What a ghost kitchen really costs to open and run. Marketplace dependency risk, real lease numbers, the unit economics that work and the ones that bury you.
Short version: a ghost kitchen done right can hit 18-25% net margins on $80k-$140k monthly revenue per stall, but only if you control three things: marketplace mix below 60%, average ticket above $32, and a direct ordering channel doing 25%+ of volume by month 9. Without those three, you're running someone else's business on your dime.
Three models, often confused:
The economics below focus on model 1, the commissary stall. Model 2's economics are usually great (almost no incremental cost). Model 3 is just a normal restaurant without the dining room — same rules apply.
Numbers below are typical for South Florida. Adjust for your market by ±15-25%.
This stall is probably losing money once you count owner labor and the amortized build-out. Most ghost kitchen stalls live here for the first 6-9 months. Many close at month 12.
This is the target. After owner pay and amortized build-out, real net is 18-22%.
The difference between the failing stall and the optimized stall isn't ingredient cost or rent — it's marketplace mix and average ticket. Drop marketplace mix from 70% to 40% (by building a direct channel) and add $40k/mo in revenue and you've roughly 10×'d your net.
A ghost kitchen at 95% marketplace volume isn't a restaurant — it's a contractor for DoorDash and Uber Eats. When the marketplace algorithm changes (it does, every quarter), your revenue swings 25-40% with no input from you. You can't market your way out because customers don't know your brand exists; they know DoorDash. The only fix is building direct ordering early. We cover the math in the DoorDash 30% post.
A dine-in restaurant gets foot traffic for free. A ghost kitchen gets zero. You're paying $9-$15 in customer acquisition cost for every first-time customer through the marketplace. Without a 25%+ direct repeat-order rate by month 9, the CAC math never works.
Most ghost-kitchen operators offer a turnkey option: they put in the equipment, you pay $1,500-$2,500/mo more in rent. Looks like a great deal because you skip the $40k build-out. The reality: you've taken on a higher monthly fixed cost that compounds for 24-36 months. If you have $30k-$50k of capital and you're confident in the concept, bring your own equipment.
Operators try to run 3-5 virtual brands out of a single stall ("Burger Mike's", "Crispy Bird Co.", "Taco Lab"). It looks like leverage. It usually isn't. Kitchen complexity scales with SKU count, and a single 240-sqft stall can't actually produce 4 menus at a competitive prep speed. Two brands max, and only if they share 70%+ of ingredient stock.
Ghost kitchens optimize for 1-2 person delivery orders. Catering needs 4-8 hour lead time, packaging the stall doesn't stock, and pickup logistics the marketplace doesn't handle. Most ghost-kitchen concepts leave $15k-$45k/mo in catering revenue on the table because they're not set up to take it. Our catering revenue playbook shows what to fix.
We built Zay-OS for Ghost Kitchens for this exact use case: one tablet running direct orders + Otter (all marketplace channels) + 14-provider delivery dispatch + DAVO sales tax. Customer CRM is the single biggest lever for ghost kitchens — without it, you have no path off marketplace dependency. Setup is 2-3 weeks. From $499/mo single-stall (Operator) or $699/mo flat for multi-brand (Concierge, up to 5 locations). No setup fee. A $2.99 per-order fee is paid by the diner at checkout — operators keep 100% of food revenue.
~$65k/mo at typical cost structure, assuming 65% marketplace mix. Drop to ~$55k if you can get to 40% marketplace mix.
Best case 4-6 months. Median 9-14. Anything past 18 months without break-even, kill the concept.
Neither. Operators have generally migrated away from facility-owned virtual brands because the data ownership and brand control are weak. Run your own brand on their real estate.
Technically yes, with 3-4 separate marketplace tablets. Practically no — staff workflow breaks above 200 orders/mo. We compared options in the Otter vs Chowly vs KitchenHub post.
If you're committed to opening a ghost kitchen stall, run this sequence:
Sunk-cost thinking ends more ghost kitchens than market conditions. The honest kill signals:
Closing a ghost kitchen at month 12 with $40k lost is a survivable mistake. Pushing through to month 24 to "save it" is how operators lose $120k+ and burn the next opportunity.
See the ghost-kitchen build of Zay-OS — Zay-OS for Ghost Kitchens · live demo · phone 321-666-1102.
30 minutes. We look at your business and tell you exactly what we'd do. client or not.
Book Your Free Audit