Restaurant sales benchmark: what September 2026 data says
What the new data says
The U.S. Census Bureau's Replace with the permanent September 2026 advance report link listed on https://www.census.gov/retail/marts/historic_releases.html (also update the sources list and primary_source_url) (release CB26-165, Oct 15, 2026) estimates that food services and drinking places sold $92.9 billion in September, seasonally adjusted. That is up 1.1% from August and up 7.0% from September 2025. Over July through September, the category was up 6.0% from the same quarter a year earlier. Before seasonal adjustment, September sales were $91.98 billion, compared with $85.19 billion in September 2025.
The day before, the Bureau of Labor Statistics published the September 2026 Consumer Price Index (also update the sources list URL). Its food away from home index, which tracks menu prices, rose 0.2% in September and 3.5% over 12 months. Full service meals rose 3.8% and limited service meals rose 3.2%.
Taken separately, either number can mislead you. Census measures dollars and is not adjusted for price changes, as the release itself says. CPI measures prices only. Combining them shows how much of the industry's growth came from more business and how much came from higher prices.
The industry's real growth, in one line of arithmetic
Strip price out of a nominal growth rate by dividing, not subtracting:
Real growth = (1 + nominal growth) ÷ (1 + price growth) − 1
- All restaurants: 1.070 ÷ 1.035 − 1 = about 3.4% real growth
- Using the full service price index: 1.070 ÷ 1.038 − 1 = about 3.1%
- Using the limited service price index: 1.070 ÷ 1.032 − 1 = about 3.7%
This is a rough estimate. The Census figure covers every kind of eating and drinking place, and the CPI covers a different basket, so you shouldn't treat the gap as an exact count of extra customers. As a benchmark it holds up: nationally, restaurants sold more food in September 2026 than a year earlier, beyond what menu prices explain. For your restaurant to keep pace, your growth after price needs to be at least in that range.
Worked example: splitting your own comp sales
This example uses made-up numbers for a single full service restaurant. Substitute your own POS figures.
| Metric | Sept 2025 | Sept 2026 | Change |
|---|---|---|---|
| Net sales | $118,400 | $124,950 | +5.5% |
| Checks (transactions) | 3,700 | 3,640 | −1.6% |
| Average check | $32.00 | $34.33 | +7.3% |
| Menu price increase taken in the past 12 months | 4.0% across the menu | ||
Step by step:
- Traffic. Checks fell 1.6%. Fewer visits is the first warning sign.
- Price. Of the 7.3% rise in average check, 4.0% is your own price increase.
- Mix. What remains is 1.0727 ÷ 1.040 − 1 = about 3.1%. That comes from guests ordering more or pricier items: add-ons, drinks, larger parties, or a shift toward higher-priced dishes.
- Real growth against your own prices. 1.055 ÷ 1.040 − 1 = about 1.5%.
- Compare to the benchmark. The industry managed about 3.1% real growth when deflated with the full service price index. This restaurant is behind by roughly 1.6 points. In nominal terms, matching the industry's 7.0% would have meant about $126,690 in September sales, or $1,740 more than it took in.
A headline of "+5.5%" looks healthy. The breakdown shows something else: every bit of the growth came from price and mix, and traffic fell. That is a common pattern when price increases outpace what guests feel they are getting, and it calls for a different response than "sales are up, keep going."
Decision table: what your split is telling you
| Your pattern vs. the benchmark | What it usually means | Where to look next |
|---|---|---|
| Traffic up, real growth above ~3% | You are gaining share | Peak-hour capacity and ticket times, so service holds up as volume grows |
| Traffic flat, check up mostly from price | Holding steady on price alone | Item-level sales after the last increase: which items lost units |
| Traffic down, check up more than price | Fewer guests spending more each (as in the example) | Daypart and day-of-week traffic, lapsed regulars, discounts and comps |
| Traffic down, check up less than price | Guests trading down and visiting less | Menu mix shift toward cheaper items; value offers; portion and price perception |
| Nominal growth below CPI menu inflation | Real sales are shrinking | Treat it as a priority: margins, labor scheduling and marketing together |
Four cautions before you act on the numbers
- Advance estimates get revised. According to the Census release's own Table 3, over the past 12 months the advance estimate of food services' month-to-month change has been revised by about 0.2 points on average. The release also notes that monthly estimates were revised on Sept 28, 2026, after the annual survey benchmark. Check the next release before you lock a target.
- Seasonal adjustment vs. your register. The 7.0% headline is seasonally adjusted. Your POS shows raw dollars, so the closer match is the unadjusted pair ($91.98 billion vs. $85.19 billion, about +8.0%). In both years September had four Fridays, four Saturdays and four Sundays, so weekends don't distort this particular comparison. That won't hold for every month.
- Use the right price index. If you run a counter-service or quick-service concept, deflate with the limited service figure (3.2%). Table service operators should use full service (3.8%). Your own menu price change matters more than either one. Pull it from your price history, not memory.
- Compare like for like. Exclude days you were closed, catering one-offs and big private events from both years, or the split will mislead you. The same goes for any summer 2026 event spikes if you extend the comparison to July and August.
A monthly routine that takes 20 minutes
- Pull net sales, check count and average check for the month and for the same month last year from your POS. KwickView, the dashboard built into KwickPOS, shows hourly, daily and weekly breakdowns, order counts and period comparisons, which gives you the raw numbers for this step.
- Write down every menu price change in the past 12 months and estimate the weighted average.
- Run the three ratios above: traffic, price and mix.
- Deflate your sales by your own price change and compare the result with the industry's real growth.
- Find your row in the decision table and pick one thing to investigate, not five.
The next data points arrive soon. BLS lists Nov 10, 2026 for the October CPI. Census has scheduled the October advance retail report for Nov 17, 2026. Running this check on October before the holiday season gives you a clean baseline for comparing November and December, when party bookings and gift-card sales make year-over-year numbers harder to read.
Frequently asked questions
How much did U.S. restaurant sales grow in September 2026?
According to the Census Bureau's advance estimate released Oct 15, 2026, food services and drinking places sales were $92.9 billion seasonally adjusted. That is up 1.1% from August 2026 and 7.0% from September 2025.
How much did menu prices rise over the same period?
BLS's September 2026 CPI shows the food away from home index up 3.5% over 12 months. Full service meals rose 3.8% and limited service meals rose 3.2%.
Why divide instead of subtract when adjusting for inflation?
Growth rates compound. Dividing (1 + sales growth) by (1 + price growth) and subtracting 1 gives the exact price-adjusted change. Subtracting is a close shortcut when both numbers are small.
Should I compare my POS numbers to the seasonally adjusted or the unadjusted Census figure?
Your register shows unadjusted dollars, so the unadjusted Census pair is the closer match: about $91.98 billion in September 2026 vs. $85.19 billion in September 2025. The seasonally adjusted figure is better for month-to-month trend comparisons.
When is the next data released?
BLS lists Nov 10, 2026 for the October 2026 CPI. Census has scheduled the October 2026 advance retail sales report for Nov 17, 2026.
Sources
- Advance Monthly Sales for Retail and Food Services, September 2026 (CB26-165), U.S. Census Bureau, 2026-10-15
- Consumer Price Index Summary, September 2026 (USDL-26-1574), U.S. Bureau of Labor Statistics, 2026-10-14
- Advance Retail Sales: Food Services and Drinking Places (RSFSDP), Federal Reserve Bank of St. Louis (FRED)
- KwickView home page (feature list), KwickView / KwickPOS
Checked November 3, 2026. Photo: RDNE Stock project / Pexels.
From the KwickOS family
Built by the team behind KwickOS restaurant platform, KwickPOS (cloud + offline hybrid POS, 5,000+ locations) and the KwickPhone AI phone agent.