Food cost: traditional method vs Masterestaurant method

Which food cost method is best when delivery already carries more than 25% of your sales?

Channel-based food costing is the better choice for any independent operator whose platforms move between 25% and 45% of sales, because the traditional method divides by a price you never fully collect.

The steakhouse in Chapinero that opened this comparison had a correct recipe card at 29,4%, an accountant who validated it, and a bank account draining month after month: 41% of orders arrived through aggregators at a weighted commission of 26%, so on a 40.000-peso dish the bank received 29.600 and the real food cost of the star product climbed to 39,7%. Ten points of gap across 41% of sales equal roughly four points of operating margin nobody was watching, and that leak hides for months because the management P&L averages channels that share almost nothing. If that is your mix, measure by channel before touching a single recipe.

Best for dining-room-only restaurants: traditional food cost still wins

Stick with a single recipe-level food cost when more than 85% of your sales are served at the table and the remainder is commission-free pickup, since the classic formula's denominator matches what actually lands in the account. A twelve-table bistro billing without aggregators gains nothing by splitting its costing into three pieces: it gains by tightening waste, portioning and purchasing, which is where the numbers still hurt. The U.S. producer price index for all foods sits 35% above its February 2020 level (USDA ERS / BLS 2026), and final-demand PPI closed 2025 at +3,0% after +3,5% in 2024 (U.S. BLS). Against that input inflation, a recipe card reviewed every sixty days is worth more than any sophisticated segmentation. Complicating the model in a single-channel business only adds administrative work without returning a peso. If your average delivery ticket falls below 35.000 pesos, the move that returns the most money is loading packaging straight into the plate cost.

Best for low tickets: packaging is product cost, not office expense

Container, bag, tamper seal and cutlery run between 1.100 and 2.400 pesos per order depending on category, and on a 28.000-peso ticket that adds 3 to 6 points of effective food cost which traditional bookkeeping files under stationery or general supplies. A burger shop shipping 900 monthly orders at 1.800 pesos of packaging moves 1,62 million per month, nearly twenty million a year, through an account nobody audits when hunting for margin. The rule we apply at Masterestaurant is blunt: anything that leaves through the door attached to the plate belongs on that plate's recipe card. Change the accounting classification and you will see which delivery menu items had been working for free since day one. Charging the same price in-app and in the dining room is the most expensive decision an owner makes by default, and fixing it returns margin within the week without renegotiating a single supplier.

Channel pricing: the fastest lever once costs will not go lower

Aggregators allow different prices per channel, and a 15% to 18% markup on the platform menu recovers most of a 26% commission without slowing orders: the delivery customer already pays a shipping and service fee, and compares against competitors inside the app, never against your printed menu. With 37% of adults ordering delivery at least once a week (UpMenu 2024) and more than 40% ordering three to five times a month, demand has enough muscle to absorb that adjustment. I got this wrong for years, recommending single pricing for brand coherence: coherence does not cover payroll, and the customer never noticed the difference. Three scenarios exist where splitting food cost by channel costs you more than it returns. First: delivery below 12% of sales, because refining a channel that contributes eight pesos out of every hundred burns management hours that pay better on the floor. Second: operations with fewer than 20 active SKUs and a monthly menu rotation, where recipe cards change before the segmentation finishes calculating.

When NOT to pick the popular option?

Third, and most common: kitchens without weekly inventory control, because a channel model fed by monthly counts produces elegant, false numbers.

If your variance between theoretical and actual exceeds 4 points, the problem is not the costing method but the storeroom, and no channel breakdown fixes what walks out the back door. Fix the count first, segment afterward. Four signals expose a tool that will not serve you. One: the system asks for a single selling price per recipe and refuses per-channel pricing, which guarantees your delivery food cost comes out understated by eight or ten points. Two: the vendor promises a universal target percentage —«hold 30%»— without asking about your channel mix or rent structure, when the healthy 32% ceiling per dish is a maximum, not a goal. Three: it will not let you load packaging as a recipe component, so 3 to 6 points end up buried in administrative expense.

Red flags when comparing costing methods and software

Four: reporting arrives monthly and without variance against physical inventory, which turns the dashboard into ancient history. One more figure worth checking before signing: SBA loan default rates for restaurants vary by as much as 8,7 percentage points across regions (Crestmont Capital 2026), and cost discipline explains a good share of that spread. If you run two brands out of one kitchen, channel costing stops being advisable and becomes mandatory, because each brand carries different commissions, packaging and promotions on top of the same labor. The repeated mistake is prorating ingredient cost by sales share and calling the analysis closed, when the virtual brand usually lives on platform discounts of 20% to 30% that eat the margin before the plate leaves the pass. Picture a secondary brand contributing 30% of sales with an apparent food cost of 31%: if 70% of its orders carry a 25% promotion, real revenue per plate drops and that 31% turns into an effective 41%, meaning the brand that looked like growth is funding its own expansion with the core business's margin.

Best for ghost kitchens or a second brand in the same kitchen

Measure it separately for eight weeks before deciding whether it lives or closes. Twenty years auditing cost structures across 43 countries leave one clear pattern, and it is not ignorance. It is obsolescence: disciplined owners, with impeccable recipe cards and suppliers negotiated to the bone, measuring with a tool that stopped describing their business the day they switched on the first aggregator. That is the diagnosis Diego F. Parra repeats in every Masterestaurant engagement, and it explains why the leak shows up in cash flow first and in the management report only months later. The useful question is not which method is superior in the abstract, because none is: it is which one describes your channel mix this quarter. Take your last ninety days of sales, split them into dining room, pickup and platform, calculate the food cost of your five best sellers against the NET revenue of each channel, and in forty minutes you will know whether you have a recipe problem or a denominator problem.

Five differences that move your cash

The denominator. Traditional divides by menu price; the channel method divides by what the platform deposits. At 26% commission a $40 dish leaves $29.60 of real income, and that single change pushes food cost from 29.4% to 39.7%. Packaging. In delivery the container, bag and seal are product cost, not overhead: roughly $0.28 to $0.60 per order depending on category, which on low tickets adds 3 to 6 pts of effective food cost that the traditional method files somewhere else. Differentiated pricing. Charging the same in app and in the dining room is the most expensive default decision an owner makes. A 15-18% platform uplift does not scare orders —aggregators explicitly allow it— and returns most of the commission to margin. How you read a promotion. Traditionally a BOGO is judged against dish food cost; by channel it is judged against that channel's net margin, and plenty of platform promotions turn out to be loss-making sales dressed as growth.

Five differences that move your cash — in practice

The link to break-even. When delivery climbs from 20% to 40% of revenue, the break-even point rises even with fixed costs untouched, because each peso sold contributes less. A single number never flags that shift.