Most places have some sort of food cost for combo situation, but they just don't want to deal with it.
Every menu has them. “Choice of fries, coleslaw, or garden salad.” “Served with your choice of Caesar or mixed greens.” “Pick two sides.”
Guests love options. Operators love the flexibility. But when it comes to food costing, “choice of” items are where most spreadsheets — and most managers — start guessing.
The problem is straightforward: if a guest can choose between three sides that cost you $0.53, $0.89, and $1.44, respectively, what number do you put on the plate cost? The wrong answer costs you money every single service. The right answer depends on how accurate you need your food cost software to be and how much data you’re willing to use.
There are three ways to approach this. All of them are defensible. One of them is significantly more accurate than the others.
Option 1: Use the Highest Cost Item
This is the conservative approach. You build your plate cost using the most expensive option the guest could choose. If your three sandwich sides are fries ($0.72), coleslaw ($0.48), and a garden salad ($0.91), you cost the plate using the garden salad. That way, no matter what the customer picks, your actual food cost will never exceed what you’ve planned for.
For higher-ticket items — a $42 steak with a choice of premium sides, for instance — this makes sense. The cost variance between sides is a smaller percentage of the total plate, and the margin of safety is worth the slight overestimate. It’s the path of least resistance, and for a lot of kitchens, it’s good enough.
But “good enough” and “accurate” are two different things.
Option 2: Average the Costs
Instead of using the highest cost, you take all the options, add up their individual costs, and divide by the number of choices. This gives you a blended average that’s closer to reality for most plates served.
Think about a breakfast menu where guests choose their protein: bacon ($1.10), sausage ($0.85), or ham ($0.78). The average cost across those three is $0.91 per serving. That’s probably closer to your actual cost per plate than costing every breakfast at $1.10 — especially if bacon isn’t the runaway favorite.
This works well for a prix fixe or table d’hôte menu, too. Say you’ve got a three-course dinner where guests select one item from each course: three appetizer options, three mains, three desserts. You build a “Choice” recipe for each course using the average cost of the options in that course, then assemble the full plate from those three averaged components. It gives you a working plate cost that reflects the full range of what guests actually order.
Same logic applies to a simple two-way choice — pancakes or French toast at brunch. Average those two recipe costs into a single “Breakfast Starch CHOICE” component, drop it on the plate, and you’ve got a number you can work with.
Better than guessing. Still not the whole picture.
Option 3: Weight the Costs Using Actual Sales Data
This is where it gets real.
Your POS system tracks what people actually order. Those side modifiers, those protein add-ons, those course selections — they all get tallied in your reports. Use those numbers.
Instead of assuming equal popularity across all options, you plug in the actual quantities sold over a recent period — a month is usually enough. If you sold 400 orders of fries, 150 coleslaws, and 250 garden salads last month, your weighted cost per side reflects what your guests are actually choosing, not what a simple average assumes.
This matters more than most operators realize. In a QSR combo that includes a burger, a side, and a fountain drink, the food cost on that combo changes meaningfully depending on whether 60% of your guests pick fries or onion rings. A basic average treats those as equal. Your sales data tells you they’re not even close.
It matters in BBQ restaurant operation where guests build their own plates — pick two meats and two sides from a list. The brisket costs you significantly more per portion than pulled pork. If brisket outsells pulled pork three to one, your average meat cost isn’t the midpoint between them. It’s skewed heavily toward the expensive cut, and your plate cost needs to reflect that, or you’re undercosting every plate that goes out the window.
The weighted-average method captures this. And once you’ve set it up, updating it is just a matter of pulling fresh sales data periodically — monthly works for most operations, though seasonal shifts (summer salads vs. winter soups) are worth watching.
Why This Was Built Into the FCCC
Here’s something that matters: the FoodCostChef Calculator handles all three of these approaches natively. You build individual base recipes for each option, wrap them into a Choice Recipe, assign your units — whether equal weights for a simple average or actual sales volumes for a weighted calculation — and the cost flows straight through to your plates.
Change a side option? Update the Choice Recipe, and every plate that uses it recalculates instantly. Romaine lettuce spikes 30% overnight? Update the ingredient once, and it cascades through your salad recipe, into your Salad Choice, through every plate that offers that choice, automatically.
That cascade is the part most spreadsheets can’t do without breaking. It’s also the part that matters most when prices are moving and your menu has dozens of interconnected components.
This feature exists in the FCCC because it was designed by someone who actually had to cost these menus. Not a developer guessing at what kitchens need — a chef who spent decades staring at the same problems you’re dealing with right now. The Choice Recipe feature exists because “choice of” is how real menus work, and the costing tool should match.
Pick What Works for You
Highest cost protects your margins. Averaging gets you closer to reality. Weighted averages using POS data get you as close to accurate as you can get without counting every plate by hand.
The right approach depends on your operation, your menu complexity, and how much variance exists between your options. A two-item choice with similar costs? Average it and move on. A six-item choice with wide cost swings on a high-volume menu? Pull your sales data and weight it properly. The difference in accuracy can be the difference between hitting your food cost targets and wondering why you’re always a point or two off.
The video above walks through this step by step using real examples in the FCCC. Worth a watch if you want to see the math in action.
It’s only good food if the math works.
