The catering numbers a restaurant owner should track

Revenue is a scoreboard. It moves for reasons you cannot see and it always arrives too late to act on. If catering revenue drops in March, you find out in April and you have no idea whether fewer people asked, fewer got quoted, or fewer said yes.

Those are three completely different problems with three completely different fixes. One number cannot distinguish them.

Which six numbers actually matter?

Each one isolates a different point of failure.

Inquiries received. How many people asked, from any channel. Form, phone, email, walkin. If this number is low, you have a visibility problem and nothing downstream matters yet.

Quotes sent. How many inquiries got a real proposal with a price. The gap between inquiries and quotes is usually response failure or unqualified leads, and both are fixable.

Orders booked. How many quotes became events. This is your close rate and it tells you whether your pricing, packages, and speed are competitive.

Average order value. Total catering revenue divided by orders. Watch this against your minimum. If your average sits barely above your minimum, you are winning only the smallest jobs.

Median response time. Hours from inquiry to first substantive reply. Use median, not average, because one terrible outlier will hide an otherwise healthy number.

Repeat rate. What share of this month's orders came from a buyer who ordered before. This is the number that separates a catering business from a series of catering accidents.

Two of those six are the ones almost nobody has, and they are the two most diagnostic: median response time and repeat rate.

How do I actually collect these without new software?

A spreadsheet and a discipline. That is the entire system.

One row per inquiry. Date received, channel, event date, headcount, quoted or not, date quoted, booked or not, order total, repeat buyer yes or no. Nine columns, filled in as things happen, not reconstructed at month end.

Whoever owns catering inquiries owns the sheet. If nobody owns catering inquiries, that is your real finding and it explains most of the numbers you were about to measure.

Channel matters and is easy to lose. Phone inquiries never touch your website analytics, so a restaurant tracking only form submissions will conclude the page does nothing when half the demand arrived by phone. Ask every caller how they found you and write it down.

At month end, six numbers off one sheet. Ten minutes. Compare to last month and to the same month last year, because catering is seasonal enough that a month over month comparison alone will mislead you.

What do the numbers tell me to fix?

Read them in order and stop at the first one that is broken.

Inquiries are low. Everything downstream is noise. You have a discovery problem, and the work is getting found for catering searches before anything else. Fixing your close rate when six people a month ask about catering is optimizing a rounding error.

Inquiries are fine, quotes are low. Either the leads are wrong or you are not answering them. Check median response time first. If that is slow, you found it. If it is fast, your page is attracting people you cannot serve, and the fix is stating headcount range, radius, and lead time clearly.

Quotes are fine, bookings are low. Price, packages, or trust. This is where publishing more information usually helps rather than hurts, and whether to publish catering pricing is the decision underneath it.

Average order value is stuck near the minimum. Your menu is not selling the larger format. A catering menu that presents pan sizes, service styles, and add ons gives buyers a way to spend more, which is one reason a catering menu page needs things a dinner menu does not.

Repeat rate is near zero. You are buying every order twice. That is a followup problem and it is the cheapest one on this list to fix.

How do I calculate contribution per catering hour?

This is the seventh number, the one nobody wants to run, and it is the one that changes decisions.

Take one real catering order. A sixty person lunch at $30 a head, so $1,800.

Food. The National Restaurant Association's 2025 Restaurant Operations Data Abstract, built on financial data from more than 900 operators nationwide, puts full service food and non alcohol beverage cost at a median of 32.0% of sales in 2024. On $1,800 that is $576. Use your own actual if you have it.

Labor hours. Five hours of prep, an hour and a half of packing, two hours of delivery and setup, an hour of manager attention. 9.5 hours.

Loaded hourly cost. This is the input people get wrong, and it is worth doing properly.

Here is the debunk. Every restaurant software and payroll blog tells you loaded labor cost is 1.25x to 1.4x base wage. There is no study underneath that range. The Bureau of Labor Statistics measures it directly. In the Employer Costs for Employee Compensation release for March 2026, accommodation and food services runs $19.92 in total compensation per hour worked against $16.12 in wages, which is 1.24 times base wage, not 1.4. Strip out insurance, retirement and paid leave, which is the reality for a lot of small operators, and wages plus legally required benefits plus supplemental pay gives 1.13 times. The 1.43 multiplier is the all private industry figure, where benefits are 30.1% of compensation against 19.1% in food service. 1.4x is almost certainly that economy-wide number imported into a low benefit industry, and using it overstates your catering labor cost by roughly an eighth.

Three caveats travel with the BLS number. Its wage figures exclude customer tips, because tips are not an employer cost. The industry code blends salaried managers with hourly staff, so it is not a line cook number. And BLS has stated that from the December 2026 release forward it will remove workers' compensation costs from this series, so figures after that date are not comparable to these.

At an $18 base and the 1.24 multiplier, your loaded rate is $22.32. Nine and a half hours is $212.

Packaging, disposables and mileage. Call it $90.

Contribution: $1,800 minus $576 minus $212 minus $90 is $922. Divided by 9.5 hours, that is $97 of contribution per catering hour.

Now do the same for a normal service hour. Say a weeknight dinner hour producing $900 in sales on eleven labor hours. Food at 32.0% is $288, labor at $22.32 is $246, contribution $366, or $33 per labor hour.

Ninety seven against thirty three. In that example catering is the better use of an hour by a wide margin, and the correct decision is to expand it rather than tolerate it.

Run it on your own three largest orders. Sometimes the number comes back below your service hour, and then you have your answer: raise the minimum, raise the price, or cut the delivery radius. I ran restaurant P&Ls for sixteen years and this comparison is the one that settles arguments. Owners defend catering on revenue and drop the argument entirely when they see it per hour.

Why is there no benchmark to compare my numbers against?

Because nobody has published one, and you should be suspicious of anyone who says otherwise.

No dataset of restaurant catering attach rates, catering channel margins, or average catering order values exists with a disclosed sample and method. It is not that the data is hard to find. It is that the platforms and consultancies who could produce it either have not, or sell it behind a paywall without publishing the method. Anyone handing you a national catering average is quoting a vendor estimate.

That is not a reason to stop measuring. It is a reason to measure against yourself. Your own trailing twelve months is a better instrument than any industry median, because it holds your kitchen, your market and your price point constant.

It also means you should not try to A/B test your way to answers at this volume, and you can settle that with arithmetic rather than a benchmark. The events you need per side is roughly 16 ÷ r², where r is the relative change you want to detect and the 16 is 2(1.96 + 0.84)² rounded, 1.96 for 95% confidence and 0.84 for 80% power. Flip it and the smallest change you could see is roughly the square root of 16 divided by events per side. Twelve catering orders a month, split two ways and pooled across a quarter, gives eighteen per side, so the smallest detectable difference is about 94%derived. Nothing you do to a catering page moves a close rate by that much. The hundred conversions per variant figure people quote is the same sum in disguise: a hundred per side is exactly what detects a 40% lift (derived), and almost nothing real moves 40%. You will never reach significance before the season, the menu and the price have all changed. Use before and after windows over long periods, say so plainly, and stop pretending it is a controlled test.

If your website is where most inquiries begin, watch behavior there too. Session recordings show you what happens in the first thirty seconds on a catering page, and they will often explain a low inquiry count faster than any report will. That only works if you have a real site to instrument, which is part of why a restaurant needs a real website even with a strong Google profile.

Where does this break down?

Two places, and the first is the common one.

When the numbers are too small to read. Four catering orders a month means your close rate moves from 50% to 25% because one buyer chose a different date. That is not a trend, it is a coin flip. With small counts, use a rolling three month window and resist the urge to explain every move.

When measurement replaces the decision. A sheet is only worth keeping if something changes because of it. If you have already decided catering is opportunistic and you will take what arrives, the six numbers are overhead. Measurement is for decisions.

What to do this week

Build the nine column sheet today. Blank is fine. Start logging from this moment rather than trying to reconstruct last year.

Go back through the last ninety days of catering orders and fill in what you can. Order date, headcount, total, repeat or not. Even partial history gives you a baseline.

Calculate median response time from your last ten inquiries. Most owners are surprised, and it is usually the fastest thing on this list to improve.

Then run contribution per catering hour on your three largest orders from last quarter, using the 1.24 multiplier on your own base wage. One hour of arithmetic that may change what you do next year.

Want me to run your six numbers?

Email me at eric@seod.com with your catering orders from the last ninety days. Date, headcount, and order total is enough, no names and no customer details. I will send back your average order value, your seasonality shape, and which of the six numbers is your actual constraint.

If you can also give me the timestamps of five recent inquiries and your first real replies, I will add your median response time to it. This is a P&L reading exercise more than a marketing one, and it is the kind of thing I did for sixteen years before starting SEOD.

There is more on catering and event business to read alongside it.