Key takeaways
- Launch only if your kitchen has real spare capacity at the hours the brand will sell.
- Build the menu from at least mostly existing ingredients and stations, or the savings vanish.
- Model margin after a realistic marketplace fee before launch; many operators already make no profit on delivery orders.
- Set kill criteria before day one and run a 90-day test with weekly reviews.
Start with the honest question
A virtual brand is a second restaurant that lives only on delivery apps and online ordering, cooked in your existing kitchen. Done well, it sells extra orders from capacity you already pay for. Done badly, it slows your main restaurant, adds packaging and complexity, and loses money on every order. The difference is usually decided before launch, by how honestly you answer a few questions.
Use this checklist as a go or no-go tool. Work through each section with your chef and manager, and write the answers down. If several sections come back "no," the right decision is to wait. For background on formats and terms, see ghost kitchens and virtual brands explained.
1. Kitchen capacity
- Do you have hours each day when the line is clearly underused, such as mid-afternoon or late evening?
- Will the brand's likely busy hours fall in those quiet periods, or will they clash with your own peak?
- Does the brand rely on a station that is already your bottleneck, such as the fryer on Friday nights?
- Do you have room on the line, in storage and at the packing station for extra orders, containers and labels?
- Can you limit the brand's hours or pause it quickly on each platform when the main restaurant is busy?
If the brand's orders would arrive mainly at your own peak, the extra margin will be paid for in slower tickets and lower ratings for your main restaurant. That is the single most common reason virtual brands are switched off.
2. Menu overlap
- Can most of the menu be built from ingredients you already buy, in quantities you already handle?
- Can it be cooked with your existing equipment and skills, without new prep that needs extra staff?
- Do the dishes travel well? Run a travel test before deciding; see how to design a delivery-friendly menu.
- Is the menu short, ideally a handful of items plus sides and drinks?
- Will it look and taste like a specialist, rather than an afterthought?
Example: a fictional Mexican restaurant considers two virtual brands. A burrito bowl brand uses its existing proteins, rice, beans and salsas and needs only new bowls and labels. A smash burger brand would need a flat-top, new buns and beef, and a cook who knows the technique. The bowl brand passes the overlap test; the burger brand does not.
3. Market and platform rules
The marketplace is already crowded. Third-party delivery is available at 58% of limited-service and 32% of full-service restaurants[1], and popular categories fill with similar listings.
Source: National Restaurant Association, 2026[1]
Show the data
| Item | Value |
|---|---|
| Limited-service | 58% |
| Full-service | 32% |
- Open each delivery app at your address and search the category. How many listings would you compete with, and how do they look?
- Is there a gap: a cuisine, dietary need or price point that nobody covers well near you?
- Do the platforms you use allow virtual brands, and what are their rules on duplicate menus and how the kitchen is identified?
- Will each brand need its own tablet, or can orders be integrated with your POS? Read taming tablet chaos before adding more apps.
- Will you sell the brand on your own ordering page too, and if so, how will guests find it?
4. Unit economics
Work out the numbers per order before you spend anything on branding.
Estimate the average order
Use the prices of similar listings near you, at the app prices you would actually charge.
Subtract the marketplace fee
Two in three restaurants pay an average 15% to 29.9% of the transaction[2]. Use the effective rate on your own statements, including promotions.
Subtract food and packaging
Cost every item and its container kit.
Subtract direct labor
Only the extra labor the brand needs, such as an extra prep shift or a packer at peak.
Compare with your main menu
If the margin per order is lower than your main menu's delivery orders, ask why you would not just sell more of your own food.
Example: a fictional diner estimates a $22 average order for a breakfast sandwich brand. At an effective fee of 25%, the platform keeps $5.50. Food and packaging come to $7.70 and extra labor to about $3 an order at the expected volume. That leaves about $5.80 before overhead. The diner decides that is worth testing, but only during weekday mornings when the griddle has room.
Remember the context. Food and labor each take roughly 33 cents of every sales dollar[3], and in a May 2026 survey 47% of full-service and 43% of limited-service operators said they make no profit on third-party delivery orders[4]. Most operators price delivery menus above in-store prices to cope: 83% of limited-service and 66% of full-service restaurants do[5]. A virtual brand has no in-store price to compare with, which gives you more freedom to price for the channel.
5. Food safety and labeling
- Can you keep accurate allergen information for every dish? The eight major allergens accounted for 90% of US food allergies when federal labeling law passed[6], and sesame became the ninth, with FDA requirements from January 1, 2023[7].
- Will shared stations increase cross-contact risk, and how will you manage it?
- If you are part of a chain with 20 or more locations, have you checked how FDA menu labeling requirements apply[8]?
- Does your local health department need to know about additional brands, and do you need to register a new trading name?
6. Brand, listing and team
A virtual brand has no dining room, so the listing and the people behind it carry the whole brand.
- Name: short, easy to search and clear about the food. Check that it is not already used locally or registered by someone else.
- Logo and photos: a simple logo that reads at small sizes, and real photos of your dishes in their containers. Stock images undermine trust.
- Description: one or two sentences on what the brand cooks and what makes it worth ordering. Consider saying which kitchen it comes from.
- Owner: one named person responsible for the brand's menu, listing, statements and weekly review. Without an owner, a virtual brand drifts.
- Training: every cook and packer knows the brand's dishes, its labels and how its tickets look, before the first order arrives.
- Guest contact: a phone number or message route for problems, so guests reach a person rather than leaving a poor rating.
Keep the launch modest. Custom bags, a website and paid promotions can wait until the test shows demand. Stickers on your standard containers, a clean listing and accurate photos are enough to learn whether guests want the food.
7. A 90-day test plan
If the checklist mostly says go, test cheaply before you invest in custom packaging or a big launch.
| Days | Focus | Actions |
|---|---|---|
| 1–14 | Build | Final menu, travel tests, photos, listing on one platform, ticket routing and labels |
| 15–30 | Soft launch | Limited hours in your quietest window; no paid promotions; fix packing and ticket issues |
| 31–60 | Measure | Extend hours only if the kitchen copes; test one small promotion with an end date |
| 61–90 | Decide | Compare results with your kill and keep criteria; expand, adjust or switch off |
Review every week with three people: the chef, the manager and whoever owns delivery. Look at orders, margin after fees, ticket times for both brands, refunds and ratings.
8. Kill or keep: decide the criteria now
Write these before launch so the decision is not made on hope.
| Measure | Keep if | Switch off if |
|---|---|---|
| Orders per day | Steady or growing after day 30 | Falling for three weeks in a row without promotions |
| Margin per order after all fees | Above your target | Below target unless discounted |
| Main restaurant ticket times | Unchanged during brand hours | Rising during brand hours |
| Rating | Holding at or above your main brand's | Falling below your main brand's |
| Refunds and complaints | Low and falling | Rising, especially about quality |
If the brand passes, expand carefully: one more platform, slightly longer hours or a second menu item. If it fails, switch it off cleanly, remove the listing and reuse anything you learned for your main menu. A failed test that cost little is a good result.
For handing over orders when volume rises, see pickup lockers vs shelves vs counter handoff. And if the brand takes off and you want to sell it directly, read branded ordering app vs mobile ordering site before building anything.
Frequently asked questions
Is a virtual brand worth it for a small restaurant?
It can be if you have real spare kitchen capacity at the hours the brand would sell, a menu built mostly from existing ingredients and a margin that holds after marketplace fees. Test it for 90 days with clear kill criteria before investing heavily.
How long should I test a virtual brand?
About 90 days is a practical test: two weeks to build, two weeks of soft launch in quiet hours, a month to measure and a final month to decide using criteria set before launch.
What makes virtual brands fail?
The most common causes are orders that clash with the host restaurant's peak, thin margins after fees, menus that need new ingredients or skills, weak listings and guests who never come back.
Can I run more than one virtual brand from my kitchen?
You can, but each brand adds listings, tablets or integrations, packaging and labels. Prove one brand works and does not slow your main restaurant before adding another.
References
National Restaurant Association (2026). National Restaurant Association comments to the FTC on the Food Delivery Fees ANPRM (Project No. P267101) (opens in a new tab) ↑
National Restaurant Association (2026). National Restaurant Association comments to the FTC on the Food Delivery Fees ANPRM (Project No. P267101) (opens in a new tab) ↑
National Restaurant Association (2026). Restaurants remain resilient despite challenging business conditions (opens in a new tab) ↑
National Restaurant Association (2026). National Restaurant Association comments to the FTC on the Food Delivery Fees ANPRM (Project No. P267101) (opens in a new tab) ↑
National Restaurant Association (2026). National Restaurant Association comments to the FTC on the Food Delivery Fees ANPRM (Project No. P267101) (opens in a new tab) ↑
US Food and Drug Administration (2004). Food Allergies (opens in a new tab) ↑
US Food and Drug Administration (2023). Food Allergies (opens in a new tab) ↑
US Food and Drug Administration (2026). Menu Labeling Requirements (opens in a new tab) ↑