TL;DR: The US restaurant industry just posted its best revenue year on record. And 42% of operators didn't make a dime — up from 29% the year before, according to the National Restaurant Association's 2026 State of the Industry report. The mathematical fingerprint of who survived and who didn't comes down to a small handful of operating ratios most independents only see at tax time. The 58% who made money saw them every day. The 42% who didn't are about to learn what prime cost is the hard way.
A record year that wasn't
In February 2026, the National Restaurant Association published its 2026 State of the Industry report. Two numbers from that report are now circulating in every operator group chat in the country.
The first: the US restaurant industry generated more topline revenue in 2025 than in any year before it. Sales are projected to hit $1.55T in 2026, up from a record-setting 2025.
The second: 42% of operators told the NRA their business was not profitable in 2025. That's up from 29% the prior year. Sixty percent of operators said conditions deteriorated. Only 15% said they were better off than the year before.
Both numbers are true at the same time. That's the whole story of 2025.
The industry sold more food than ever. The industry made less money doing it than at any point in the last decade. Coverage in Restaurant Business Online and NRN has the same headline in different words: persistent cost increases, enduring demand, and a growing share of operators who watched the year close out and realized that "busy" did not mean "winning."
Your competitors aren't lying when they tell you they had a record sales year. They're also not lying when they tell you they're behind on rent. Both can be true. For 42% of independent restaurants, both are true.
The 8.7-point gap
If you want to know which side of that 42/58 line a restaurant ended up on, the cleanest single predictor is labor cost as a percentage of sales.
Toast's 2025 industry data — pulled from a sample of tens of thousands of restaurants across the country — shows it bluntly. Profitable full-service operators averaged 34.2% labor cost. Unprofitable full-service operators averaged 42.9% labor cost.
The gap is 8.7 percentage points.
On a $1.5M restaurant, 8.7 points of labor cost is roughly $130,000 a year. That's the difference between writing yourself a paycheck and not. That's the difference between paying down debt and refinancing. That's the difference between staying open in 2026 and quietly handing the lease back to your landlord.
The food cost gap is similar in magnitude. Profitable independents tend to land between 28% and 32% of sales on food. Unprofitable ones drift up to 33%, 35%, sometimes higher — usually without realizing it until quarter-end. Eight-tenths of a point on food cost is real money. Three points on food cost over a year is the year.
Those two lines added together — food + labor — are what restaurant accountants call prime cost. Most healthy independent full-service shops run prime cost between 60% and 65% of sales. The 42% who lost money in 2025 generally pushed prime cost north of 70% without seeing it happen.
The arithmetic is unforgiving. Once your prime cost crosses 70%, the remaining 30% of every dollar has to cover rent, utilities, insurance, debt service, marketing, and ownership pay. It doesn't.
Why most operators don't see it until it's too late
If prime cost is the single number that predicts whether a restaurant survives the year, why don't more independents watch it?
Because most of them can't.
A typical independent full-service restaurant ends each month by handing receipts and timecards to a bookkeeper or an accountant. Those numbers come back as a P&L six to eight weeks later. By then, the month is two months in the past. By the time you see September's prime cost, October is already over and November is half-baked.
This is not a complaint about bookkeepers. They do exactly the work they're hired to do. The problem is that the operating cadence of an independent restaurant — daily covers, daily food deliveries, daily shift schedules — runs on a clock measured in hours, while the visibility cadence runs on a clock measured in weeks.
You make ten labor decisions a week. The data telling you whether those decisions worked arrives a month and a half later.
Compare that to how chains operate. A Domino's franchise running on PULSE — the operating system the company has used internally since the early 2000s — sees food cost % and labor cost % update in real time, by daypart, on a screen in the office. A Starbucks store manager sees the same. Chick-fil-A's operators do too. The corporate ones, the franchised ones, all of them.
When labor cost spikes 3 points on a Tuesday lunch because someone over-scheduled, a chain operator sees it on Tuesday and adjusts on Wednesday. When food cost drifts up because a supplier raised wholesale beef 8% and the prep team didn't notice the bigger waste in the bigger portions, the chain catches it inside a week.
Independents catch the same thing inside two months.
That gap — between when a problem starts and when an owner can see it — is what produces the 42%.
The three ratios the surviving 58% watched daily
Talk to operators who closed 2025 in the black, and you find a remarkably consistent answer about what they pay attention to. It's almost never one number. It's three, and they're all percentages of sales.
Food cost percentage. Total cost of goods sold divided by sales. Healthy independents run 28%-32% in full-service, slightly higher in concept-driven shops, slightly lower in pizza or QSR. The trick is not the level. The trick is catching the drift. Food cost rarely jumps. It creeps. A supplier raises chicken thigh by 6% in March. Your prep team's portions shift by half an ounce. A new sauce gets added that costs 30 cents more per plate. By August you're at 34% and you haven't noticed.
Labor cost percentage. Total labor (wages + payroll taxes + benefits) divided by sales. The 8.7-point gap. The single biggest predictor. Labor drifts the same way food does — up half a point in April when a manager promotes a server to lead, up another half point in June when summer demand drops but staffing doesn't, up another point in September when a long-tenured employee gets a raise that nobody re-baselined against the schedule.
Prime cost percentage. Food plus labor, as a single number. This is the one most independents have never calculated for themselves. Prime cost is the line that decides everything else. If prime cost is at 60%, you have 40 points to play with. If prime cost is at 72%, you have 28 points to cover everything else, and "everything else" is rent + utilities + insurance + processing fees + software + debt service + your own pay. It can't.
Operators who watched these three numbers in 2025 — daily, not monthly — saw the cost squeeze coming and adjusted: trimmed labor variance, repriced the items where ingredient inflation crossed 10%, killed two underperformers from the menu, switched a supplier when a relationship turned predatory.
Operators who only saw their P&L six weeks late made all the same mistakes for an extra six weeks.
The mistake that produces "busy but broke"
Most operators in the 42% don't go broke from a single bad decision. They go broke from a hundred reasonable decisions made without the right data.
The most common failure mode of 2025 was a blanket menu price increase. An operator saw food cost rising in late 2024, applied a 5%-8% across-the-board hike to the menu, and assumed margins would normalize. They didn't, because the price increase was uniform but the cost increases weren't. The three or four items that actually drove the food cost spike were each up 10%-25%. The other items were flat. The blanket hike under-priced the bleeders and over-priced the items customers were already buying — pushing volume away from the high-margin winners and toward the low-margin losers.
Sales went up. Profit went down. We covered this exact mechanic yesterday on the blog.
The second most common failure mode was over-scheduling labor against forecasted volume that never showed up. The 2025 customer count actually dropped year-over-year for many independents — 40% of consumers told NRA pollsters they were cutting back on restaurant visits. Operators who scheduled like 2024's traffic but priced like 2025's still ended up with too many hands on a slow shift. That's the 42.9% labor cost showing up as a number on a spreadsheet.
The third most common failure mode was treating the P&L as a quarterly story instead of a daily one. By the time the Q3 P&L arrives in mid-October, Q3 has already happened. Q4 is the only thing you can still affect. And Q4 is the holiday quarter — the one where over-staffing and under-pricing compound the fastest.
None of these failures is dramatic. None of them shows up as a single bad day. They show up as a slow march of half-points across forty-eight months until the year closes and the accountant says, "you broke even."
What changes when you watch prime cost daily
When prime cost lives on a dashboard that updates every shift, the operator's job changes from forensic accounting to forward steering.
Instead of finding out in late November that October's labor was 41%, you see Tuesday's labor at 41% on Tuesday afternoon and cut a half-shift on Wednesday. Instead of waiting until quarter-end to find out that food cost crept up two points, you see chicken thigh's contribution margin slip on Friday's P-Mix report and renegotiate the supplier on Monday. Instead of running a blanket 8% price hike, you run a surgical 18% hike on the three items that actually drove the cost increase and leave the rest of the menu alone — protecting the items that customers were already choosing because they like the price.
The math doesn't change. The clock changes.
A restaurant catching a 3-point drift inside a week loses three days of margin. A restaurant catching the same drift inside a quarter loses ninety. Multiply that across labor, food, and the long tail of small line items — packaging, paper goods, third-party delivery commissions, processing fees — and you have the entire 8.7-point labor gap, in arithmetic form.
Chains have known this since the 1990s. PULSE, R365, Ctuit, Compeat, MarketMan — every chain operator in America has had real-time prime cost visibility for at least a decade. Most independents still don't. That's the asymmetry that produced the 42%.
What the 58% will do differently in 2026
If 2025's lesson was "record sales don't guarantee a record year," 2026's lesson will be "the operators who win this year are the ones who can see what's happening in their business before the month closes."
That doesn't have to mean enterprise software. It does have to mean three things.
First, item-level food cost tracking. Every plate has a recipe. Every recipe has a cost. When a supplier raises a price, you should know which items are affected and by how much that same week — not at quarter-end. The technology to do this has been commodity for ten years. Most independents still don't run it because the tools that ship it are priced for chains and packaged for chains.
Second, hourly labor visibility against forecasted demand. You don't need an AI. You need a forecast that updates against actual covers and an alert when scheduled hours diverge from the forecast by more than 10%. That's a one-page dashboard. Every chain has it. Most independents have a clipboard and an instinct.
Third, a prime cost dashboard the owner can read in 30 seconds. Not a P&L. Not a thirty-tab Google Sheet. A single screen — food cost %, labor cost %, prime cost %, sales, covers — refreshed daily. The chains call it the "today screen." Independents who built one in 2025 didn't end up in the 42%.
This is precisely what KitchenRush's Intelligence Layer is built to do for independents. The Intelligence Layer pulls in your POS data, your inventory counts, your time cards, and your supplier invoices, and it computes the three ratios that actually decide whether you'll be in this year's 58% or this year's 42%. It runs on a per-tenant, real-time basis — not a once-a-month basis — and it surfaces drift before it becomes loss.
That isn't a marketing claim. It's the literal technology. The chains have had this for thirty years. Independents are about to.
The bottom line
42% of US restaurants told the NRA they made no money in 2025. The industry's revenue was at an all-time high while almost half its operators couldn't pay themselves. The single sharpest predictor of who lost money was an 8.7-point labor cost gap. The mechanical reason for that gap was visibility — chains see prime cost in real time, most independents see it once a quarter.
If you don't know what your prime cost was last Friday, you have one of the two operating problems that defined the 42% in 2025. The fix isn't bigger prices. It isn't a bigger marketing budget. The fix is the same data the chains have been quietly running on for a generation.
Pulled the right way, your existing POS, your existing time clock, and your existing supplier invoices already contain everything you need to compute prime cost daily. The work is putting it on one screen.
---
Run a free Pulse Check on your restaurant at kitchenrush.app/pulse and we'll show you, on a single screen, where your prime cost actually sits today — no signup required.
