The median American household spent $0 on food delivery in 2025, according to new federal data — yet DoorDash and Uber Eats keep posting growth. The explanation, economists say, is a narrow band of superusers spending thousands of dollars a year.

Consumers in their mid-40s or younger who order hundreds or thousands of dollars of food annually appear to form the core customer base for delivery apps, according to data from the Federal Reserve's Survey of Consumer Finances, collected in 2025 and released in October.

"Those are where probably most of their revenue is coming from," said Jeremy Horpedahl, an associate professor of economics at the University of Central Arkansas. "It's super important for them to retain those customers."

The same data suggest many people do not use delivery services at all, Horpedahl said.

What does the survey actually show?

The Survey of Consumer Finances runs every three years and asks 5,000 households how they earn, spend, save, and invest their money. Horpedahl analyzed the food delivery responses by dividing spending into percentiles — ten groups of 10% each, from the lowest spenders to the highest.

The results were stark:

• For the eight bottom percentiles, the median consumer spent $0 on food delivery — suggesting many respondents have never used delivery apps at all.
• The median consumer under 35 in the top 10% of spenders spent $1,800 a year on food delivery.
• Among top-10% spenders aged 35 to 44, the median figure rose to $2,400 a year — roughly $200 per month on delivered food.

Put differently: for a large majority of households, delivery is not a spending category at all, while a small top tier spends at rates comparable to a recurring utility bill.

Why is delivery still growing while retailers struggle?

The numbers frame a divergence in the retail economy. Walmart and Dollar General have both warned in recent quarters that shoppers are pulling back or hunting for better value — a trend the retailers say now extends to households with six-figure incomes.

DoorDash and Uber, by contrast, have reported steady growth in orders and spending, with no comparable pullback.

A small group of users willing to spend hundreds or thousands of dollars a year could explain why delivery services keep posting strong earnings while much of the retail sector warns about cautious consumers, Horpedahl said.

"It could be true that 95% of people are pulling back," he said. "But if this group is still spending, that's why Uber Eats could be saying 'Yes, we're still seeing strong revenue.'"

How are the companies responding?

A DoorDash spokesperson acknowledged the limits of the federal snapshot. "Surveys like this are a helpful snapshot, but they capture only one slice of how people use DoorDash today," the spokesperson said.

The company has recently added more grocers and other retailers to its platform, the spokesperson added. It is also growing its paid DashPass membership program — a mechanism aimed squarely at retaining its most frequent users, the very cohort the Fed data identifies as the revenue engine.

What are the caveats and the strategic stakes?

The survey's food delivery question does not name DoorDash or Uber specifically; it covers other services as well, including pizza delivery. Horpedahl also noted it likely misses respondents who ordered delivery once or twice but forgot those purchases while answering.

For the industry, the data point to two distinct opportunities, he said. Companies can still lure first-time users, widening a base that remains largely untapped. But poaching superusers from rivals could prove particularly lucrative.

"If you get just one percent of those going to the other company, that would be huge," Horpedahl said.

The arithmetic is simple: in a market where most households spend nothing, each heavy spender a platform wins — or keeps — carries outsized weight.