When you manage dining services for a school system with nearly 200,000 students and some 250 school sites in an era of high fiscal stress, you have to be very efficient. That is the challenge faced by Mary Kate Harrison, food & nutrition services director for the Hillsborough County (FL) Schools, one of the country's ten largest districts.
Harrison is one of the new breed of school nutrition directors looking to meet the challenge of the times through a more rigorous, numbers-oriented approach to school foodservice, an approach that includes utilizing KPI (Key Performance Indicators) to evaluate the program, determine points of emphasis and gauge results in an objective, measurable way.
In the K-12 environment, KPI metrics include standard measures like food and labor cost, meals per labor hour and participation. But KPI based management is designed to go further, to drill down into those broad numbers, as well as into more arcane ones.
It requires sophisticated data gathering and analysis systems to work effectively, as well as a compatible database of comparable statistics against which they can be measured and benchmarked (for more on that, see the sidebar on p. 20). And, just as critically, it requires a savvy management acumen that can take advantage of the data to make effective and efficient changes in the program.
Making the Units Accountable
The KPI oriented approach at Hillsborough County went into high gear three years ago when Harrison decided her existing system couldn't meet the mounting challenges faced by her department.
“We'd gathered profit-loss statements for years from our schools and given them back information about how much money they had spent on food and labor, how much revenue they've brought in and so forth,” Harrison explains. “Then we would break it down into their cost per plate and either their profit or loss per plate. That was good information, but I felt it had to be more meaningful for employees to really be able to manage their programs because I think they were just looking at the numbers and saying, ‘Oh, I made money’ or ‘I lost money.’ But they didn't know how they got there, and they didn't know what to do about it if they needed to make changes.”
To make the numbers more useful, Hillsborough now has a “site progress report” (SPR) that “breaks down all those things that managers can do something about,” Harrison says. “With schools, there are certain things over which we don't have a lot of control. The school board decides what our salary line increase will be, how much we'll make and so forth. We wanted to make a point to our managers that even if there are variables that we can't control, we still have to create enough revenue to cover all of our costs.”
The SPRs provide site managers key data on revenues and expenses, along with individual site goals. The revenue metric centers around participation, which is further broken out to measure breakfast, lunch and a la carte separately. In the first year, each school site was given its own distinct, realistic goal individualized to its circumstances, historic performance and potential.
“We looked at where each was and where we thought it could go for the next year based on the percent of their students that they weren't serving,” Harrison says. “The goal was to get numbers up where they realistically could be expected to go up.”
The expectations were incremental that first year — a program with 35% breakfast participation might have been expected to bump that to 38%, for example. “In that first year, we just wanted to get them used to having goals and to understand that they had to strive to meet those goals,” Harrison explains.
On the expense side, food cost was set as a percentage of revenue. Unlike participation, it was pegged to general criteria like school size rather than being based on individual school circumstances.
To make goals clearer and simpler to understand, the data on the SPRs are couched not only in percentage terms but in raw numbers. So, for example, a manager will be given not only the participation percentage that site needs to meet, but also specifically how many additional student customers are needed to hit the goals.
“And they really get that,” Harrison says. “While I think percentages are really meaningful and paint a good picture, when you're dealing one-on-one, serving kids through a line, it may be more useful to think in terms of ‘I need to serve 10 more kids today at breakfast’ than ‘I need to increase breakfast participation three percent.’”
Another effective enabler is a calculator application on the ordering program provided to each site that lets managers take better charge of procurement costs to stay in line with their expense goals. With the program, the manager is better able to tailor orders so that they meet the site's procurement cost goals.
“For example, they input so many cases of this and so many cases of that and it will tell you that you will spend, say, $1,800,” Harrison explains. “So, if they know they have to cut say, $34 a day, they can go back and make decisions: ‘Can we get away with one case of this?’ or ‘I really don't need that right now.’”
In this way, unit managers are empowered and encouraged to find the kind of hidden incremental cost savings that would be harder to ferret out with a heavy-handed top-down approach. The goals for each school, as well as the progress toward meeting them, are posted in each kitchen and are also gone over during each day's “power rally” staff meeting. Harrison says this kind of ongoing communication also has a positive effect, not only on managers but on employees.
Continue on next page
Getting the Numbers
Harrison says just getting the needed numbers needed to implement a KPI approach was one of her initial challenges. “When I first got here, getting financial data to run this program was impossible and you kind of did it by the seat of your pants. Most of the information came through the district financial system, so you really needed to have financial expertise on your team to be able to figure it all out. But that expertise is critical if you are to make knowledge-based decisions.”
Those “knowledge-based decisions” extend much further than simple cost-cutting. In fact, the attention paid to the metrics has actually allowed the nutrition program to make investments in areas where they can make the biggest difference.
Those investments have included using more fresh and higher grade items — as well as more attractive packaging and presentation — in the meal program to encourage more participation. Another investment: a new executive chef for the district who will not only work with the recipes but look for ways to make production of higher quality dishes easier for the staff.
“During this whole thing we have not cut any kind of quality out of our program,” Harrison maintains. “But at the same time, I think we've really realized some financial savings, just because of the control factor at each site.”
Harrison says that while food cost reductions have been among the biggest changes resulting from the KPI approach in the past three years, there has also been a noticeable bump up in middle and high school participation. She attributes that to the quality improvements made possible by the KPI-enabled cost discipline.
“I'm talking about going from 45 cents a portion to 65-70 cents,” she says proudly. “We made that investment and changed a lot of the presentations, and the high- and middle-school kids responded.”
Metrics That Matter
Standard metrics like participation rates and food and labor costs — teased out into a variety of submetrics based on schools size, population, free/reduced percentage, etc. — lie at the core of KPI based management of school nutrition programs. Analysis gives managers clues to potential problem areas, says Lora Gilbert, director of food & nutrition services for the Orange County Public Schools, another large Florida district.
“For example, we want to know how many of our free students are eating breakfast at each school because, if a school can't even give away breakfast, there must be some kind of problem: the buses are late, the administration won't let them in the cafeteria, or maybe the breakfast isn't good enough to attract customers there.”
At Orange County, the KPIs being tracked include the scores from regular surprise onsite evaluations of each site.
“We take a complete look at what the operation should look like when we walk in the door,” says Gilbert. “I get a report every month of what our onsite audits are saying.” The audits evaluate things like equipment, staffing, marketing, sanitation and food safety. The scores not only track the progress of each individual site but provide basis for what areas to focus training efforts on.
Orange County also has KPIs that track broader goals like the level of community involvement. That has led to initiatives such as the lending of the department's mobile food vans to parent and school groups to use for special events and fundraisers (see photo on p. 20).
At St. Paul, MN, Public Schools, a KPI approach has been deployed to find cost savings in inventory, says Jean Ronnei, director of nutrition and commercial. “We started comparing inventory values on a monthly basis since this is obviously money sitting on the shelf,” she says.
Now, instead of the spot checks conducted previously, the department does actual inventory audits, which in turn have yielded data on the significant differences in inventory levels among different sites. That data has then been used to establish target inventory levels for all the site managers.
To help them meet those targets, managers are being encouraged to evaluate their ordering practices. “The biggest key is looking at the menu, and then looking at what you have in supply, knowing when your next order's coming in and paying more attention,” Ronnei explains.
Amy Thering, business systems manager of nutrition and commercial services for St. Paul Public Schools, says the department's inventory tracking metric is part of a new report designed specifically to be more useful to site managers than what had traditionally been provided them.
“We used to give them a monthly P&L that would include their revenue, their labor expense, their food expense and what their net was,” she says. “But we found people weren't using the report to make changes.”
Following some discussion with site managers, a new report was developed that included the kind of specific data managers find useful. These include participation, inventory value in dollars, food and supply costs, meal equivalents and health inspection data. In addition to the metrics, there are now also specific goals for each category.
At Denver Public Schools, one metric tracked is percent of meals served on time, says Leo Lesh, executive director of enterprise management. He also uses KPI metrics to judge the impact of staff reductions before deciding on sanctioning replacements.
Lesh says it's important to track metrics at both the individual site and the district level. At the site level, the numbers provide comparative data that can be used to identify substandard performers and target the areas that need improvement.
“For example,” he says, “if we are comparing meals per labor hour between two elementary schools of similar size, and one has meals per labor hour of 20 while other has 15, that gives us further evidence to drill down and see what's going on in that particular school so that we can make corrections in a timely manner.”
Or better yet, give the school itself the information, the means and the incentive to make the improvement itself.
Paving the Way
KPIs are used in many ways — not just to identify problem areas and track progress toward their correction. They also provide critical input when planning major initiatives. In Denver, Lesh and his team analyzed reams of data to justify a planned conversion to a from-scratch production system, and to anticipate its impact on operations, personnel and revenues.
“We looked at all these things and tried to do a pro-forma before we even started,” Lesh says.
Continue on next page
“What will it cost in training? What will it cost in equipment we would need but don't have? After all, when you start slicing and dicing your own fruits and vegetables, you'll need knives. We didn't have knives before, so those are the kinds of things you have to figure in. The cost may seem small, but when you buy knives for 142 schools, the costs add up.”
He also calculated anticipated changes in food and labor costs. Food costs certainly would rise some, given the emphasis on fresh ingredients and a conversion to premium items like grass-fed beef. On the other hand, supply costs would diminish since food would now be dished up fresh to trays rather than prepackaged in separate cups and dishes.
The effect on the staff was harder to anticipate. Some employees might not be able to convert to a more demanding way of doing things, and others might take their new-found skills to seek other opportunities.
In the early going, the experiment is holding up at the 30 pilot sites, with a four-percent participation increase and labor and food costs within anticipated parameters.
The key to the KPI-based approach is objective, meaningful data.
“It's the way I do my business,” says Orange County's Gilbert. “I'm constantly reminding my staff, ‘If you don't have data or you didn't write it down, then it didn't happen.’”
Once it does “happen,” and it's quantified, though, then it becomes a powerful teaching aid.
“In our staff meetings we go through some of the issues that they've come across and we solve those as a team,” Gilbert continues. “It's both a way to share best practices and a chance for my senior management team to see how other people are solving things. We're constantly looking at how we can reduce our cost per meal and find those things we're doing that we don't need to be doing.”
“What KPI does for me is that helps keep the pulse of the business and of the children. So when I'm looking at participation rates per day, for example, I know what it was last year and the year before so I can benchmark.”
“If you can't measure change, then how do you know you've made progress?” Lesh asks rhetorically. “I hate to say we run it like a business, but we do. In our case, it is a business of about $30 million, and it's easy to lose track of. Without the right data, you're left with, ‘Well, my gut feeling is that we have an increase in free/reduced participation.’ But how do you know that if you don't track the measurements?”