How to open a dark kitchen from scratch: checklist vs Masterestaurant method

What exactly is a dark kitchen, and how does it differ from a regular restaurant?

A dark kitchen produces exclusively for delivery and aggregators —Rappi, Uber Eats, DiDi, iFood— with no dining room, no floor staff, and not a single square meter spent on decor.

That structural savings is the promise that draws so many owners in, and it's also why so many pick up the wrong checklist: they copy the operational list from a traditional restaurant (hygiene, staffing, inventory) and call it complete. Half of it is missing. Diego F. Parra, of Masterestaurant, sorts out financial viability first —price, product mix, aggregator commission— and only then builds the operational checklist on top of that foundation, because building the perfect kitchen on top of a margin that doesn't exist is the fastest route to closing. 79% of dark kitchens in Latin America fail within the first 18 months, and 64% of that failure has nothing to do with food quality: it comes from never running this math before signing the lease on the industrial unit.

The top 5 mistakes almost everyone makes (and what each one costs in real money)

First, ignoring the aggregator's real commission: DoorDash charges plans of 15%, 25%, or 30% depending on the tier (CloudKitchens, 2024), and an owner who calculates food cost against sale price without subtracting that commission is lying to himself about a margin he doesn't have —the cost of getting this wrong is running out of cash by month four. Second, not validating product mix before buying equipment: if your average ticket doesn't cover commission plus food cost plus packaging, every order delivered loses money instead of making it. Third, underestimating packaging as a fixed cost, when it can eat 4-6 points of margin nobody budgeted for. Fourth, hiring two full-time cooks from day one without validating volume —fixed payroll against variable delivery income is the combination that bankrupts a new dark kitchen fastest. Fifth, not negotiating terms with the aggregator using real data from your zone: paying list-price commission instead of asking for the rate that matches the volume you'll actually move gives away 3 to 5 points of operating margin that never come back.

Why does the traditional checklist let doomed businesses through on the very first order?

Because it checks that the kitchen is clean, that inventory adds up, and that staff paperwork is in order, but it never asks whether the pricing model can sustain the operation.

With a 30% aggregator commission, 28% food cost, and a resulting 8% gross operating margin, the kitchen can be spotless and the business will still fail within six months —the traditional checklist would approve every box while the cash runs out. Masterestaurant reverses the order: it first asks what price and product mix sustain the model given the real commission of the channel you're using, then builds the operational checklist on top of that validated number. This sequence speeds up the profitability cycle 3.2 times —8 to 12 weeks against 24 to 36 weeks with the traditional method— because it doesn't spend capital fixing a pricing error that a spreadsheet could have caught before a single piece of equipment was purchased.

How do you actually run this checklist day to day, without it turning into a forgotten PDF?

The owner —not a delegated manager, because in a dark kitchen micro-operation the owner IS the operation— reviews five financial items every Monday before opening:

effective commission paid the previous week, real food cost per dish sold, average ticket, operating margin, and order volume by channel. That review takes 20 minutes if the aggregator app's data is on hand, and it catches a pricing or mix deviation before it turns into three months of accumulated loss. The operational checklist —hygiene, inventory, packaging, prep times— gets reviewed daily, at shift opening, because an error there costs an hour, not a month. With an assistant and the owner covering both fronts, the full checklist cycle —weekly financial, daily operational— doesn't require a new administrative hire: it requires 20 minutes of discipline that almost nobody schedules because it feels less urgent than the kitchen line.

How do you audit real compliance with the checklist, using evidence instead of memory?

Every financial item needs a written number, not a gut feeling: effective commission gets audited against the aggregator's weekly statement, not against the percentage quoted at sign-up —many aggregators raise the rate with volume or change plans without clear notice.

Real food cost gets audited against the month's purchase invoices divided by dishes sold, not against the theoretical recipe card, which is almost always outdated. Operating margin is calculated by subtracting commission, food cost, and packaging from the average ticket, then compared week over week in a simple spreadsheet, not in the owner's head. If any of those three numbers drifts more than two percentage points from budget, the checklist flags a red alert and forces a price or mix adjustment before the next purchase cycle —the audit isn't an end-of-month exercise, it's the only way to catch the leak before it becomes structural.

What staffing mistake costs the most when building a dark kitchen from scratch?

Assuming, as the traditional checklist does, that you need at least two professional cooks from day one of operation. Masterestaurant starts with a real micro-operation:

the owner plus one assistant, scaling staff only once already-validated margin can sustain it, not when order volume just "looks good" on the aggregator dashboard. This difference isn't cosmetic —fixed payroll running against variable delivery income is the quiet cause behind much of the 64% failure rate wrongly blamed on generic bad unit economics. Hiring early feels prudent because it gives operating headroom to absorb order spikes, but in practice it freezes working capital exactly when the business most needs flexibility to correct price or mix without breaking cash flow. The right criterion: every new hire should be backed by at least four consecutive weeks of positive operating margin, not by an optimistic growth projection. Treating them as "just another channel" is exactly the mistake baked into the traditional checklist, and the one that erodes a new dark kitchen's margin fastest.

Are aggregators just another sales channel, or do they need different treatment when building the model?

Masterestaurant treats them as business architecture: negotiated from scratch, with real local market data, before the menu and pricing are set —not afterward, once equipment is bought and a printed menu that's expensive to correct is already in circulation.

Food delivery penetration in Colombia rose from 19.8% in 2024 to a projected 29.2% by 2026 (Statista), confirming the channel keeps growing, but also that competition for space on each aggregator will squeeze commissions and visibility. Negotiating the right plan —15%, 25%, or 30% depending on volume and service tier, following the structure documented by CloudKitchens— from the first contract, instead of after six months of operating blind, is the difference between building on solid architecture or on a channel that can shift terms at any renegotiation. The traditional checklist celebrates order volume and five-star reviews as signs of success, and that's the second most expensive mistake in the sector: both metrics can climb while operating margin quietly sinks.

How do you know the model actually works, beyond good reviews?

Masterestaurant tracks both in parallel —if volume rises AND operating margin rises, the model holds; if only volume rises, you're buying customers with your own profitability.

58% of customers prefer ordering directly through a restaurant's own app or website instead of an aggregator (NCR Voyix, 2024), a fact almost no new dark kitchen exploits because it starts out thinking exclusively about the external channel. Opening a direct ordering channel, even a modest one, cuts dependence on aggregator commission and gives the owner a second source of margin that pays off precisely when platform competition squeezes menu prices. Traditional method burns capital on validation; Masterestaurant validates BEFORE spending, then spends only on execution. Traditional checklist is sequential (equipment first, then operations, then pricing); Masterestaurant is parallel: pricing and operations day one. Traditional measures success by volume/reviews; Masterestaurant by unit economics + reviews in parallel — if both rise, the model holds.

Key differences

Aggregators in traditional method are just another channel; in Masterestaurant they're core architecture, negotiated from zero with real local-market data. Staffing: traditional checklist assumes 2+ professional cooks from day one; Masterestaurant starts micro (owner + 1 assistant) and scales only as real margin grows.