Restaurant KPIs: the numbers a franchise owner should watch

The restaurant KPIs a franchise owner should watch each week are net sales, orders, average check, prime cost, food cost, hourly labor percentage, sales per labor hour, customer rating, speed of service and catering or channel sales. That is ten numbers. Read each one against the store's own last few weeks and against similar stores in the brand, and you will know where to look before the P&L arrives.

A KPI, or key performance indicator, is just a number you have decided to watch. The hard part is not finding numbers. Your POS has hundreds. The hard part is picking the few that tell you what changed and what to do about it.

Why fewer KPIs work better

A scorecard with 40 lines does not get read. Whoever runs the store skims it, nothing stands out, and the one number that mattered is lost in the middle.

Pick the numbers that answer three questions:

1. Are more people coming in?
2. Are they spending more?
3. Is the store keeping enough of each dollar?

Everything else is a detail you look up when one of those answers changes.

Sales KPIs: traffic, check and mix

Net sales and same-store sales

Net sales is sales after discounts, before tax and tips. Same-store sales compares a store with itself a year ago, which removes the effect of new openings. It is the clearest single read on whether a store is growing. Our guide on how to increase same-store sales covers what moves it.

Orders

Orders, or transactions, is your traffic count. It is the number most owners skip and the one that warns earliest. Sales can hold up for weeks on price increases while fewer people come through the door.

Average check

Average check is net sales divided by orders. Read it next to orders, never alone. A rising check with falling orders often just means the small orders went away. See how to increase average check for the full method.

Channel and catering sales

Split sales by channel: counter, drive-thru, your own online ordering, third-party delivery and catering. Each has a different check size and a different cost. Catering is worth its own line because a single order can be worth a busy hour at the counter.

Cost KPIs: food, labor and prime cost

Food cost percentage

Food cost percentage is cost of goods sold divided by net sales. Weekly, most stores use purchases plus a count of high-cost items. Monthly, use a full inventory count.

Hourly labor percentage and sales per labor hour

Hourly labor percentage is hourly wages divided by net sales, straight from the POS or scheduling app. It is not the same as labor on the P&L, which adds salaries, payroll taxes and benefits. Keep the two apart and label them clearly.

Sales per labor hour is net sales divided by hours worked. It tells you whether the schedule followed the business. Our guide to sales per labor hour shows how to use it to set next week's schedule.

Prime cost

Prime cost is cost of goods sold plus total labor, as a share of net sales. It is the biggest controllable cost on the P&L. Many operators aim for somewhere near 60%, though the right number depends on the concept. Our guide on how to calculate prime cost walks through it.

Guest KPIs: rating and speed

Customer rating

Track your Google rating over the last 90 days, not the all-time number. The all-time rating barely moves. The recent one tells you whether something changed this month. Watch the number of new reviews too, since a rating built on five reviews means little.

Speed of service

For drive-thru and counter concepts, time from order to hand-off at the peak hour is a sales number as much as a guest number. A slow line at lunch caps how many orders a store can take.

People KPIs belong on the monthly view

Turnover, time to fill open shifts and how long the store's manager has been in the role all matter. They move slowly, so a monthly check is enough. When they go wrong, they show up first in the weekly numbers, as rising labor, slipping speed or a softer rating.

A weekly scorecard example

Here is one week at Switchback Grill in Provo, where Mike is the GM, against the store's own eight-week average.

Read it as a story, not as nine separate numbers.

• Sales fell 2.5%, but orders fell nearly twice as fast. The check is holding sales up, so traffic is the real story.
• Labor hours went up while sales went down. That is why hourly labor rose almost two points and sales per labor hour fell.
• Catering fell by $410, which is more than half of the $700 sales gap.
• Food cost and rating moved a little. Not enough to act on this week.

So the week has two clear things to work on. Put next week's schedule closer to the sales forecast. And call the store's regular catering accounts to get them booked again. Everything else waits.

Compare against the right baseline

A KPI without a comparison is just a number. 24.1% hourly labor is fine at one store and a problem at another. Use three comparisons:

1. The store against its own recent weeks. This is the best baseline, because it is the same building, menu and market.
2. The store against the same week last year. This handles seasons and holidays.
3. The store against the middle store in its brand. This shows what is possible with the same menu and systems.

Avoid comparing a store only with the system average. A few very large or very small stores drag the average around. The middle store is a steadier yardstick. Franchisors comparing many stores can read more in our guide on how to benchmark franchisee performance.

Common KPI mistakes

• Tracking too many numbers. If a line has not changed a decision in three months, drop it.
• Reading single days. A rainy Tuesday tells you nothing. Use weeks.
• Chasing one ratio. Cutting hours to hit a labor target can cost more in lost orders than it saves.
• Mixing hourly labor and P&L labor. They measure different things.
• Averaging across stores. A five-store average can look fine while one store struggles.
• Numbers nobody owns. Every KPI on the scorecard should have someone who acts on it, whether that is the owner or the GM.

A simple weekly habit

1. Every Monday morning, pull the scorecard for each store: last week against the eight-week average.
2. Circle any number that moved more than its usual swing.
3. Read the circled numbers together and write one sentence on what happened.
4. Pick one or two things to work on this week, per store. Not five.
5. Next Monday, check whether those numbers moved.

For a single store, this takes 15 minutes. For several stores, our guide on how to manage multiple locations shows how to keep it to an hour.

Franlever's AI COO runs this comparison every morning. It reads each store's numbers against its own recent weeks and the middle store in its brand, and sends the owner, and the GM where there is one, at most one lever per store. A store running normally gets no lever that day.

General information for restaurant and franchise operators, not legal, tax, accounting or financial advice. Figures in examples are illustrations, not predictions or promises of results. Laws and platform policies change and differ by location, so check the current rules and talk to a qualified advisor about your situation.