Table of contents
- What is a ghost kitchen and why the delivery-first model changes how you launch
- Choose your kitchen model before you spend a dollar
- What it costs to start a ghost kitchen
- Build a menu designed for delivery, not the dining room
- The platform math every ghost kitchen operator needs before going live
- Permits, licenses, and inspections: what to file in week one
- Your 30-day launch calendar: week-by-week from signed lease to first order
- Staffing and operations for a kitchen with no front of house
- The operators who survive past 90 days track data, not just volume
- Frequently asked questions about starting a ghost kitchen
On paper, the overhead costs for a ghost kitchen look clean: no dining room, no host stand, no servers. But a lean cost structure is not the same as an easy launch. The catch is that a ghost kitchen has zero built-in foot traffic. Your launch is a digital listing going live on a delivery app, and your visibility has to be built from scratch, on purpose, starting day one. According to IBISWorld, there are already 7,606 active ghost kitchen businesses competing for those same delivery platform placements.
Getting lost in that crowd is easy. Getting found takes a plan.
This is that plan: a week-by-week, 30-day critical-path calendar that takes you from concept to first order, with the margin math, permit reality, and platform sequencing that most guides skip entirely.
Key insights
- Platform onboarding takes 5–10 business days and health inspections book 1–3 weeks out. Both must start on day one of your plan, not after your kitchen is set up, or your 30-day launch date slips
- If your food cost is 30% and the platform takes 27% commission, you're already at 57% cost before labor or packaging. Price your menu backwards from a target margin, not forwards from your dine-in prices
- Launch on one platform first. Going live on DoorDash, Uber Eats, and Grubhub simultaneously before stress-testing your kitchen drives order errors and early bad reviews that suppress your app ranking for months
- Ghost kitchens have zero built-in foot traffic. Plan for 15–20% of projected revenue in marketing spend, hold paid ads until your rating clears 4.5 stars, and treat day 31 as the start of the real work, not the finish line
What is a ghost kitchen and why the delivery-first model changes how you launch
A ghost kitchen (also called a dark kitchen or virtual kitchen) is a delivery-only food operation with no dining room and no customer-facing storefront. Orders come in through third-party delivery apps. Food goes out in a bag. That's the entire customer interaction.
Two terms are used interchangeably; let’s define them.
- The ghost kitchen is the physical space.
- A virtual brand is the food concept: the name, menu, and identity that operates inside that space.
One ghost kitchen can run multiple virtual brands simultaneously, which is one of the model's real revenue advantages. One team, one set of equipment, multiple menus and income streams.
The launch sequence is different in one key way: there is no grand opening. No ribbon cutting, no foot traffic from a busy street. Your launch is a listing going live on DoorDash or Uber Eats, and your rating from the first ten orders shapes your algorithmic ranking for months.
A bad week one is hard to recover from. That's why the 30-day plan below is built around parallel workstreams, not sequential steps, and why platform registration and health inspection scheduling must start on the day you go live.
Choose your kitchen model before you spend a dollar
There are four models to choose from, and the one you pick affects your ghost kitchen cost, your timeline, and how much control you have over the food that leaves your kitchen.
1. Shared or commissary kitchen. You rent by the hour or shift inside a licensed commercial kitchen that other operators also use. Lowest upfront cost, minimal equipment expense if the facility is already equipped. The trade-off: limited hours, shared storage, and less control over your environment. Best for testing a concept before committing to a lease.
2. Dedicated leased space. You lease your own kitchen, set it up the way you want, and operate on your own schedule. More control, more storage, better brand consistency, and higher fixed costs. Typically takes 2–4 weeks to set up and clear a health inspection. Best if you already have volume proof or a clear demand signal.
3. Sublease inside an existing restaurant during off-hours. If you already run a restaurant, your kitchen sitting idle from 9 PM to 11 AM is an underused asset. Some restaurant owners will negotiate off-hours access. Often the lowest total cost of any model, but it requires a good working relationship and a clear written agreement.
4. Purpose-built ghost kitchen campuses. Multi-tenant facilities can get you operational quickly. The risk: vendor lock-in. Before signing anything, get the payment model, internal pricing, and your expected net margins in writing. Do not rely on verbal sales claims.
The practical decision rule: if you have no volume data, start in a shared commissary. If you're an existing restaurant operator adding a delivery-only revenue channel, use your own kitchen during off-peak hours or move to a dedicated ghost kitchen space.
What it costs to start a ghost kitchen
The honest range by model:
- Shared/commissary kitchen: $500–$2,000/month in rental fees, minimal equipment cost if the facility is equipped
- Dedicated leased space: $3,000–$8,000/month in rent (market-dependent), plus $10,000–$30,000 in equipment
- Sublease arrangement: negotiated, often the lowest total cost
One-time startup costs to budget explicitly:
- Commercial equipment (ranges, fryers, refrigeration, smallwares) if not provided
- Permits and licenses: $500–$2,000 depending on your city
- Delivery-grade packaging for the first 30 days
- First 30 days of food inventory
A lean shared-kitchen launch can come in under $10,000 all-in. A dedicated space with full equipment can run $30,000–$50,000 or more. OysterLink's industry research puts the broader startup cost range at $75,000–$200,000 when you factor in labor costs, supplies, insurance, and marketing ramp-up.
The costs operators consistently underestimate:
- Digital marketing: budget 15–20% of projected monthly revenue for ads and promotions from the start
- Delivery-grade packaging: packaging that fails in transit generates bad reviews before you've built any rating history
- Order management technology: order management and POS are not optional. Otter's Online Ordering product starts at $55/month per location, so you can budget it from day one
- Cash reserves: have at least 60 days of operating costs in the bank before you go live
Realistic net margin for a well-run ghost kitchen in its first six months is 10–20% after platform fees and food costs. Claims of 35%+ exist but are outliers tied to specific volume and cost conditions, not a baseline to plan around.
The platform math every ghost kitchen operator needs before going live
Commission reality: DoorDash, Uber Eats, and Grubhub typically charge 15–30% per order depending on plan tier and negotiation. On a $15 order at 27% commission, you receive $10.95 before food cost, packaging, and labor costs.
The profit margin math, worked through:
Cost category | % of revenue |
Food cost | 30% |
Platform commission | 27% |
Subtotal | 57% |
Packaging | 2–4% |
Labor costs | 15–20% |
Marketing | 5–10% |
That math must be solved at the menu stage, not after go-live.
Sequence your platform launches
Start with one third-party delivery platform. Most operators in the US begin with DoorDash, which holds the highest order volume in most markets. Stabilize your kitchen operations, hit a consistent 4.5-star rating and 95%+ acceptance rate, then add Uber Eats as your second platform.
Going live on multiple delivery platforms simultaneously before your kitchen flow is tested creates multi-tablet chaos, order errors, and cancellations. Bad ratings in week one hurt your app ranking for months.
- How to get on DoorDash: create a merchant account at merchant.doordash.com, submit your menu, photos, and business information. Onboarding typically takes 5–10 business days. Start this in week one.
- How to get on Uber Eats: register at merchants.ubereats.com with the same materials. Expect a similar 5–10 business day review window.
Managing separate tablets for each platform is the top operational complaint from ghost kitchen operators in their first weeks. Otter consolidates orders from DoorDash, Uber Eats, Grubhub, and other third-party delivery apps into a single POS interface, eliminating the multi-tablet problem before it creates errors and bad reviews.
Permits, licenses, and inspections: what to file in week one
Despite being delivery-only, ghost kitchens face the same regulatory requirements as any commercial food operation. This catches many first-time operators off guard and is the single most common reason a 30-day launch slips.
Required in most US markets:
- Business license: file with your city or county clerk
- Food handler permit: a food manager or food handler certification (ServSafe is the most widely accepted) takes 1–3 days online and is required before you handle food commercially
- Food establishment permit: issued by the local health department after a physical inspection of your commercial kitchen
- Seller's permit: required if your state collects sales tax on prepared food
The health department inspection is your critical path item. Inspection slots often book 1–3 weeks out. Schedule your appointment on day one or two of week one. Waiting until week two to call could delay your opening by 30 days.
Shared kitchen exception: some commissary kitchen facilities already hold a facility-level health permit. Verify explicitly whether it covers your operation.
City-specific requirements: NYC, LA, Chicago, and other major markets may have additional zoning restrictions on ghost kitchens. Check with the local planning or zoning department before you sign a lease.
Run permit applications in parallel with kitchen setup and menu work. Filing everything in week one is what makes the 30-day timeline possible.
Your 30-day launch calendar: week-by-week from signed lease to first order
Week 1 (days 1–7): start the long-lead items immediately
- Sign your lease or confirm kitchen access
- File your business license application the same day
- Schedule your health department inspection immediately
- Begin your food handler permit certification online
- Register merchant accounts on DoorDash and Uber Eats now. The 5–10 business day onboarding clock starts today
- Finalize your 8–12-item menu
- Order packaging materials and first-week food inventory
Week 2 (days 8–14): set up and prepare
- Receive and set up equipment
- Complete food photography before go-live
- Finish food handler permit certification
- Follow up on platform onboarding status
- Confirm supply chain with food distributors
- Set up your POS and order management system
- Begin training any kitchen staff
Week 3 (days 15–21): inspection and systems check
- Health inspection (if scheduled in week one, it typically lands here)
- Confirm platform menus are live, photos are approved, and pricing reflects your delivery margin math
- Run a full internal stress test: fire 10–20 mock orders, time prep-to-pack cycles, identify any station bottlenecks
- Fix layout problems before a real customer experiences them
Week 4 (days 22–30): soft launch on one platform only
- Go live on one delivery platform only
- Accept real orders and monitor acceptance rate, prep time, and customer ratings daily
- Do not run paid ads yet. Stabilize your operations and rating baseline first
- On days 28–30, review first-week data: what sold, what didn't, where errors occurred
- Adjust menu and operating hours based on what you learned
- Plan second-platform activation for week five or six, not before
Staffing and operations for a kitchen with no front of house
The ghost kitchen labor model is genuinely leaner. No servers, no hosts, no bussers. Your payroll is kitchen-only, and that's one of the model's real advantages.
Minimum viable crew for a lean launch: one lead cook or operator (often the owner in the early stage) and one prep and packing person. Add staff as order volume justifies the cost. Staff to demand data, not to anticipation.
No dining room buffer. In a traditional restaurant, the front of house absorbs volume spikes while the kitchen catches up. Ghost kitchens have no such buffer. A surge in delivery orders hits the kitchen directly. Design your menu and station layout to handle 10–15 simultaneous orders without breakdown before you go live.
Use a KDS, not a ticket printer. In a delivery-only environment, a Kitchen Display System helps your team see, sequence, and close orders clearly during peak windows.
Your packing station is your quality checkpoint. It's the last step before food reaches the customer. Build a packing checklist for every menu item. Use tamper-evident packaging on every order. Include reheating instructions for items that benefit from them.
Schedule for delivery demand curves. Delivery peaks on Friday and Saturday evenings and weekend lunches. Your staffing must mirror delivery platform demand patterns.
The operators who survive past 90 days track data, not just volume
The 30-day plan gets you to the first order. What you do with data in days 31–90 determines whether your ghost kitchen becomes a viable business.
Key metrics to track from week one:
- Order acceptance rate (target 95%+)
- Average prep time
- Order error rate
- Customer rating per platform (4.5+ stars is the threshold for strong app ranking)
- Revenue by menu item
- Peak order windows by day and hour
DoorDash Merchant Portal and Uber Eats Manager both provide weekly performance data. Review them weekly, not monthly. A rating problem identified in week two can be corrected. Ignored until week six, it becomes structural.
Menu iteration cadence: after 30 days of real order data, cut the bottom 2–3 items by order volume. They're consuming food cost and adding kitchen complexity without proportional revenue. A tighter menu in month two is almost always more profitable than the launch menu. This is menu engineering in practice: let real sales data drive what stays and what goes.
Telly’s Charburgers took exactly this approach. As co-owner Nicole Kuti put it, “Since getting Otter, we’ve cut out three items that were really just costing us money to have on the menu.”
Platform expansion timing: add a second platform only when your first-platform acceptance rate is consistently above 95% and your average rating is 4.5 or higher.
Otter's analytics surfaces missed orders, prep time trends, and sales performance across all connected delivery platforms in one dashboard. You catch problems faster than piecing together insights from three separate app portals.
Hold paid ads until week five or six, after you've stabilized operations and built a rating baseline. A promoted listing with a 3.8 rating accelerates customer churn, not growth.
A ghost kitchen launched with a tight 30-day plan, sound platform margin math, and daily data habits is a disciplined operation. The operators still running at month twelve are the ones who treated day 31 as the start of the real work.
Ready to run your ghost kitchen orders from one system, from day one? Start today with Otter.
Frequently asked questions about starting a ghost kitchen
How much does it cost to start a ghost kitchen?
Ghost kitchen cost varies by model. A shared or commissary kitchen launch can start under $10,000, covering rental deposits, permits, packaging, and initial food inventory. A dedicated leased space with your own equipment typically runs $30,000–$50,000 or more depending on your market. Budget an additional 15–20% of projected monthly revenue for digital marketing. Ghost kitchens have zero built-in foot traffic and visibility must be actively built.
How long does it take to start a ghost kitchen?
A lean shared-commissary launch can be operational in 2–3 weeks if you start permits and platform registration on day one. A dedicated leased space typically takes 4–6 weeks to clear inspection and go live. The two most common delays are health department inspection scheduling (which can book 1–3 weeks out) and delivery platform onboarding (5–10 business days per app). Both must start in week one or your launch date slips.
How do I start a ghost kitchen on DoorDash?
Create a merchant account at merchant.doordash.com and submit your menu, food photos, business address, and banking information. DoorDash typically takes 5–10 business days to review and approve your listing. Commission rates range from 15–30% depending on your plan. Start this registration in week one of your launch plan, not after everything else is set up.
How do I start a ghost kitchen on Uber Eats?
Register at merchants.ubereats.com with your menu, food photos, and business details. The process mirrors DoorDash: expect a 5–10 business day review window. Most operators should start with one platform, stabilize operations and ratings, then add Uber Eats as a second channel. Going live on both simultaneously before your kitchen is stress-tested often leads to order errors and bad early reviews.
Do I need a commercial kitchen to run a ghost kitchen?
Yes. Commercial food delivery operations cannot legally run from a home kitchen in most US states. You need a licensed commercial kitchen space, whether that's a rented commissary, a dedicated leased facility, or an off-hours sublease inside an existing restaurant. The space must pass a local health department inspection before you can begin fulfilling delivery orders.
What permits and licenses do I need for a ghost kitchen?
You need four things in most US markets: a business license, a food handler or manager certification, a food establishment permit from your local health department, and a seller's permit if your state taxes prepared food. The one that controls your timeline is the food establishment permit, because it depends on a passed health inspection and those appointments can book weeks out. Zoning is the quieter risk: some cities limit where delivery-only kitchens can operate, so confirm your address is cleared before you commit to a lease.
What is the difference between a ghost kitchen and a virtual brand?
Think of the ghost kitchen as the address and the virtual brand as the sign on the door: one is the licensed physical space, the other is the named concept and menu customers actually order from. The distinction matters at launch because a single kitchen can host several virtual brands at once, each with its own delivery-app listing, so you can test a second concept without signing a second lease. The trade-off is that every added brand is another menu, another set of photos, and another rating to manage, so it pays to prove one brand before you stack on more.
How profitable is a ghost kitchen?
A well-run ghost kitchen can realistically net 10–20% after platform commissions and food costs in its first six months. Claims of 35%+ margins exist but are outliers tied to specific volume and cost conditions. The biggest margin risks are menus priced without accounting for 15–30% platform commission, underestimated marketing spend, and expanding to multiple delivery platforms before operations are stable enough to handle the volume.
Launch your ghost kitchen with Otter