The October retail report contains a modest headline and an important qualification. The Census Bureau estimated that retail and food services sales reached $748.2 billion in October, up 0.2% from September after seasonal and calendar adjustments. But the 90% confidence interval around that change includes zero, so the agency says the data do not provide sufficient statistical evidence that the underlying month-to-month change differs from zero. The result is best read as an early estimate with an uncertain direction, not proof that shoppers spent more. U.S. Census Bureau

Keep the month measured separate from the release date

Census published the estimate on November 17, but it describes activity in October. That distinction matters when comparing stories: the publication date tells when the estimate became available, while the reference month tells what period it covers. In the same release, September’s seasonally adjusted sales were revised to $746.6 billion and its monthly increase remained 0.8%. October’s $748.2 billion is therefore compared with the revised September estimate, not an unrevised number from an older headline. U.S. Census Bureau

The report also gives a 6.8% increase from October 2025, with a 90% confidence interval of plus or minus 0.5 percentage point. That longer comparison and the 0.2% monthly estimate answer different questions. Neither should be mistaken for a measure of how many more items households bought.

A confidence interval changes what “up” can mean

The release reports the October monthly change as 0.2%, plus or minus 0.4 percentage point. In practical terms, the interval extends from a small decline to a larger increase and includes zero. Census explicitly says there is insufficient statistical evidence to conclude that the actual change differs from zero. That does not prove sales were exactly flat. It means the estimate is not precise enough to separate a small increase from no change using the stated 90% interval. U.S. Census Bureau

A useful analogy is a household survey estimate that points slightly upward but has a margin of uncertainty larger than the measured move. The direction of the point estimate is still information, yet it should not carry the same weight as a result clearly outside its uncertainty range. October’s estimate is a weak standalone signal about momentum, especially when one month can include ordinary volatility and later revision.

The total is in dollars, not inflation-adjusted volume

Census says the sales values are adjusted for seasonal variation and holiday and trading-day differences, but not for price changes. They are dollar receipts. If an item’s price rises while the number sold stays constant, the dollars recorded can increase even though unit sales do not. If prices fall, dollars can understate a change in units. That simple example explains why a nominal sales gain is not automatically a gain in real consumption. It does not tell us how much of October’s result came from prices versus quantities; this release does not make that decomposition. U.S. Census Bureau

The year-over-year 6.8% figure has the same limitation. It compares reported dollars with the prior October and does not subtract inflation. Readers looking for a price-adjusted measure need a separate source and method, and should avoid simply treating this sales percentage as a household purchasing-power change.

Retail sales cover a defined slice of household activity

The Advance Monthly Retail Trade Survey, or MARTS, tracks retail and food service companies. Census describes its scope as businesses selling merchandise and related services to final consumers, along with food services and drinking places. The measure is broader than shop merchandise alone, but it is not a complete account of consumer spending. Many household expenses—such as rent, medical care, education, and financial services—are outside this retail-sales total. U.S. Census Bureau

The unit also matters. Sales by type of retail business are organized around the establishment’s primary business, so a store category is not a tally of every product sold there. The figures can help describe broad demand at retailers and restaurants, but they do not identify what a particular family bought or how a particular business performed. U.S. Census Bureau

The advance estimate is designed to arrive early

MARTS is an early indicator built from a subsample of the larger Monthly Retail Trade Survey. Census describes the advance sample as about 4,800 firms. It derives each industry’s advance change from the subsample and applies that ratio to the previous month’s preliminary estimate, then combines the industry results. This design delivers an early reading; it also means the number can change when later, broader survey data are incorporated. U.S. Census Bureau U.S. Census Bureau

Census reports that the average absolute revision to the advance estimate is about two-tenths of one percent. That historical average is context, not a prediction that this particular estimate will be revised by exactly 0.2 points or in either direction. The best practice is to label the October number as advance, preserve its uncertainty, and check the later release before describing a lasting turn in consumer demand. U.S. Census Bureau

A careful reading in one sentence

A restrained summary would be: Census’ advance estimate put October retail and food services sales at $748.2 billion, 0.2% above revised September in dollars, but the 90% confidence interval includes zero and the series is not adjusted for prices. That sentence names the period, comparison, uncertainty, and measurement basis without turning one early estimate into a verdict about households or the wider economy. U.S. Census Bureau

For the next comparison, look for the revised estimate, check whether the reference period changed, and keep nominal retail receipts distinct from inflation-adjusted spending. Those checks make the monthly release useful without asking it to answer a question it was not designed to measure.

THE EVIDENCE

Sources & further reading

  1. Advance Monthly Sales for Retail and Food Services, October 2026 ↗U.S. Census Bureau
  2. About the Advance Monthly Retail Trade Survey ↗U.S. Census Bureau
  3. Monthly Retail Trade Survey Methodology ↗U.S. Census Bureau