Twenty percent. That's the commonly accepted percentage of labor costs to sales in the restaurant industry. It's easy arithmetic:

Sales ÷ payroll = percentage of labor costs to sales

for any given time period.

Sales is the unruly member of the equation. It changes season to season, day to day, hour to hour. Payroll is much more well behaved, much more predictable. And it's that part of the calculation we're going to talk about today: managing labor costs.

Labor costs are best controlled when:

• Performance is monitored in deep detail.
• It's based on accurate sales forecasts (a subject unto itself – stay tuned, we’ll publish a post on this topic in the next few weeks).
• Staffing is precisely matched to that forecast.
• Staff knows how—or is taught—to do more than one job at a time.

Let's start with performance. Labor costs are controlled when time and attendance, length of breaks, sales per shift, preparation time—you know the list—are all tracked and reported. Of course, you have to instantly spot and terminate non performers. But correctable weaknesses (like a reluctance to upsell) need be identified and remedied. And you have to take preventive actions as well. Think about time clock functions, for instance, like enforcing lockouts to prevent employees from clocking in before start of shift.

Accurate forecasting is detailed forecasting. Even end-of-day numbers aren't good enough. You have to know sales volume by time of day—that way you have the right coverage on the floor all the time instead of most of the time. You have to know sales volume by type of meals—that way you don't have the wrong chefs in the kitchen. You have to know the sales volume at the bar—that way you don't leave a bartender with little to do but wash clean glasses.

Controlling costs isn't just about how many people you schedule at any given time: it's also about which people you schedule. Many restaurants take the easy approach: put their A Team on the floor at the same time during peak periods. That's shortsighted though: the only way for newer staff to learn the ropes is under pressure, with top performers to guide them. The more strategic approach, then, is to have the right mix of skills on the floor—some of your best working alongside newer staff.

That works for peak periods, but what about during slow periods? To some extent the same principle applies: now some of your promising staff mentors trainees. The issue expands though; it's not just about performance now, but skills as well. Key to labor management is having staff that can handle more than one type of job. When bar sales are slow, schedule servers who can mix drinks or greet and seat. Turn bussers into runners.

Skills are not just about what someone knows: it's also about what someone can learn. Once again analyzing labor management data, you can easily spot, for each employee, where existing skills need strengthening or new ones can be developed. If a server is consistently rated high in customer feedback, teach them to handle the podium. A prep chef rated "quick to learn" can be taught to handle the grill. Just as important, you need to know—in fact you need to be alerted—to employees for whom additional training would have little value.

Approaching the magic 20% is done with data, analysis, monitoring, strategic scheduling and more. It's about knowing the details, and knowing how to use them.

Want to learn more? We've got a whitepaper that delves more deeply into the topic, which you can download below.