You reduce staff turnover in a food business by fixing the systems that push people out: unpredictable scheduling, vague onboarding, and unclear roles. These three drive most voluntary quits, and fixing them retains staff better than raises, which matters because each departure costs 50 to 200 percent of that role's annual salary.
Restaurant turnover runs brutal, often well over 70 percent a year. Owners blame the labor market, the generation, the pay. In my kitchens, turnover dropped when I stopped blaming people and started fixing systems. Bad systems make good people quit. Good systems make average people stay and improve. That is the whole game. The moment you accept that the role is churning and not the people, you stop hiring your way out of the problem and start engineering your way out of it.
Turnover Is a Systems Problem, Not a People Problem
When the same role keeps churning, the role is the problem, not the people who keep leaving it. Three system failures cause most of it, and all three are inside your control. If you have replaced the same dish station four times in a year, the dish station is broken, not the four people. Maybe the schedule for that role is always the one that changes, or the training is always rushed, or nobody ever told that person what good looked like. Fix the role and the next person stays.
- Unpredictable scheduling that wrecks people's lives outside work
- Onboarding that leaves new hires lost and embarrassed
- Unclear roles where nobody knows what they truly own
- Managers who lead by yelling instead of by system
What Turnover Actually Costs You
Owners underestimate this badly because most of the cost is invisible. It is not just the job ad. It is the lost productivity, the overtime to cover the gap, the training time, and the errors from a short-staffed line. The visible cost is the tip of the iceberg, and the part below the waterline is two or three times bigger.
| Cost driver | Per departure | Notes |
|---|---|---|
| Recruiting and ads | 500 to 2,000 | Direct, visible |
| Training and ramp | 1,000 to 3,000 | 4 to 8 weeks to productive |
| Lost productivity and overtime | 2,000 to 5,000 | Hidden, largest |
| Total per hourly departure | 3,500 to 10,000 | 50 to 200 percent of salary |
The row most owners ignore is the third one, lost productivity and overtime, even though it is the largest. When someone leaves, the work does not pause politely until you find a replacement. The rest of the team absorbs it, usually on overtime, while the line runs slower and wastes more food short-handed. Then the new hire ramps for six to eight weeks at reduced output. That entire stretch, from the day someone gives notice to the day their replacement is fully productive, is a productivity sinkhole that never shows up as a clean line item. It is the part of turnover that quietly eats your margin while your P and L looks fine, which is exactly why owners chronically underestimate what each departure really costs.
The Three Fixes That Move the Needle
Predictable schedules posted two weeks out. A written onboarding playbook per station. A one-page role definition so everyone knows what they own. None of these cost a raise, and together they outperform throwing money at wages. Here is the order I would install them, fastest win first.
- Move your schedule to a two-week posting window this cycle, the fastest free win.
- Write a one-page playbook for your highest-turnover station, then the next two.
- Give every role a one-page definition of what they own and who they report to.
- Assign every new hire a named mentor for their first two weeks.
- Run a 30-day and 60-day check-in to catch quiet quitters before they leave.
A Worked Example of the Turnaround
One location I worked with was running 85 percent annual turnover on a 22-person hourly team, roughly 19 departures a year. At a conservative 5,000 dollars each, that was 95,000 dollars a year walking out the door. We did not touch wages. We posted schedules two weeks out, wrote playbooks for the four busiest stations, and gave every role a one-page definition. Within two quarters, turnover fell to around 50 percent, about 11 departures. That is 8 fewer departures a year, roughly 40,000 dollars saved annually, from changes that cost a few weekends of writing and a discipline about the schedule.
Put a real dollar figure on your turnover before you spend another cent on ads.
Calculate your turnover costWhy Raises Alone Do Not Stop the Bleeding
A raise feels good for a paycheck or two, then the same broken schedule and the same role confusion wear people down again. You cannot out-pay a bad system. Stability and clarity are what people actually stay for, and they are far cheaper than a permanent wage spiral. A raise is also permanent overhead that compounds with every hire, while a two-week schedule is a one-time discipline change that costs nothing and keeps paying back forever.
Find Your One Worst Role First
When turnover is high everywhere, the temptation is to fix everything at once, which usually means fixing nothing well. Instead, find your single worst role, the one position that churns more than any other, and fix that one completely before moving on. In most kitchens it is dish or prep, the roles with the least training, the worst schedules, and the vaguest expectations. Pour all your systems energy into that one role: write its playbook, give it a predictable schedule, define exactly what it owns, assign its new hires a mentor. Get the churn on that one role down, then take the same playbook to the next worst role.
This approach works because turnover is rarely uniform. It clusters in a few roles that have been neglected the longest, and those few roles often account for the majority of your departures and your hidden cost. Fixing your worst role first gives you a visible, fast win that funds the patience to fix the rest. Trying to boil the ocean gives you a half-finished mess across ten roles and no clear proof that any of it worked. The fast win matters for momentum, too. When the team sees that the dish station finally stopped churning and the people there are happier, they believe the next fix is worth supporting. Change in a restaurant lives or dies on whether the crew believes it is real, and a single completed turnaround is the proof that earns you the goodwill to keep going.
The Common Mistake: Treating the Exit, Not the Entrance
When people quit, owners pour energy into the exit: counteroffers, exit interviews, scrambling to backfill. The leverage is at the entrance and the first month, not the exit. By the time someone gives notice, the system already failed them weeks ago. Spend your attention on the first 30 days, where a good schedule, real onboarding, and a clear role decide whether this person becomes a two-year veteran or a 90-day statistic. The cheapest departure to prevent is the one that never reaches the exit conversation.
The Bottom Line
Turnover is not a hiring market you are stuck with. It is a systems leak you can close. Fix scheduling, onboarding, and role clarity before you touch wages, and watch the same people who used to quit decide to stay. Start by measuring what the leak is costing you, then plug the biggest hole first. The labor market is the same for the place across the street that keeps its people. The difference is the system.