Restaurant Performance Index September 2026: Define Same-Store Sales by Location, Same Days, Net of Menu Price Before You Report It Up
Same-store sales measure the change in sales at locations that were open in both periods, over matching trading days, with the price component shown separately. Fifty-four percent of operators told the National Restaurant Association their September same-store sales rose. Before you say the same about your locations, define the number the same way at every store.
Why this matters now
On 2026-10-30 the National Restaurant Association published its Restaurant Performance Index for September 2026. The index stood at 100.7, up 0.1 percent from August, and readings above 100 mark expansion in the Association's framework. The Current Situation Index and the Expectations Index both read 100.7. Four in ten operators expect higher sales in six months, and 20 percent expect the overall economy to improve.
The detail behind the headline is in the Association's same-store sales and customer traffic indicator, also dated 2026-10-30. Fifty-four percent of operators said same-store sales rose between September 2025 and September 2026, up from 51 percent in August. Thirty percent said sales declined and 16 percent said they held about the same. Traffic ran the other way: 42 percent reported more customers and 44 percent reported fewer, the 19th net traffic decline in the last 20 months.
Put a price index beside those two readings. The Bureau of Labor Statistics Consumer Price Index for September 2026, released 2026-10-14, has food away from home up 3.5 percent over the year, with full service meals up 3.8 percent and limited service meals up 3.2 percent. A majority of operators selling more dollars to fewer guests, while menu prices run 3.5 percent above last year, means most of the reported increases are price. The question for an operator with more than one location is whether each store's number was built the same way, so the price, traffic, and check pieces can be compared across stores.
One more date matters for anyone running the comparison inside QuickBooks. Intuit's Fall 2026 release notes, published 2026-10-26, say reports now keep customizations, automatically pick the right prior period for comparisons, and show budget reports that match expectations. The note does not describe how the automatic prior period treats a 4-week or 5-4-4 restaurant calendar, so the comparison range on the report header is something to read, not assume, before a QuickBooks comparison column becomes your same-store figure.
What same-store sales has to mean before anyone computes it
A self-contained definition first. Same-store sales, also called comparable sales, is the percentage change in net sales at the set of locations that were open and trading normally in both the current period and the comparison period, measured over matching trading days. Locations opened, closed, or shut for remodel in either period are excluded. The metric exists to separate growth at existing stores from growth that came from adding stores.
That one sentence hides five decisions. A group with three locations and three managers will make them three different ways unless someone writes them down.
| Decision | Common default | What it does to the number | Group rule |
|---|---|---|---|
| Comparable set | Every store currently open | A store opened in March 2026 shows runaway growth; a store shut two weeks for a hood fire shows a collapse | Open and trading normally for all of both periods; name each excluded store and the reason, every period |
| Period | Calendar month as QuickBooks reports it | Weekend counts shift year to year; October 2026 has five Saturdays, October 2025 had four | Compare like days: the same 4-week period, or the calendar month adjusted for the day-of-week swap |
| Sales base | Whatever the sales accounts total | Comps, discounts, delivery commissions, and gift card loads land differently by store | Net sales: gross less comps and discounts, delivery at gross with commission as expense, gift card redemptions in and loads out |
| Price | Ignored | A 3.5 percent menu increase reads as growth | Log each menu price change with its effective date and compute the weighted price effect per location |
| Channel | Blended | A store that added third-party delivery shows traffic growth the dining room never saw | Report dine-in, takeout, and delivery separately, then total |
The table is the whole control. Once those five answers exist in writing, the same-store number from the downtown store can sit next to the number from the airport store and mean the same thing.
One location, October worked through
Call it Oak Street, a full service location open since 2022. The figures are illustrative. The method is the point. The bookkeeper pulls October 2025 and October 2026 from the daily sales summaries posted to the general ledger, one entry per day, and from the POS check counts for the same days.
Step one: the headline
October 2025 net sales were $212,000 on 12,400 checks. October 2026 net sales were $224,700 on 12,250 checks. The headline is 6.0 percent same-store sales growth. Checks fell 1.2 percent. Average check rose from $17.10 to $18.34, or 7.3 percent.
Step two: the calendar
Both Octobers have 31 days, but October 2025 started on a Wednesday, so it held five Wednesdays, five Thursdays, and five Fridays. October 2026 started on a Thursday: five Thursdays, five Fridays, and five Saturdays. The 2026 month swapped a Wednesday for a Saturday. At Oak Street a Saturday averages $9,800 on 520 checks and a Wednesday averages $5,200 on 300 checks. On a same-day basis, October 2026 is $220,100 on 12,030 checks. Growth drops from 6.0 percent to 3.8 percent, and checks are down 3.0 percent rather than 1.2 percent. The calendar alone was worth 2.2 points.
Step three: price
The menu price log shows a 4 percent increase on entrees effective February 2026 and no change on beverages. Weighted by Oak Street's own sales mix, the price effect is 3.5 percent, which lands near the national food away from home figure. Use the log, not the index. The index is the check, not the input.
Step four: the split
Same-day growth of 3.8 percent decomposes multiplicatively. Price contributed 3.5 percent. Checks contributed negative 3.0 percent. What remains is check mix: 1.038 divided by 1.035 divided by 0.970 is 1.034, so the guests who came spent 3.4 percent more per check before price, through attach rate or trading up. The one-line summary for the owner reads: Oak Street, October, same-store sales up 3.8 percent on matched days, of which price 3.5, traffic negative 3.0, mix 3.4. The unadjusted 6.0 percent included 2.2 points of calendar.
Run the same four steps for the other two locations and the comparison finally says something. A store showing 6 percent growth with traffic up and a store showing 6 percent growth with traffic down 5 percent are not performing the same, and only the split shows it. The daily sales entry that feeds this is the one described in the POS reconciliation guide, and the per-location structure it rolls into is in the multi-location restaurant P&L guide.
Accounting controls for same-store sales
Source data: the daily sales summary posted to the general ledger for each location and each day, with gross sales, comps, discounts, sales tax, tips, and gift card loads and redemptions on separate lines; the POS check count and channel split for the same days, dine-in, takeout, and delivery; the menu price change log with effective dates and the items affected; each location's opening date and any closure days with the reason; and the written comparable-set rule.
Calculation: a fixed sequence anyone can re-perform. Net sales equal gross sales less comps and discounts, with gift card redemptions included and loads excluded. Sales are a flow for the period. The gift card balance and tips payable are point-in-time liabilities that the comparison never touches. Polarity is the first check: sales are credits, comps and discounts are debits to contra-revenue, and a comp posted as an expense instead of contra-revenue inflates net sales at that store and nowhere else. Match trading days by swapping out the day-of-week difference at that store's own daily averages. Compute the price effect from the log, weighted by the store's own mix. Then split same-day growth into price, checks, and check mix multiplicatively, so the three factors multiply back to the matched-day growth.
Review: a second person rebuilds one location from the daily entries and lands within rounding; the sum of daily net sales agrees to the general ledger sales accounts for the period, by class or by entity; check counts agree to the POS period report; every excluded location has a written reason; the price effect ties to the dated log and not to an index; and any location whose unadjusted and matched-day growth differ by more than two points carries a one-line explanation naming the calendar swap.
Decision: the owner or operating partner sets the comparable-set rule and the period convention once a year and does not change them mid-year. Each general manager owns the daily sales entry and the price log for their store. Finance owns the worksheet, the split, and the monthly comparison page, which is the same page the food cost percentage test gets read against when margin moves.
Limits and common mistakes
The Association's figures are shares of operators, not sales growth rates. Fifty-four percent reporting higher same-store sales says nothing about how much higher, and nothing about your market. The CPI food away from home figure is a national average across every restaurant type, so your price log is the input and the index is only a sanity check. Oak Street's numbers are illustrative.
The first mistake is comparing calendar months without matching days. A fifth Saturday is worth about two points at a full service location, and it reverses the next year.
The second is leaving new and closed stores in the set. A store opened mid-period shows growth that belongs on the expansion line, not in same-store performance.
The third is counting gift card loads as sales in December and redemptions as nothing in January. Loads are a liability when sold and sales when redeemed, and a store with a strong corporate gift program will show a December spike and a January hole unless the entry is right.
The fourth is treating a third-party delivery channel added this year as traffic growth. Report channels separately, decide once whether delivery is recorded gross with commission as an expense or net of commission, then apply that at every store.
The fifth is accepting a report comparison column without reading its date range. An automatically selected prior period is only the right period if it matches your trading calendar, and an operator on 4-week periods has to confirm that on every report.
Frequently Asked Questions
What did the September 2026 Restaurant Performance Index say about same-store sales?
The index released October 30, 2026 stood at 100.7, up 0.1 percent from August. Fifty-four percent of operators said their same-store sales rose between September 2025 and September 2026, up from 51 percent in August, while 30 percent said sales declined. Customer traffic was net negative, with 42 percent reporting more customers and 44 percent reporting fewer.
How do you calculate same-store sales for a restaurant?
Take net sales for locations that were open and trading normally in both periods, match the trading days so each period holds the same count of each weekday, and divide the current period by the comparison period. Then split the change into menu price, check count, and check mix so the growth that is price can be seen separately from the growth that is guests.
Why does the day-of-week calendar change same-store sales growth?
Because weekend days sell more than midweek days. October 2026 has five Saturdays and October 2025 had four, so a calendar-month comparison credits 2026 with an extra Saturday. At a full service location that swap alone can add about two points of growth that reverses the following year. Compare 4-week periods or adjust the month for the swapped days.
Should menu price increases be included in same-store sales?
Report the total, then show the price effect as its own line. Food away from home prices were up 3.5 percent over the year to September 2026, so a store reporting 3.5 percent same-store sales growth with no price adjustment sold about the same volume to about the same guests. Compute the price effect from your own dated menu price log, weighted by each store's mix.
How should a multi-location operator compare same-store sales across stores?
Write one rule for the comparable set, the period, the sales base, the price effect, and the channel split, and apply it at every location. Then put the stores side by side with the same four lines each: matched-day growth, price, check count, and check mix. Two stores with the same headline growth and opposite traffic are not performing the same, and only the split shows it.
Once every location's same-store number is built the same way, the monthly comparison stops being an argument about whose month had the extra Saturday. FinBoard reads each location's QuickBooks file, or each class in a shared file, into one view with prior-period and prior-year columns on the same page, so matched-day sales by store sit next to the ledger they came from and every figure drills to the daily entry behind it. See how FinBoard puts every location's sales comparison on one page.
