The owner called me last March.
He wasn’t ready to talk about numbers yet. He just needed to say the thing out loud. “We’re bleeding and I don’t know where. I need you to look.”
Four units. One concept across all of them. Casual dining, mid-check average, same menu, roughly the same volume. On paper the group should have been printing. In the P&L he was looking at Monday morning, food cost was running 32.8%, two full points over where it should have been for a concept at his price point. Nobody in the room could tell him why. The GMs each had a theory. None of the theories agreed. And when he pulled the four P&Ls side by side, no single unit was dramatically worse than the others. All four were quietly running hot.
That’s the shape of the leak that gets an operator into trouble before a holiday season. Nothing dramatic. No single bad month. Just a cost line that’s been drifting for long enough that nobody remembers what good looks like.
What we found
When you lay four units of invoices on a conference table and actually read them, a few things usually show up.
The first is vendors nobody has questioned in two or more years. Not because the vendor is bad. Because the standing order has been running on autopilot and prices have moved on the vendor’s side at a pace that quietly outran everybody’s attention. In this case, the primary protein vendor hadn’t been re-quoted since the group opened the fourth unit. That was eighteen months ago, through the two biggest cost-move years the market has seen in a decade.
The second is invoices that get approved without being read. Signatures, scanned, filed, paid. Nobody comparing this week’s unit price to last week’s. Nobody catching the two percent that moved on four items at once. Nobody noticing when the pack size changed and the unit price went up with it. The system was built to pay vendors, not to catch drift, and it was doing exactly what it was built to do.
The third is that nobody had a view across the four units at the same time. Each GM saw his own numbers. The owner saw the consolidated P&L. Nobody saw the four unit P&Ls together, next to each other, in the same room. Which meant a vendor who was handling one unit well and three units badly looked, from any single angle, acceptable.
What changed
Not everything, and not fast. The GMs were already working hard, and the point of any of this isn’t to add work. The point is to put the work they were already doing in front of the right numbers.
Three moves. Each one small on its own.
Primary protein and primary produce got re-quoted across all four units. Not as a replacement decision. As a reset. The incumbent kept both categories, with new terms the owner could point at.
A standing Tuesday-morning review went on the calendar. One hour, same hour every week, the owner and one of the GMs looking at the top twenty items by spend across all four units together. Not a meeting. A read, with a short list of questions.
And the invoices started getting read before they got paid. Not every line of every invoice. The top twenty items, cross-referenced to the week before, flagged if anything moved more than two percent.
Three habits. None of them novel. All of them specific to what the four units actually needed.
Where the numbers went
By May, food cost was 31.9%. By July, 30.6%. In the P&L he looked at this past Monday, pulling September’s close, 30.4%.
Two and four-tenths points off food cost on a group doing roughly a hundred and sixty thousand a month in combined food spend. Call it thirty-eight thousand dollars a month back on the P&L. Over the four months to Christmas, that number is more than the group’s second-worst unit will clear in profit this year.
The owner’s calling it a reset, not a win. There’s a difference, and I think he’s right about it. The number is back where it should be. The habits that put it there are still six months old. The holidays are what will prove whether they survive volume.
What this means for anybody reading in October
Holiday season is where the gap between the group that put the discipline in and the group that didn’t will be most visible. More volume means more invoices, more deliveries, more decisions, less time to look at any single one. The groups heading in with a weekly review hour on the calendar and a top-twenty price check in place won’t be surprised by January. The groups heading in without those systems will spend February trying to figure out where the money went between Thanksgiving and New Year’s.
Which group yours ends up in is mostly already decided by what you install in the next three weeks.
One warning the experienced operators reading this already know, worth saying out loud for the ones who are newer to a holiday run at scale. November and December food cost numbers can lie to you. Catered events bring volume, and planned event menus come with better margins built in, and together they can make a food cost line look stable or even improved while the a la carte business underneath is quietly bleeding. The real test comes in January, after the catering tide goes out and the bills from a six-week sprint come due. Operators who read a healthy December number as proof the system is working, when the system is actually being carried by banquet business, get a nasty surprise around Groundhog Day. I’ll come back to this in two weeks. For now, the point is that whatever your December close says, the ongoing habits either earned it or they rode on catering, and the two paths read very differently in February.
Why I’m telling this one now
October is the only month that’s still long enough to do real work before the holidays. By early November the volume starts to turn and the window to install new habits closes. The group above wasn’t doing anything you can’t do. They were looking at the right numbers on a schedule, with somebody responsible, before the truck backed in.
Four weeks ahead, this newsletter is going to run four of these stories back to back. October is when case studies actually matter, because this is the month to see what’s possible before the storm hits. If you’ve been reading since Labor Day, now is when the ideas you’ve been nodding at get expensive to leave on the page.
That’s the start of Holiday Run.
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.