Free tools Windows power users keep installed
One-click scans. No signup required.
iTechGuides is reader-supported. When you buy through links on our site, we may earn an affiliate commission. As an Amazon Associate I earn from qualifying purchases. Learn more
U.S. retail-sales data is a monthly estimate of sales by retail and food-service businesses to final consumers, produced by the Census Bureau’s Monthly Retail Trade and Food Services Survey (MRTS). It is an important gauge of demand for goods, but it is not a complete measure of household spending: services outside retail and food services are largely missing, and the series records where goods are sold rather than where they were produced.
To read the report correctly, identify whether the figure is an advance or regular estimate, seasonally adjusted or unadjusted, note the industry mix, and check later revisions. The distinction matters because an early percentage change can differ from the revised estimate, and a higher dollar total can reflect prices rather than more merchandise being purchased.
| # | Preview | Product | Price | |
|---|---|---|---|---|
| 1 |
|
Consumer Economics: Issues and Behaviors | $78.80 | Buy on Amazon |
| 2 |
|
Consumer Economics | $63.97 | Buy on Amazon |
| 3 |
|
Consumer Education And Economics, Student Edition | $8.99 | Buy on Amazon |
| 4 |
|
Economic Education for Consumers | $33.00 | Buy on Amazon |
| 5 |
|
Consumer Education And Economics, Student Activity Manual | $14.04 | Buy on Amazon |
What the monthly retail-sales report measures
The MRTS measures dollar sales and end-of-month inventories reported by firms classified in U.S. retail trade and food services under the North American Industry Classification System (NAICS). Census publishes estimates for individual industries and aggregates, in both seasonally adjusted and unadjusted forms.
The Tool Desk
Outbyte Driver Updater FREEFix the driver behind crashes, sound loss and screen glitchesFind Drivers →Outbyte PC Repair FREEClear out junk files and repair common Windows errorsFree Scan →Retail sales are one of Census’s primary measures of consumer demand for durable and nondurable goods. The headline total is therefore a measure of transactions through retail and food-service businesses, not a direct tally of every purchase made by households. A purchase of a service from a doctor, lawyer, software consultant, airline, or many other providers is generally outside this retail series.
#1 Best Overall
The survey measures sales by the place of sale. A television imported into the United States and sold by a domestic retailer appears in retail sales just as a domestically manufactured television does. The series does not identify the production origin of the goods.
How Census collects the data
A stratified business sample
MRTS uses a stratified, one-stage sample rather than attempting to contact every store or company. Census divides the population into 85 primary industry strata and then into several annual-sales-size strata. The largest firms in each industry are selected with certainty; other businesses are selected randomly.
The current survey description covers approximately 13,000 paid-employee retail businesses, including approximately 2,500 certainty-selected units. Those figures describe the survey design, not the number of stores in the country and not a complete census of every retailer.
Quick wins for a faster PC:
Clear out junk files and repair common Windows errorsFree Scan →Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →Repair Windows errors before they cause bigger problemsFix Now →What businesses provide
Sampled firms report sales for the reference month and inventories held at the end of that month. Census combines the reported observations with imputed values for eligible businesses that did not provide usable data, then applies sampling weights to estimate totals for each industry.
Weighting, benchmarking and sample updates
Monthly weighted estimates are benchmarked to the latest available annual survey totals. This keeps the monthly series aligned with a more comprehensive annual measurement as new annual information becomes available.
Rank #2
Census refreshes the sample for business births and deaths each quarter. New weights are phased into the published estimates over three months, reducing the chance that an abrupt sample change will create a false jump in the time series. In the April 2025 annual revision, nonemployer businesses were removed from the MRTS time series to align it with the Annual Integrated Economic Survey. Comparisons that cross that revision should use a consistent, current vintage.
How a monthly estimate is produced and released
Advance estimate: MARTS
The Advance Monthly Retail Trade Survey (MARTS) provides an early national estimate of total retail and food-service sales and the month-to-month change. Census normally releases it about 10 business days after the end of the reference month.
Recommended Free Tools
Because many responses are not yet available, MARTS uses a link-relative estimator. It calculates the weighted ratio of current-month sales to the previous month for the businesses providing information, then applies that ratio to the prior month’s benchmarked level. This design makes the advance report timely but leaves more room for later revision.
Regular MRTS estimate
The regular MRTS release arrives approximately six weeks after the reference month. It incorporates the more complete MRTS responses and replaces the corresponding advance estimate. Additional revisions can occur when Census incorporates annual benchmarks, updated classifications, corrected reports or other revisions to the underlying survey.
| Feature | Advance MARTS | Regular MRTS |
|---|---|---|
| Typical release timing | About 10 business days after the reference month | About six weeks after the reference month |
| Data available | Fewer completed responses; link-relative estimator based on current-to-previous-month weighted sales ratios | More complete MRTS responses and updated estimates |
| Revision risk | Higher, because the first estimate is based on less information | Lower than the advance estimate, but later revisions remain possible |
| Best use | Early signal of national sales and month-to-month direction | Analysis requiring a more complete monthly estimate |
Always label a number by its vintage. “September sales rose 0.4%” is incomplete unless it also says whether the figure is an advance or revised estimate and whether it is seasonally adjusted.
Rank #3
Seasonally adjusted versus unadjusted data
Retail activity follows strong recurring patterns: holidays, back-to-school shopping, weather, the number of trading days, and other calendar effects can move sales sharply even when underlying demand has not changed. Census seasonally adjusts the series with X-13ARIMA-SEATS and an X-11 filter-based procedure. The adjustment accounts for recurring seasonal, holiday and trading-day patterns.
| Series | What it does | When it is most useful |
|---|---|---|
| Seasonally adjusted | Removes estimated recurring seasonal and calendar effects | Comparing one month with the previous month to assess underlying momentum |
| Unadjusted | Retains the actual seasonal pattern and dollar level | Examining the dollars recorded in a particular month, holiday-period activity, or comparisons where seasonality is part of the question |
Seasonal adjustment is an estimate, not a permanent transformation. As additional observations arrive, the adjustment factors and historical values can be revised. For a month-to-month signal, use the seasonally adjusted series; for the actual seasonal size of sales, use unadjusted data and explain the calendar context.
What retail sales can—and cannot—say about consumer spending
What it captures well
- Spending on durable goods sold through retail channels, such as many vehicles, appliances and electronics.
- Spending on nondurable goods sold through retailers, including many household and personal items.
- Food-service purchases made through the businesses included in the survey.
- Changes in sales by NAICS industry, allowing analysts to see whether a total is broad-based or concentrated in a few industries.
What it leaves out or measures differently
- Many consumer services are outside retail trade and food services, so retail sales are narrower than total household consumption.
- The series is reported in dollars. A rise can result from higher prices, greater quantities, or both; the headline does not by itself provide a volume measure.
- Sales are assigned to the location and industry of the seller, not to the factory where a product was made.
- Online and other remote sales are included when they are made by businesses in the covered retail industries, but the report is not a separate measure of “e-commerce spending” across the whole economy.
A strong retail-sales month therefore supports a conclusion about demand for covered goods and food services, not a claim that all consumer spending or real household purchasing power increased by the same percentage.
Retail sales versus BEA personal consumption expenditures
The Bureau of Economic Analysis (BEA) uses Census retail data in estimating personal consumption expenditures (PCE), especially the goods components. Its retail-control method uses sales by kind of business from Census monthly and annual retail surveys to extrapolate many goods PCE categories.
| Question | Retail sales (Census MRTS) | Personal consumption expenditures (BEA) |
|---|---|---|
| Coverage | Retail trade and food-service businesses selling to final consumers | Broader household consumption, including extensive service categories as well as goods |
| Primary source | Monthly and annual Census retail surveys of businesses | Multiple source series, including Census retail data, administrative records and other economic statistics |
| Price/volume interpretation | Reported sales dollars; a separate price adjustment is needed to assess real volume | BEA publishes nominal and inflation-adjusted PCE measures using national-accounts methods |
| Imports | Includes imported goods when they are sold by covered U.S. businesses | National-accounts treatment separates domestic production from imports and uses them in the GDP framework |
| Revision pattern | Advance, regular and later Census revisions | BEA’s own monthly, quarterly and annual revision schedule, incorporating updated source data |
Retail sales and PCE can move differently without either series being wrong. PCE includes services that retail sales omit, while retail sales can react immediately to a change in goods purchases that is a smaller part of total consumption.
Do these 3 things before closing this tab:
1Repair Windows errors before they cause bigger problems2Scan for outdated or missing drivers - takes under a minute3Clear out junk files and repair common Windows errorsRank #4
How retail sales feed into GDP
Retail sales help BEA estimate consumer spending, which is the “C” in the expenditure identity C + I + G + X − M: consumption plus investment plus government spending plus exports minus imports.
The import issue is essential. Retail sales count an imported product when a U.S. retailer sells it. GDP measures domestic production, so the value of imports is subtracted in the “− M” term. A jump in sales of imported phones can raise retail sales and the consumption component while producing little or no increase in U.S. domestic output; the import adjustment prevents that foreign production from being counted as American GDP.
Census describes its data as an input to BEA’s GDP estimates and national-accounts input-output tables. Retail sales are therefore an important building block, not a one-for-one substitute for GDP or for the full PCE series.
How to read a retail-sales release step by step
- Identify the vintage. Check the reference month, release date, and whether the number is advance MARTS or regular MRTS.
- Choose the adjustment status. Use seasonally adjusted data for month-to-month momentum. Use unadjusted data when the actual seasonal dollar pattern is the subject.
- Check the comparison. Month-to-month changes are sensitive to seasonal adjustment and calendar effects; year-over-year changes avoid some of those effects but can be distorted by the unusually strong or weak base month.
- Inspect industry contributions. Determine whether the total reflects broad-based movement or a small number of industries. A headline increase concentrated in one category carries a different signal from a similar increase spread across many categories.
- Separate dollars from volume. Ask whether prices may explain part of the change. Retail sales alone do not tell you how many units were purchased.
- Compare with broader spending measures. Use BEA PCE when the question includes services or total household consumption, and account for the different release and revision schedules.
- Read the uncertainty and revision notes. Treat a small movement cautiously when it is within the survey’s stated sampling uncertainty or when the estimate is an early vintage.
Why retail-sales numbers are revised
More complete responses
The advance estimate is produced before all businesses have reported. As late responses arrive and imputations are replaced, the regular estimate can change. Corrected business submissions can also alter prior values.
Annual benchmarking
Census benchmarks monthly totals to the latest annual survey information. Annual data may show that the level or industry distribution of sales differed from the assumptions used in earlier monthly estimates, requiring a revised history.
Business births, deaths and changing weights
Retail businesses open, close, merge and change classification. Census updates the sample quarterly for births and deaths and phases revised weights in over three months. This process improves representativeness but can change recent observations.
Seasonal-factor updates
Seasonal adjustment uses patterns estimated from the historical series. New observations can change those patterns, leading to revisions even when the underlying business reports have not changed.
Uncertainty and common interpretation mistakes
Census reports both sampling and nonsampling error. Sampling error exists because a sample, rather than every eligible business, is measured. Census describes the margin of sampling error as a 90 percent confidence interval. A small reported change should not be treated as a precise reading of the entire retail economy when the interval is wide.
Nonsampling risks include nonresponse, overcoverage, undercoverage, misreporting, recording and coding mistakes, misunderstanding of questions, and processing errors. These risks are not eliminated by a large sample.
- Mistake: treating an advance number as final. Solution: record the vintage and revisit the regular estimate.
- Mistake: treating nominal sales as real consumption growth. Solution: examine prices and use an appropriate inflation-adjusted measure for volume.
- Mistake: equating retail sales with all consumer spending. Solution: include PCE and service spending when answering a household-consumption question.
- Mistake: assuming a strong total means every retailer is strong. Solution: inspect industry detail and concentration.
- Mistake: ignoring imports. Solution: remember that sales of imported goods enter retail sales but are removed from GDP through imports.
- Mistake: mixing vintages. Solution: compare a consistent revised history, especially across the April 2025 change that removed nonemployers from the time series.
A practical interpretation example
Imagine a hypothetical release showing a 0.8% month-to-month increase in seasonally adjusted retail sales. The correct conclusion is limited: covered retail and food-service sales were estimated to be 0.8% higher than the previous month in that vintage. Before calling it stronger consumer spending, check whether the increase is concentrated in one industry, whether prices rose, whether the figure is advance or regular, and whether the change is large relative to the 90% confidence interval.
If the same month’s unadjusted sales are lower than the prior month, that is not necessarily contradictory. The unadjusted series may reflect the normal seasonal pattern, while the adjusted series removes the expected seasonal decline. Neither series is “the real number”; each answers a different question.
Checklist for citing or using the data
- Name the U.S. Census Bureau MRTS or MARTS series.
- State the reference month and release vintage.
- Specify seasonally adjusted or unadjusted.
- Give the comparison basis: month-to-month or year-over-year.
- Identify whether the figure is nominal sales dollars, an estimated percentage change, or an inflation-adjusted measure from another source.
- Note important industry concentration and any relevant calendar or holiday effect.
- Use PCE for a broader household-spending answer and GDP data for domestic production.
- Recheck the number after the regular release and after major annual revisions.
Bottom line
Retail sales data is a carefully weighted sample estimate of sales at U.S. retail and food-service businesses. It offers a fast, valuable read on demand for goods, but it is narrower than consumer spending, vulnerable to price and seasonal effects, and subject to revisions. The most defensible interpretation combines the release vintage, adjustment status, industry detail, uncertainty, and the broader PCE and GDP concepts that turn store sales into a measure of national economic activity.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
