The restaurant industry has a math problem. Rent for a good location in most cities consumes 10-15% of revenue before you've served a single plate. Add staffing for front-of-house, interior buildout, furniture, decor, and maintenance, and you're looking at 60-70% of your operating costs tied to the physical experience of sitting in your restaurant. The food, which is the whole point, competes for the remainder.
A cloud kitchen solves this by removing everything that isn't the food. No dining room. No servers. No ambiance budget. No lease on a prime corner lot. You rent a commissary space or a commercial kitchen, cook for delivery only, and run the entire operation from a screen.
The economics are compelling. Cloud kitchens typically spend 30-40% less on overhead than a comparable dine-in restaurant. But the lower barrier to entry has created a wave of operators who launch fast and fail faster, because the things that kill a cloud kitchen are different from the things that kill a traditional restaurant, and most of the guides out there don't account for that.
This walkthrough covers how to actually open a cloud kitchen, from choosing your model to taking your first order, with the parts most people skip included.
What a Cloud Kitchen Actually Is (And What It Isn't)
A cloud kitchen is a commercial cooking facility that produces food exclusively for delivery or pickup. No tables. No walk-ins. No reservation system. Customers discover you through delivery apps, social media, or your own online storefront, and the food arrives at their door.
What it isn't: a shortcut. The kitchen still needs health permits, food safety protocols, proper equipment, and consistent output. You're just redirecting the 30-40% of capital that would go toward the dining experience toward marketing, packaging, and technology instead.
There are a few models to choose from, and the one you pick shapes everything downstream.
Independent Cloud Kitchen
You operate your own brand, your own menu, your own kitchen space. Full control, full responsibility. This is the most common starting point for chefs and restaurant operators who want to go delivery-only without answering to anyone else.
Multi-Brand Kitchen
You run two or more distinct brands out of the same physical kitchen. Different menus, different names, different online listings, all cooking in the same space. The advantage is that you capture more orders per time slot. A customer who doesn't want burgers might want tacos, and you sell them both from the same fryer station. This model works well when brands share prep workflows and can batch ingredients.
Franchise or Virtual Brand Partnership
You license an existing brand's menu and systems, cook in your own space, and operate under their name. Lower risk, lower margin. This can be a reasonable on-ramp if you want to test the model without building a brand from scratch, but you're limited by the franchisor's menu, pricing, and promotional decisions.
Step 1: Choose Your Concept and Menu
Most cloud kitchen failures start here. The concept sounds great in a brainstorm, but it doesn't survive contact with the delivery environment.
The core constraint is this: your food has to travel. It has to look good and taste good 20-40 minutes after it leaves your kitchen, inside a container, in a delivery bag, on a motorcycle or bicycle. That eliminates a lot of dishes that are phenomenal fresh and terrible after transport.
What Works in Delivery
Fried rice holds. Curries hold. Bowls hold. Pizzas hold if you control the crust geometry. Burgers hold if you separate the wet ingredients. Pasta with oil-based sauces holds. Sandwiches hold if you wrap the hot and cold components separately.
What Doesn't
Crispy fried items lose their texture within 10 minutes. Salads wilt in enclosed containers. Soups leak. Anything that relies on a temperature contrast (warm cake with cold ice cream) collapses in transit. Delicate sauces break.
Most teams miss this part. They design a menu around what they want to cook, not around what survives a half-hour ride in a box. The result is food that was excellent in the kitchen and disappointing at the door.
Keep your initial menu tight. 8-12 items is enough to offer variety without overwhelming your prep line or your inventory. You can always expand. Starting with 25 items almost guarantees inconsistent quality and wasted ingredients.
Step 2: Secure Your Kitchen Space
You don't need a restaurant location. You need a permitted commercial kitchen.
Commissary Kitchen
Shared commercial kitchens that rent by the hour, day, or month. They come with existing health permits, commercial-grade equipment, and oftenprep stations you can reserve. Costs range from $15-40 per hour or $500-2,000 per month depending on your city and the terms.
The advantage is low upfront cost and flexibility. The disadvantage is scheduling. You're sharing equipment and storage with other operators, which means your peak prep time might conflict with someone else's.
Dedicated Cloud Kitchen Space
Purpose-built facilities designed for delivery-only operations. Brands like CloudKitchens and KitchenHub offer individual kitchen pods within a larger facility, complete with dedicated prep space, storage, and loading zones for delivery drivers.
These cost more ($2,000-6,000/month in most markets) but eliminate scheduling conflicts and give you predictable access during peak hours. For operations running 50+ orders per day, the cost premium is usually worth it.
Build Your Own
Lease a commercial space and build it out. This is the most expensive option ($20,000-80,000 for basic buildout depending on your market) and only makes sense if you're running a multi-brand operation or-scaling beyond 100 daily orders.
In practice, this usually fails when operators sign a lease before validating demand. Start in a commissary. Prove the model. Then upgrade.
Step 3: Get Legal and Permits Right
The paperwork for a cloud kitchen is roughly 70% of what a traditional restaurant requires, but the 30% you skip is the dining room, not the kitchen. You still need every food-related permit.
What You Need
Business license for your city or municipality. Food establishment permit from your local health department. Food safety certification (ServSafe or equivalent) for at least one person on your team. Fire safety inspection and certificate. Liability insurance that covers food production and delivery. If you're doing your own delivery, vehicle insurance for commercial use.
What Changes by Location
Different cities classify cloud kitchens differently. Some require a traditional restaurant license even though you have no dining room. Others have specific cloud kitchen or ghost kitchen permit categories. Some require a separate delivery permit if you're running your own drivers. Check your local regulations before signing anything.
The key takeaway is that your health department inspection will happen, and it will be thorough. They don't care that customers don't sit in your space. They care that food is being stored, prepared, and packaged safely. Have your HACCP plan ready before the inspector arrives.
Step 4: Set Up Your Technology Stack
This is where cloud kitchens either scale smoothly or become a daily nightmare. Your tech stack is the operating system of your business. It handles orders, payments, menu updates, customer communication, and analytics.
At minimum, you need:
An Ordering Platform
This is your storefront. It can be a whitelabel online store, a listing on delivery apps, or both. Most cloud kitchens start on delivery apps for the visibility and add a direct ordering channel over time.
If you simplify it: delivery apps are for discovery. Your direct channel is for profit.
The commission math matters here. Delivery apps charge 20-30% per order. Your own ordering platform charges a flat fee or a small percentage. At 50 orders per day with an average ticket of $25, that's the difference between paying $375-562 per day in commissions versus $5-15 per day on a flat-fee platform.
Kitchen Display System
Orders need to flow into your kitchen in a way that your team can actually see and manage. A kitchen display system (KDS) consolidates orders from multiple channels into a single screen. Without one, you're juggling tablets, phones, and printouts, which works at 10 orders per shift and collapses at 50.
Inventory and Menu Management
Your menu needs to update in real time. When you 86 an item, it should disappear from every ordering channel simultaneously. When you add a special, it should appear the same way. Manual updates across three delivery apps plus your own site guarantee that something will be wrong at any given moment.
Payment and Invoicing
If you're running a direct ordering channel, you need payment processing that works. This isn't just about accepting cards. It's about reconciliation across multiple channels, automatic invoice generation, and clear reporting on what you earned versus what you collected after commissions and fees.
Step 5: Design for Delivery Operations
The physical layout of your cloud kitchen matters more than most first-time operators expect. You're not building a restaurant kitchen. You're building a production line.
Station Design
Organize your kitchen around workflow, not traditional kitchen hierarchy. Prep station, cooking station, plating and packaging station, handoff station for delivery drivers. The path from raw ingredient to sealed bag should be as short and linear as possible.
Packaging as Product Design
This is something traditional restaurants never think about, but cloud kitchens have to. Your packaging is the restaurant. It's the only physical interaction the customer has with your brand. Leaky containers, soggy food, and collapsed boxes kill repeat orders faster than mediocre taste.
Invest in packaging that maintains temperature, prevents leaks, and looks intentional. Tamper-evident seals are non-negotiable. Custom branding on packaging turns every delivery into a marketing interaction. A customer eating from a plain foil container forgets who made the food. A customer eating from a branded box remembers.
Delivery Handoff Zone
If you're using third-party drivers, they need a place to pick up orders that doesn't involve them walking through your kitchen. Designate an exterior handoff shelf or counter, clearly labeled with your brand name. Drivers grab and go. Your team stays focused on cooking.
Step 6: Launch, Measure, and Adjust
The first two weeks of operation will teach you more than any planning document. Here's what to watch.
Order Volume by Channel
Which platform is sending you the most orders? Which has the highest average ticket? Which generates the most repeat customers? Track this from day one.
Prep Time per Item
Time every dish from order to bag-seal. Know your actual prep times versus what you estimated. If your average order takes 22 minutes to fulfill but you promised 15, your reviews will suffer and your delivery app ranking will drop.
Customer Feedback Patterns
Read every review. Not to feel good or bad about your food, but to identify patterns. If three people say your noodles arrive clumped, that's a packaging problem, not a cooking problem. If five people say the portion is small, that's a pricing problem, not an opinion problem.
Contribution Margin per Order
Revenue minus food cost, packaging cost, commission, and payment processing. Know what you actually keep on each order. A $25 order on a 28% commission platform might net you $12 after all costs. A $25 order through your direct channel might net you $20. Same food. Different economics.
What Most Cloud Kitchen Operators Get Wrong
They launch on three delivery apps simultaneously. Each app has its own tablet, its own menu format, its own driver ecosystem, its own customer expectations. Managing three platforms plus your own channel from day one is an operational mess. Start with one. Add others only when your processes are solid.
They underprice to get reviews. Discounting your way to volume works until you stop discounting and the orders disappear. Price your food for sustainably from day one. If your concept can't survive at full price, the concept needs work, not a 40% off coupon.
They ignore the direct ordering channel. Every order on a delivery app builds the app's customer base, not yours. From your first week, start directing customers to your own storefront. A QR code in every delivery bag. A link in your social media profiles. A discount code that only works on your direct store. Build the habit early.
They treat the kitchen as the business. The kitchen is production. The business is marketing, technology, data, and customer relationships. If you spend all your time perfecting recipes and none of your time studying your order data, you're running a hobby, not a business.
How Ekada Supports Cloud Kitchen Operators
Running a cloud kitchen means managing everything from one screen. Ekada gives you the infrastructure to do that without juggling five apps and three spreadsheets.
- Whitelabel digital storefront — Your own ordering page, branded, mobile-optimized, shareable via WhatsApp, Instagram, or QR code. No marketplace logo. No competitor suggestions next to your menu. Your brand, your customers, your data.
- Real-time menu management — Add items, change prices, 86 sold-out dishes, and everything updates across all channels instantly. No more calling three apps to tell them the special is gone.
- Order management dashboard — See every order from every channel in one place. No switching between tablets. No missed tickets. No confusion about which driver picked up which bag.
- Integrated payments and invoicing — Every order tracks revenue, costs, commissions, and net margin automatically. You know what you earned the instant the order completes, not 48 hours later when the payout report loads.
- Customer data ownership — Every customer who orders through your direct store is yours. Email, phone, order history, preferences. Use it for re-engagement, promotions, and loyalty. The delivery apps will never give you this.
- Analytics that matter — Not just order counts. Prep time tracking, channel contribution margin, repeat customer rates, and menu item performance. The data you need to make decisions, not the data that looks good on a chart.
One platform. Zero commission on your direct orders. Every process a cloud kitchen needs to run smoothly.
Free to start. No credit card required.
Start Your Free Ekada Account | Book a Personalized Demo
The overhead savings of a cloud kitchen are real. The operational complexity is real too. The operators who succeed are the ones who treat their kitchen like a production facility and their technology stack like their most important employee. Get both right, and the margins speak for themselves.
Frequently Asked Questions
How much does it cost to start a cloud kitchen? Startup costs range from $5,000-15,000 for a commissary kitchen setup with basic equipment and permits, to $50,000-100,000+ for a dedicated space with full buildout. The biggest variables are your market (rent), your equipment needs (concept-dependent), and whether you're building from scratch or moving into an existing facility. Most operators can launch a lean cloud kitchen in a commissary space for under $10,000 including first month's rent, permits, initial inventory, and packaging.
Do I need a restaurant license for a cloud kitchen? Yes. Even though you have no dining room, you need a food establishment permit, business license, and food safety certification. The specific permits vary by city. Some municipalities have specific ghost kitchen or cloud kitchen permit categories. Others require a standard restaurant license. Contact your local health department before signing a lease or committing to a space.
Can I run multiple brands from one cloud kitchen? Yes, and many operators do. A multi-brand kitchen runs two or more distinct menus from the same physical space. The key is choosing brands that share prep workflows and can use overlapping ingredients, so you're not buying separate inventory for each concept. Most delivery platforms allow multiple brand listings from one physical address as long as each brand has a distinct menu and identity.
How do cloud kitchens handle delivery? Three options: third-party delivery apps (Grab, Foodpanda, Gojek, etc.), in-house delivery drivers, or a hybrid approach. Most new operators start with third-party apps for their built-in driver network and customer base, then add direct ordering with their own delivery or flat-rate courier partnerships as volume grows. The important thing is to build a direct ordering channel from day one, even if most early orders come through apps.
What's the biggest mistake new cloud kitchen operators make? Launching on too many platforms at once before operations are stable. Juggling orders from three delivery apps plus your own channel creates chaos in the kitchen and guarantees missed orders, slow prep times, and bad reviews. Start with one channel, build consistent operations, then expand. The second biggest mistake is ignoring the direct ordering channel, which means paying 25-30% commission indefinitely instead of building a customer base you own.