers.

Owens says that in her group’s StubHub comparison the observed fee gap became much smaller in incognito mode. She cautions that outsiders cannot see exactly how the algorithms work, so this result cannot explain every difference.

Owens suggests sharing experiences publicly, pointing to consumer pressure as one possible check on practices shoppers consider unfair. She still sees policy, rather than individual vigilance alone, as the durable answer.

The pair discuss digital menu boards, customer data and ways to raise concerns publicly. Owens says publicity has sometimes prompted companies to reconsider proposals, though a possible use of technology should be distinguished from a demonstrated use.

They cite a WIRED reporter’s request for his own McDonald’s data, which they say produced hundreds of pages of purchase and loyalty information. For Owens, the example illustrates the scale of information a routine account can accumulate.

The reported data-access example includes purchase history and estimates of future visits. Von and Owens react to the asymmetry: customers may know little about the models built from their own routine activity.

Von wonders whether technology can serve people more directly. Owens argues that the effort put into modeling customer behavior could instead go toward better products and more transparent service.

They consider consumer organizing through communities, churches or unions. Owens argues that people can use collective purchasing power and solidarity to seek better terms rather than relying entirely on individual workarounds.

Owens mentions a shopping tool that its maker says can compare a basket across retailers. She expressly says she has spoken with its creator but has not used the product herself, so her description is not an endorsement based on personal testing.

AI assistants could help customers with time-consuming jobs such as canceling subscriptions, Owens says. But an assistant may also gain access to sensitive information and operate under incentives that are difficult for a shopper to inspect.

The hosts describe an early study using fictional profiles to test shopping recommendations across AI models. They say the study found that recommendations sometimes varied with apparent wealth even when the prompt asked for a cheaper option; they discuss it as research, not proof of a particular commercial deployment.

The show reports that the study tested many fictional profiles and several model families. Owens interprets the result as a warning that an assistant’s recommendation can diverge from what its user explicitly requested.

Owens considers transaction fees and advertising as possible business models for assistants. Von asks whether merchant fees might be reflected in prices; the two frame this as a risk and an open question rather than a settled outcome.

The speakers consider whether a transaction-based assistant might save time while also introducing a new intermediary. They do not settle how such services will be priced or whether merchants will pass costs on.

An assistant called an agent should act in the customer’s interest, Owens argues. Her concern is that opaque recommendations, advertising and commercial relationships may pull it in a different direction.

The show recalls a 2015 ProPublica investigation of geographic differences in SAT preparation prices. The presenters say higher prices were more common in some ZIP codes with larger Asian populations; the reported correlation alone does not establish the company’s intent.

Owens argues that income is not the same as willingness or urgency to pay. Her concern is that a pricing system could exploit a person’s immediate need or lack of alternatives, even when that person is not wealthy.

Owens outlines a proposed shoppers’ bill of rights. Her priorities include clear all-in prices, a straightforward refund when a product fails, and limits on using personal information to set individualized prices.

The pair discuss a New York proposal that Owens says had passed the legislature and was awaiting the governor’s action at the time of recording. She advocates signing it and describes industry lobbying against the measure; its later status is not established in the conversation.

Owens says the proposed New York law would restrict individualized pricing based on personal information. She describes similar, narrower efforts elsewhere and argues that the state could set an example for others.

To hold officials accountable, Owens urges voters to ask for investigations into pricing practices and for stronger consumer protections. She says Congress can demand information that an outside research group cannot obtain on its own.

Owens wants hearings to ask company executives how pricing and shopping assistants work. In her account, an independent organization can examine public materials but cannot compel the internal records that lawmakers can seek.

Von thanks Owens for explaining the issue and congratulates her on the book. She closes by pointing listeners to her organization’s research and encouraging continued public scrutiny of corporate pricing.