Catering & event costing: +6.1 EBITDA points after closing the capital leak with the Restaurant Model Canvas and the Standard Recipe Generator

The starting point: record sales, zero cash

Catering wasn't failing on low pricing but on nonexistent per-event costing: that was Diego F. Parra's diagnosis when he opened the books of this casual 32-table bistro, 11 years in business. The operation ran 18 staff in high season, a $41 dining-room ticket and a $148 per-guest catering ticket. The events arm brought in 38% of total sales and was growing, yet at month-end nothing was left. The owner kept repeating the line Diego hears again and again across +8,400 restaurants in 43 countries: 'we bill more than ever, but there's no cash left'. Median pretax profit in full-service was 2.8% of sales in 2024 (National Restaurant Association, 2025); this operation wasn't far from average, and that is precisely the problem: the industry average is a warning, not a target to aim for. Catering and events costing demands a per-event P&L because each service is a mini-operation with its own purchasing, logistics, staffing and waste, not an extended dining-room service.

Why catering costing is a different animal?

When this bistro quoted 'by eye', copying its menu food cost, margin evaporated in production: the gap between theoretical and actual food cost hit 9.4 points.

It's not rare. Foodservice food surplus was worth $157 billion in 2024, equal to 14% of sector sales (ReFED, 2024), and at an event that waste spikes without dedicated costing. Diego split every 40-to-220-cover event as a business unit: its own purchasing, its own Prime Cost, its own break-even. The $148 ticket stayed intact; what changed was measuring the real cost per event instead of hiding it inside the restaurant's general overhead where nobody could see it bleeding out. The Masterestaurant tool applied was the ecosystem's Per-Event Costing Matrix (herramientas_restaurantes.html), a P&L that isolates each service with its real Prime Cost. Diego loaded onto each event the three costs previously diluted: production food cost, direct setup labor and logistics (transport, tableware rental, extra staff).

The tool: per-event P&L and the Masterestaurant costing matrix

The result was brutal: real catering Prime Cost was 71%, not the 58% the owner believed. With raw data, the theoretical-actual gap fell from 9.4 to 1.8 points in three months. The Producer Price Index for services rose +3.2% in 2025 (BLS, 2025) and the all-food index sits 35% above the February 2020 level (USDA ERS / BLS, 2026): without per-event costing, that input inflation eats the margin without the owner seeing it until the monthly close arrives. The bistro was financing its event clients without knowing it: it collected at 34 days and paid suppliers in advance, leaving the operation structurally in the red before each event. That cash mismatch is lethal in a thin-margin business. In full-service, median profit was 2.8% (National Restaurant Association, 2025), a cushion that can't absorb financing 34 days of working capital. Diego imposed the ecosystem rule: a 50% deposit at contract signing and the balance 7 days out from the event.

The other hole: the cash cycle was financing the client

The cash cycle aligned and the operation came out of the pre-event red. Regional variation in SBA loan default rates for restaurants reaches 8.7 percentage points (Crestmont Capital, 2026); a healthy cash cycle is what separates profitable catering from one that bills hard and ends up borrowing to make payroll. Price was never the problem: with the same $148 per-guest ticket, catering EBITDA went from 2.9% to 9.0% in a single quarter. No figure in this case comes from raising prices; every gain came from costing correctly. The theoretical-actual food cost gap dropped from 9.4 to 1.8 points (per the case), real Prime Cost came under control, and the 50% deposit pulled the operation out of the cash red. For scale: Mexico's restaurant industry is worth 300 billion pesos (CANIRAC, 2024) and Canadian foodservice sales reached CAD 96.5 billion in 2024, +4.0% versus 2023 (Statistics Canadá, 2024); in markets growing like that, catering is the most wasted margin lever there is.

The measurable result: same price, EBITDA triples

This bistro didn't sell a pricier dish: it stopped giving away the margin on its own events. The transferable lesson is that every event needs its own P&L, and the first step depends on the size of the operation. Small independent (1 site, occasional catering): this week, build a simple sheet with production food cost + direct labor + logistics from your last event and compare it to what you charged; real Prime Cost almost always tops 65%. Mid-size (bistro with a catering arm, like this case): apply the 50% deposit on your next contract and measure the theoretical-actual gap in your event food cost —if it exceeds 3 points, you have a leak. Multi-site group: standardize a single Per-Event Costing Matrix across all sites and audit per-event Prime Cost monthly; with sector sales projected near US$1.55 trillion in 2026 (National Restaurant Association, 2026), margin is won through costing discipline, not volume.

Limits of this case: where NOT to expect the same result

This result does not replicate in every context, and saying so avoids survivorship bias. First: in catering where the selling price already sits below real Prime Cost —price wars on low-ticket corporate events— better costing only confirms you must raise price or abandon the channel; here the $148 ticket had margin to rescue. Second: in operations without event volume (fewer than 8-10 a month), the cost of building a per-event costing system doesn't pay off, and a simpler batch calculation is smarter. Third: in markets with runaway input inflation —the food PPI sits 35% above February 2020 (USDA ERS / BLS, 2026)— not even the best costing compensates if contracts are fixed-price at 6 months; there the lever is an adjustment clause, not the P&L. Costing cleans up what has margin; it doesn't invent margin that isn't there. BEFORE costed the event as an extended dining-room service; AFTER treats it as a business unit with its own P&L, its own Prime Cost, and its own break-even point.

What changed at the root between before and after?

BEFORE hid production and logistics waste inside the restaurant's general overhead; AFTER charges them as direct cost of the event that caused them, which is why the theoretical-actual gap dropped from 9.4 to 1.8 points.

BEFORE financed the client unknowingly (34-day collection, supplier paid upfront); AFTER requires a 50% deposit and aligns the cash cycle, pulling the operation out of the structural pre-event red. BEFORE believed the problem was price; AFTER proved the price was right and the problem was costing: same $148 ticket, EBITDA from 2.9% to 9.0%.