What Food Cost Actually Means (And What Killed Flaco's Cafe)

A friend of mine ran a cafe in Florida called Flaco’s. His name was Felix.

His parents came straight from Cuba and cooked the menu themselves — paella on weekends, traditional Cuban plates Monday through Friday. The kind of food you eat at a Sunday family party at someone’s tía’s house. The room was warm. Regulars walked in and got hugged. Service was the kind you can’t fake, the kind a corporate hospitality consultant could run a six-day workshop on and still not capture.

Flaco’s closed.

Not because the food was bad. The food was great. Not because the service was bad. The service was excellent. Not because the rent was crushing or the location was wrong or the menu was confused. Flaco’s closed because the family ran out of money — and they ran out of money because nobody was running the math on the generosity.

Free meals to staff every shift. Comps for regulars. Plates out the kitchen for friends, family, the AC tech, the priest, anyone who walked in hungry. The food itself didn’t cost a lot to make — proteins were sourced cheap, simple cuts done well, sides built around rice and beans. On paper, Flaco’s should have worked. In reality, the math nobody was running ate the business one comped plate at a time.

This is the story I want every operator to read before they open their doors. Because Flaco’s isn’t an outlier. Flaco’s is the modal indie restaurant closure: love-driven, generosity-fueled, math-blind.

Permission note: Felix has consented to me telling his story here. The lessons are universal. The names are real because his family’s craft deserves to be remembered, not erased.

Food cost percentage — what it actually measures #

Most operators learn food cost percentage as a math definition: cost of food sold divided by food revenue, times 100. That’s correct. That’s also the part that doesn’t matter.

What food cost percentage actually measures is your discipline density — how many times per week you’re forced to look at what’s going out the back door versus what’s coming in the front. An operator running food cost weekly catches drift in week three and fixes it in week four. An operator running it monthly catches drift in month two and fixes it in month three. An operator running it quarterly catches drift after the lease renewal and after the loan call.

Flaco’s wasn’t running food cost at all. Not weekly. Not monthly. The closest they had was the bank balance at the end of the month — and the bank balance is a lagging indicator of a problem that started six months earlier. (I have, regrettably, met operators who think the bank balance is the food cost calculation. They are also operators with closures in their futures, and I tried to tell them.)

The four buckets of “free food” that close restaurants #

In an indie restaurant where the owner is also the cook, the math gets fuzzy because the food and the love and the family and the regulars all blur into one operating reality. But the dollars don’t care about that. The dollars are tracked in four buckets.

1. Staff meals #

Every restaurant feeds its staff. Some at cost. Some free. Some with a per-person daily budget. The point isn’t that staff meals are bad. The point is that staff meals are a real cost line. A four-person kitchen eating $4 in food per shift is $480 a month. Doubled across the FOH on a busy operation is closer to a thousand. None of it is malicious. All of it has to come out of margin somewhere.

National Restaurant Association guidance treats employee meals as a normal operating cost that should be tracked separately from cost of goods sold (NRA Uniform System of Accounts for Restaurants, 10th ed.). Most modern POS systems support a “staff meal” comp button that captures the food cost without rolling it into COGS. Use it. The button does the discipline for you.

2. Comps for regulars #

The buyback culture I wrote about for bar pour cost applies in restaurants too. The owner sees the regular, sends a complimentary appetizer, comps a dessert. On a slow Tuesday it’s hospitality. On a busy Saturday across every regular at every table, it’s a real cost.

Same fix: comp it in the POS every time. The point isn’t to stop comping. The point is to count the comps so you can decide whether the loyalty return justifies the spend. Flaco’s never counted. Eventually nobody knew how much they were giving away because nobody was tracking. Then the bank balance spoke for them.

3. Family and friends #

The hardest one. The cousin who shows up with three kids. The priest who blesses the place at Christmas. The vendor’s son who graduated and brought his college roommates. In a family-run operation, these are the people the business exists to feed. Saying no to them isn’t an option, and an operator who tries to say no is going to spend Christmas Eve in the cold.

But not counting them is also not an option. The fix is the same: a family-and-friends button in the POS, rung every time, with the food cost captured. The relationships stay intact. The numbers get tracked. The two are not in conflict, no matter how much it feels like they might be.

4. The kitchen ate it themselves #

The loaf that didn’t sell. The fish that came in slightly off and got cooked for staff. The mistake plate the cook ate at 2pm because it was already plated when the order got recalled. Food waste consumed by staff is the most invisible cost line in the entire restaurant. It’s already in inventory, already in COGS, but it’s not generating revenue. It just disappears.

The Toast operations playbooks recommend a daily waste log for exactly this reason (Toast: Restaurant Waste Management Guide). MarginEdge does the same in its inventory module. The point isn’t that staff shouldn’t eat the mistake plates — they should, that food was paid for. The point is that the cost should be visible so you can manage it.

The “love doesn’t have a price tag” problem #

Felix’s parents are not unique. Every family-run operation faces the same conflict between hospitality and discipline. The food is love. The regulars are family. The staff is family. Charging for love feels obscene.

But the rent isn’t love. The vendors aren’t love. The payroll taxes aren’t love.

The fix isn’t to stop being generous. The fix is to decide the budget for generosity in advance. A 5% comp budget on food sales is roughly $400 a week on a $80,000-a-month operation. That’s enough to comp two regulars per shift, feed the staff, and slip a free dessert to the table that bought the bottle of wine. It’s a real budget. It’s funded. It doesn’t hide.

Flaco’s didn’t have a budget. Flaco’s had a culture of generosity without a corresponding culture of accounting. That works for six months. It does not work for sixty. (Sixty months, for the record, is approximately the runway most lenders give you to prove a concept. Six months is approximately the runway most operators give themselves emotionally before they start ignoring the math.)

The operational discipline that prevents this #

Here is the entire playbook. There are four steps. None of them are technical. All of them are habits.

1. Count inventory weekly. Every Sunday night, every Monday morning. Walk the walk-in, count the dry storage, count the freezer. Same valuation method every week. The count itself is what matters; the precision is secondary. A 30-minute Sunday count beats a perfect monthly count every time.

2. Pull the COGS calculation weekly. Opening inventory plus purchases minus closing inventory equals cost of goods sold. Divide by food sales. That’s your weekly food cost percentage. Track it in a spreadsheet. Watch the trend, not the absolute number.

3. Set a comp budget and ring every comp. Pick a percentage. 3%, 5%, whatever fits your concept. Ring every staff meal, every regular comp, every family-and-friends plate, every mistake meal. The button matters. The discipline matters more than the budget.

4. Have the uncomfortable conversation when the number drifts. A two-point food cost slip on $80,000 in monthly food sales is $19,200 a year. That’s the difference between an operator who renews a lease and one who closes. When the number drifts, sit with it. Walk through the four buckets above. Find the leak. Fix the leak. Don’t let it sit until next quarter, because next quarter is also when your accountant calls about your taxes.

What I think about when I think about Flaco’s #

There is a version of Flaco’s that is still open. Same family. Same recipes. Same warmth. Same hugs at the door. The difference is that someone in the family was running food cost weekly. The first time staff meals ran a hundred bucks more than expected, somebody noticed. The first time a regular got six free desserts in a month, somebody asked the question. The math didn’t kill the love. The math protected the love.

That’s the entire reason this site exists. Generosity is the soul of indie restaurants. Generosity without accounting is the leading cause of indie restaurant closure. Both things are true. The calculator on this site is one half of the answer. The discipline of using it weekly is the other.

Felix and his family deserved a version of this story where Flaco’s stayed open. I’m telling the version we got because every operator reading this still has time to choose differently. The math is on your side if you let it be.

Try the food cost calculator #

Run your numbers. The food cost calculator on this site does Quick mode (purchases ÷ sales) and Detailed mode (full COGS with adjustments for transfers, employee meals, and comps). Both show the math. Use it weekly.

Sources cited #

• National Restaurant Association — Uniform System of Accounts for Restaurants, 10th ed.
• Toast: Restaurant Food Cost Calculator + Operations Playbook
• Restaurant365: Calculating Restaurant COGS
• MarginEdge — Plate & Menu Costing 101
• Industry comp budget benchmarks cross-referenced with Toast and Lightspeed publications