RestauNax Logo
Home FoodPricingAffiliate
Starting a Restaurant

How to Start a Virtual Restaurant Brand From a Ghost Kitchen or Your Own Line

A virtual restaurant brand is a delivery-only menu cooked in a kitchen you already have. This guide covers the concept, the marketplace rules, the launch costs and a worked example.

November 14, 2026

10 min read

a close up of a person holding a container of food

Photo by Kris Len Lu on Unsplash

To start a virtual restaurant brand, pick a delivery-only concept your existing kitchen can cook with ingredients and equipment it already has, cost and photograph a short menu, check your health permit covers it, list the brand on delivery marketplaces and your own ordering site, then review sales weekly and adjust.

The rest of this guide walks through each step, with the marketplace rules that have been published, the costs to plan for and a worked example.

What is a virtual restaurant brand?

A virtual restaurant brand is a restaurant that exists only online: a name, a menu and photos, with the food cooked in a kitchen that belongs to another operation. Uber Eats defines it as "an online-only food delivery business that works out of the same kitchen as a brick-and-mortar restaurant". DoorDash (2026) calls it a "delivery-only brand" that lets a restaurant launch an additional restaurant.

It is not the same thing as a ghost kitchen. A ghost kitchen is a place: a kitchen with no dining room. A virtual brand is a menu, and it can be cooked in a ghost kitchen or on the line of a busy dine-in restaurant. What is a ghost kitchen covers the kitchen side.

Virtual brand in your restaurantVirtual brand in a ghost kitchen
KitchenThe one you already runA rented delivery-only kitchen
Extra rentNoneYes
Capacity limitYour slow hours and spare stationsThe size of the bay you rent
Main riskSlowing down dine-in serviceCovering rent with delivery sales alone
Swipe to see more →

Can I run several brands from one kitchen?

Yes. One kitchen can cook for several brands, and that is the usual reason to start one: the same fryers, cooks and inventory earn more per hour. The limits are practical and regulatory.

  • Capacity. Every brand adds tickets to the same line. If Friday dinner is already at the limit, a second brand makes both slower.
  • Marketplace rules. Marketplaces do not want the same menu listed five times. In March 2023 Uber Eats set standards that, as Nation's Restaurant News reported, required a virtual brand with a brick-and-mortar parent to have at least 60% of its menu items differ from the parent's menu, images on at least five items, a description on every item, a 4.3-star rating or higher for parent and virtual brand, no more than 5% of orders cancelled by the operator, and no more than one new virtual brand per location every four weeks. Rules change, so ask each marketplace for its current policy before you design the menu.
  • Licensing. Your health department decides whether a second trade name operating under your existing food permit needs anything more. Ask before launch. This is general information, not legal advice.

How do you choose a concept for a virtual brand?

Choose a concept that uses what you already buy and cook, fills a gap in local delivery, and travels well. Work through these questions:

  1. What do you already have? List your equipment and the ingredients you stock in volume. A pizzeria has ovens, dough, cheese and chicken wings in the walk-in. A wing or flatbread brand needs almost nothing new.
  2. What is missing nearby? Open the delivery apps at your address at lunch, dinner and late night. Note the cuisines with few options or poor ratings.
  3. When is your kitchen idle? A breakfast burrito brand suits a dinner-only restaurant. A late-night brand suits a kitchen that closes its dining room at nine.
  4. Does it survive the ride? Test every dish after 25 minutes in its packaging.
  5. Is it different enough? A second listing of your main menu under a new name is the pattern marketplaces remove.

How do you start a virtual restaurant brand, step by step?

Start with the menu maths and finish with a small launch you can measure. The steps:

  1. Write a short menu. Eight to fifteen items is enough. Share prep across items so a slow day does not create waste.
  2. Cost every item. Add ingredient cost and packaging cost per order, then set prices that still work after the marketplace's commission. The menu pricing calculator does the per-item sums.
  3. Name and brand it. Pick a name that says what the food is. Check that the name is free as a web address and on social platforms, and is not someone else's trademark in your category.
  4. Photograph the food. Every item needs a photo and a plain description. Uber Eats made both a published requirement in 2023.
  5. Confirm permits and tax. Ask your health department and your state tax agency whether a new trade name under your existing business needs a registration.
  6. Set up the channels. Create the marketplace listings, and open a direct ordering site for the brand so repeat customers can order without commission.
  7. Train the line. Agree which station cooks which items, how the tickets are told apart, and who packs and checks the bag.
  8. Launch quietly for two weeks. Limit hours at first. Watch prep times, cancelled orders, ratings and item sales.
  9. Review and trim. Cut the items that do not sell. Menu engineering gives you a method for that decision.

What does it cost to launch a virtual brand?

A virtual brand cooked in your existing kitchen has no build-out and no extra rent, so the launch cost is mostly menu development, photography, packaging and the tools you sell through. Uber Eats and DoorDash both describe the model as needing little upfront investment and neither publishes a figure, so the table below lists what to budget rather than inventing an average.

CostWhat drives itOne-off or ongoing
Recipe development and tastingIngredients and staff time for testingOne-off
Name, logo and menu designWhether you do it in-houseOne-off
Food photographyNumber of items; your own phone or a photographerOne-off
PackagingContainers, bags, labels per orderOngoing, per order
Extra ingredientsOnly what the new menu adds to your ordersOngoing
Extra laborMore tickets on the same lineOngoing
Marketplace commissionDoorDash publishes 15%, 25% or 30% on Marketplace delivery; Uber Eats publishes 20%, 25% or 30%Ongoing, per order
Direct ordering siteYour ordering platform's subscription and card processingOngoing
A rented ghost kitchen, if you use oneLocal rents; get quotesOngoing
Swipe to see more →

Commission rates are from the DoorDash and Uber Eats merchant pricing pages, checked November 14, 2026.

Worked example: a wing brand run from a pizzeria

The numbers below are assumptions chosen to show the method. Replace each one with your own.

A pizzeria launches a wing brand. It already stocks wings, sauces and fryer oil. In a typical week the brand takes 40 orders at an average of $28.

All 40 orders on a marketplace20 marketplace, 20 direct
Sales$1,120.00$1,120.00
Marketplace commission at 25%$280.00$140.00
Card processing on direct orders (2.9% + 30¢)$0.00$22.24
Food cost at 32% of sales$358.40$358.40
Packaging at $1 per order$40.00$40.00
Left before labor and overhead$441.60$559.36
Swipe to see more →

Direct card processing: $560 x 2.9% = $16.24, plus 20 x $0.30 = $6.00. Delivery fees on direct orders are assumed to be paid by the customer at checkout.

Two things stand out. First, the brand adds $441.60 a week from a kitchen and a crew that were already there, provided the line can absorb 40 more tickets. Second, moving half the orders to a direct channel adds $117.76 a week, or about $6,100 over 52 weeks, without selling one more order.

How do virtual brands get customers without the apps?

Virtual brands get customers outside the apps by turning each marketplace order into a direct one next time. Nobody walks past a virtual brand, so the first order usually comes from a marketplace. The second does not have to.

  • Put a card in every bag with the brand's own ordering address and a first-order coupon.
  • Give the brand its own website and app, so a search for the brand name finds you and not only a marketplace listing.
  • Collect contact details on direct orders and send email and text offers.
  • Run a loyalty program that rewards direct orders.
  • Use the parent restaurant. If the brands share an audience, mention the new brand on the parent's menu, receipts and social accounts.
  • Post the food on the social platforms where your customers are. See the restaurant marketing guide for a full plan.

Third-party delivery's effect on restaurant margins explains why this shift matters more as volume grows.

What goes wrong with virtual brands?

Most virtual brands that fail do so for one of five reasons, all avoidable:

  1. The menu is a copy. It duplicates the parent menu, so the marketplace removes it or customers feel misled.
  2. The kitchen cannot keep up. Dine-in guests wait longer and ratings fall for both brands.
  3. The prices ignore commission. Items sell well and lose money.
  4. The food does not travel. Soggy food earns low ratings, and ratings decide placement.
  5. There is no direct channel. Every repeat order pays commission forever.

Be open with customers about who cooks the food. A brand that names its kitchen on its website loses nothing and avoids the complaint that the restaurant "is not real".

Where RestauNax fits

RestauNax gives a brand its own ordering website and branded app, loyalty, coupons, email and text marketing, a kitchen display and delivery through on-demand couriers for $99/month per location with 0% commission; card payments are 2.9% + 30¢. One thing to know before you plan around it: RestauNax does not have a multi-brand dashboard. Each brand is set up as its own restaurant with its own menu, website and app, and marketplace orders stay on the marketplace's tablet. Talk to us about how several brands in one kitchen would be set up: see ghost kitchen software or book a free demo.

Virtual restaurant brand FAQ

Is a virtual restaurant brand legal?

Yes, when the kitchen is properly licensed and the brand is honestly described. Whether a second trade name needs its own registration or permit depends on your state and local health department, so ask them. This is general information, not legal advice.

Do I need a ghost kitchen to start a virtual brand?

No. Most virtual brands start in a restaurant kitchen that already exists. Renting a ghost kitchen makes sense when your own line has no spare capacity or you want to reach a delivery area your restaurant does not cover.

How many virtual brands should one kitchen run?

As many as the line can cook well at peak, and usually that is fewer than owners expect. Start with one, measure prep times and ratings for a month, then decide. Uber Eats' 2023 rules limited merchants to one new virtual brand per location every four weeks.

How long does it take to launch a virtual brand?

It depends on how much is new. A brand built from ingredients and equipment you already use needs a menu, photos, packaging and listings; a brand that needs new suppliers and training takes longer. We have not found a reliable published average, so plan from your own task list.

Should a virtual brand have its own website?

Yes. A website and app under the brand's name give customers a way to reorder without a marketplace, give you their contact details, and make the brand findable in search.

Sources

  • Uber Eats for Merchants, "What is a virtual restaurant?"
  • DoorDash for Merchants, "What Is a Virtual Kitchen? How to Start a Virtual Restaurant" (September 29, 2026)
  • Nation's Restaurant News, "Uber Eats is starting to crack down on redundant and poorly rated virtual brands" (March 28, 2023)
  • DoorDash and Uber Eats merchant pricing pages, checked November 14, 2026
Tags:
virtual restaurant
virtual brand
ghost kitchen
delivery-only restaurant
online ordering
restaurant growth

About the Author
RestauNax Team
RestauNax Team

RestauNax

The team building RestauNax, writing about restaurant technology and the real cost of third-party delivery.