Volume hides leakage. November is where you find out which.
There was a phone on the wall of a kitchen I was working in close to thirty years ago. The vendors called it to take our orders. If the chef or the sous was expecting an order to go in on a Monday for Tuesday delivery, the vendor would call us. We’d grab the phone with one hand, keep working the line with the other, and read off the list.
That was the system. It worked fine on a Tuesday night.
It did not work on the Friday of a holiday weekend, when the house was packed and the kitchen was two cooks short because somebody didn’t show up and nobody had time to replace them. On nights like that, when the phone finally rang, whichever one of us was closest would wipe his hands on a rag, grab the receiver, and thank God the vendor had called, because we certainly weren’t going to find a window to call out. “Same as last week, plus two more cases of chicken, Tuesday morning, bye.” Thirty seconds, back on the line.
If the vendor didn’t call, we didn’t call out. We were too busy. The order got placed when it got placed, which was usually too late, which was usually how we ended up short of something the next week, which was usually how we ended up in a worse version of the same scene a week later.
That is what peak looks like in a restaurant kitchen. It has always looked like that. You are always understaffed. You are never fully prepared. The volume comes in peaks and valleys that make staffing financially impossible to size correctly, so you size for the valleys and absorb the peaks. The tools have changed since the nineties. The phones are mobile and the orders go in through portals. But the shape of the problem is identical. When the volume peaks, the discipline you installed six weeks ago bends. Sometimes it holds and sometimes it doesn’t, and whichever way it goes, you find out in January.
What peak actually does to the system
The receiver who’s been running careful counts since September has three tickets in his hand on a Friday night in late November. The vendor rep is at the back door. The sous is covering the hot line because somebody called out. The chef is on expo. The owner is doing the floor because it’s that kind of night.
Something gets skipped. The count that was scheduled for 4 PM doesn’t happen. The notation on the invoice gets rushed and loses the detail the credit department will need next month. The weekly review hour that was supposed to run Tuesday morning gets pushed to “later this week” and then gets forgotten. None of it looks like a problem on the night. All of it compounds into a January close that nobody will be able to fully explain.
Three things make peak harder than the rest of the year, and they are structural, not personal.
First, every attention is on the loudest problem. In a slammed kitchen there is always a loudest problem. The quiet, scheduled work loses to it every time. The weekly review and the top-twenty count are not loud problems. That is precisely the kind of work that doesn’t survive peak.
Second, volume absorbs waste. A twenty-percent shrink rate on a slow Tuesday is visible. The same rate on a two-hundred-cover Friday feels like a busy night. The number on the invoice doesn’t scream the way it would in July. Everybody is tired. Everybody moves on.
Third, your vendor knows you’re busy. Mistake invoices, price creep, and short deliveries show up during peak because the pushback is weakest. The rep who would get a hard call on August 12 gets a shrug on December 12. After hour fifteen of a shift, the hard call doesn’t come.
What peak-volume leakage actually costs
A point of food cost loss on your November-December volume costs more than a point in any other month. If your group runs $250,000 a month in food spend normally and $400,000 a month at peak, a one-point loss during peak is $4,000 a month more than the same point in the summer. Across the six highest-volume weeks of the year, that is real money.
The deeper problem is that peak-volume leakage is also the hardest leakage to see on the P&L. The composite numbers in November and December are bigger across the board. Even a bleeding operation looks acceptable on a quick glance. By the time it’s obvious, which is somewhere in the January close, the money is spent and the staffing decisions you made for Q1 are already in motion.
The order itself is a different animal
There’s another layer to this that is specific to the holidays and almost nothing else. Peak doesn’t just mean more volume. It means your ordering changes shape.
Thanksgiving week is not a bigger version of a regular week. It is a different week. The menu runs items you don’t order the rest of the year. The volumes on your standing items swing wildly, with specific proteins and produce going up by multiples and other items dropping to nothing. The weekly order you’ve been placing on autopilot for ten months doesn’t apply, because the week you’re ordering for isn’t a normal week.
Then layer in the catering and the events. Every booked event during the holiday is its own ordering problem on top of the shifted a la carte problem. A hundred-cover corporate dinner on December 11 is not inside the standing truck day. It needs its own list, its own timing, its own back-door window. In a shop with a strong event book, there are often three or four of these a week in December.
All of this means the ordering muscle you’ve been running on autopilot does not work in November and December. You can’t say “same as last week, plus a little.” Last week wasn’t a normal week either. You have to actually look, item by item, line by line, and that’s the exact work you have the least time to do when the kitchen is slammed.
How to actually hold the line
Nothing you install during peak is going to work. That is the hard truth about November. The disciplines that will hold are the ones that were already habit by Halloween. If you are reading this today, that means you have about three weeks of time where new habits still have a chance to stick before the volume takes them.
Three moves that specifically protect the system through peak, assuming the fundamentals are already in place.
Name the person responsible for protecting the receiving process during peak. Not everyone. One person. Give them explicit authority to slow a delivery down when the kitchen is slammed. The receiver who feels authorized does the work. The receiver who feels like he’s interrupting lets the driver leave.
Move the weekly review to the easiest hour of the week, not the most convenient hour on the calendar. For most groups that means 7 AM Tuesday, before service prep begins, not 2 PM Wednesday when it is supposed to be slow and never actually is in November. The review needs to survive peak. That means it goes early.
Shorten the count if you have to, but do not skip it. In a normal week it is twenty items. In peak, if the count is going to live at all, it may be ten. Ten honest items beat twenty items that never happened.
For operators reading this today
Thanksgiving is three and a half weeks out. The real volume ramp starts the week before that. The disciplines you have in place today are approximately the disciplines you’re going to have through December.
If you aren’t sure which of your habits is going to hold under peak, ask your chef and your receiver. They will tell you. The ones they shrug at are the ones that will get eaten first.
That is Hold the Line, week one. Next week we go deeper on the receiving door specifically, because the receiving door is where most of this is going to happen or not happen.
Related issues
Behind the Line goes out every Tuesday for multi-unit operators who would rather hold the line than explain where the money went.
Want to know what your own numbers look like? ChefMod runs procurement for multi-unit independents: the agreements, the price checks, the receiving discipline, and the weekly numbers underneath all of it.