What "Labor Cost Percentage" Is Actually Measuring (And What It Isn't)
Before you can fix it, you need to know what you're actually looking at.
Your labor cost percentage is your total labor expense divided by your net sales, expressed as a percent. Simple enough. But that single number is a blended average across every hour you operated, every position on your floor and line, and every shift across your entire week.
A 34% labor cost at 11am on a Tuesday looks nothing like a 34% labor cost at 7pm on a Saturday. One of those is a problem. One of those is just the cost of opening your doors.
The weekly number tells you that something is off. It does not tell you where, when, or why. And until you know those three things, every "fix" is just a guess.
The Four Places Labor Cost Actually Creeps
In our experience auditing Toast data across restaurant concepts, labor cost creep concentrates in four areas. They're not exotic. They're hiding in reports you probably already have access to.
- Daypart inefficiency — the gap between your schedule and your covers. Your slowest daypart is almost always your most expensive in terms of labor percentage. You know this in your gut. What you may not know is exactly how wide that gap is, or whether it's gotten worse. In Toast, you can run a sales summary by daypart alongside your labor hours for the same window. What you're looking for is cover count per labor hour — how many guests your team is actually serving per hour they're on the clock. When that number drops and your staffing doesn't flex with it, your labor percentage climbs. The problem isn't usually a manager making bad calls. It's that the schedule was built around a sales projection that no longer matches reality, and nobody flagged it because the daily labor dollar number looked acceptable.
- Overtime bleeding through without triggering alarms. Overtime is the quietest tax on your labor line. A server or line cook hitting 38–40 hours across a week doesn't feel like a crisis — until you're paying time-and-a-half on eight hours of labor you could have spread differently. Toast captures actual clock-in/clock-out data. Pull your labor detail report, sort by hours worked, and look at who's consistently landing between 38 and 44 hours. In most cases, there's a simple scheduling adjustment — a shift trade here, a rotating off-day there — that drops those people back under 38 and recovers the overage. We've seen this single adjustment move labor percentage by half a point at a two-location concept. Across twelve months, that's real money.
- Manager and admin labor that's invisible in the weekly review. Salaried manager labor often doesn't get included in the labor percentage conversation because it's a fixed cost. And if your managers are driving efficient operations, that's reasonable. But here's what we see frequently: operators don't know how much hourly admin and off-floor labor they're accruing. Training hours. Pre-open prep time. Closing paperwork. Side work that runs long because someone missed a task. This labor shows up in your clock-in data, but because it doesn't map cleanly to a shift or a cover count, it gets absorbed into the weekly total without scrutiny. Ask yourself: do you know what percentage of your total clocked hours last week were productive service hours versus ancillary labor? Toast knows.
- Tip share and back-of-house pay structure miscalibration. This one is more structural, and it doesn't fix quickly, but it's worth naming. As tip share arrangements have evolved — particularly tip pooling models that bring BOH into the calculation — many operators have added real labor cost without fully accounting for it in their margin models. If you've changed your tip structure in the last two years, go back and look at your total labor percentage trend from that inflection point forward. We've seen concepts absorb a full percentage point of labor cost from a tip share change they made in good faith, without recognizing the impact until they did a backward-looking audit.
How to Pull the Right Labor Report in Toast
Here's the short version. In your Toast back-end:
Start with a trailing 4-week view
Look at labor percentage by daypart, not just by day
Then: Reports → Labor → Employee Detail
Pull the same 4-week period
Sort by total hours — flag anyone over 38 hours in any given week
Cross-reference those two reports against your Sales Summary by Daypart. You're building a picture: when did you have the most labor on the floor, and how did guest volume match it?
If you've never done this side-by-side comparison before, it's going to be clarifying. You will almost certainly find one or two dayparts where your labor percentage is running 6–10 points higher than your overall average. That is your leverage point.
What Good Looks Like (And What the Benchmarks Actually Mean)
You've probably heard 30% as the benchmark for labor cost. Maybe 28% for fast casual, 35% for full-service.
Treat those numbers with appropriate skepticism. They're averages across concepts with vastly different ticket averages, service models, and market wage rates. An independent fine dining operator in a major metro running 38% labor might be more efficient than a casual counter-service concept in a secondary market running 30%.
The benchmark that matters is your benchmark — what your labor percentage looks like when your volume is right, your schedule is tight, and nobody's on overtime.
And if you've never established that baseline, that's the first thing to build. Pull your best four-week stretch from the last eighteen months — the one where everything felt like it was clicking — and use those numbers as your target. Toast has all of it.
The One Thing That Won't Fix This
Cutting staff.
This is where operators go first, and it's almost always the wrong move. Under-staffing doesn't reduce labor cost percentage — it reduces sales, which makes labor percentage worse while also degrading the guest experience and burning out your people.
The fix for labor cost creep is precision, not reduction. It's closing the gap between when your labor is deployed and when your guests actually show up. That requires data. It requires running the reports. And it requires being honest about which dayparts you're subsidizing out of habit rather than necessity.
we look through in every audit.
In most engagements, it's also the one where operators find the fastest recovery — not because labor was poorly managed, but because nobody had sat down and run these reports side by side before. A 30-minute conversation will tell you whether an audit makes sense for your operation. No pitch, no proposal on the first call.
Book a Free Discovery Call →