guides — 2026-05-18

Ghost Kitchen Menu Design: How Virtual Brands Win on Delivery Apps

How to design a ghost kitchen menu that converts on delivery apps in 2026 — concept narrowing, item count, photo coverage, pricing, and running multiple virtual brands from one kitchen.

A Ghost Kitchen Menu Is a Different Animal

A traditional restaurant menu is a document a guest reads while sitting in your dining room. They've already arrived. The menu's job is to help them choose what to order, then maximize the order size.

A ghost kitchen menu is your entire restaurant. The guest never visits. They never see a dining room, never meet the staff, never smell the kitchen. Every decision they make about your concept — whether it exists, whether it's good, whether to order it — happens inside a delivery app listing they spend 8-15 seconds scanning.

This changes how the menu has to be designed. The principles that work for dine-in menus (variety, ambient cues, server-driven upsells) actively hurt on a delivery app. The principles that work for ghost kitchens (focus, photo coverage, search-friendly naming) are often the opposite.

This guide is the practical playbook.

The Ghost Kitchen Menu Constraint Set

Before any design decisions, internalize the constraints that shape every other choice:

Customer attention is 8-15 seconds. That's how long a guest spends scanning your listing on DoorDash or Uber Eats before swiping. Your menu has to communicate concept, value, and order intent in that window.

The first 3-5 items carry the listing. Most guests never scroll past the first screen of items. Your top 3-5 dishes account for 60-80% of orders.

Photos are not optional. Items without photos consistently underperform by 30-50%. On delivery apps where photos are the only sensory cue, this gap widens.

Search-friendly names beat clever names. "Nashville Hot Chicken Sandwich" wins searches that "The Big Cluck" loses. Optimize for what guests type, not what sounds fun.

Multiple brands compound complexity. If you run 3 virtual brands from one kitchen, each menu needs its own design, photos, naming, and pricing. Shortcuts here are visible to customers and platform algorithms.

Margins are squeezed by 15-30% commissions. Pricing has to absorb the platform take while still feeling competitive. Underpricing destroys margins; overpricing kills volume.

These six constraints drive every recommendation that follows.

The Right Number of Items

The single most common ghost kitchen menu mistake is too many items. Operators who came from a sit-down concept default to 40-60 item menus. On delivery apps, this hurts.

The right number is 15-25 items for a single ghost kitchen concept. The reasoning:

Customer decision fatigue is real. A 40-item menu on a 5-inch phone screen means guests choose the cheapest familiar item, not your hero dish.
Operational complexity multiplies. Each additional item means more inventory, more training, more 86-out risk, more potential for negative reviews.
Photo coverage gets harder. 50 items with photos = 50 photos to keep current. 20 items = 20 photos, refreshed quarterly without breaking a sweat.
Platform algorithms reward focused concepts. A menu with 20 items in one cuisine outranks the same kitchen offering 60 items across three cuisines.

The exception: build-your-own concepts (bowls, sandwiches, pizzas) where 4-6 "base" items with rich modifiers create variety without inventory complexity. Even here, the visible item count stays under 25.

Designing for the First Screen

The first screen of your delivery app listing is the only screen most customers will see. Three rules for what goes there:

The hero item leads. Your highest-margin, highest-popularity, most-photogenic dish goes first. Not your cheapest. Not your simplest. Your signature.

Items 2-5 are your bestseller deck. The dishes that, between them, represent 60-80% of orders. Photograph them beautifully. Price them in a tight band that signals consistency.

Items 6+ are depth. Important for the customer who's been before and wants to try something new — but they're not winning new customers.

The error most operators make: treating the menu as a structured document with starters → mains → sides → desserts. Delivery app guests don't browse that way. They scroll the first screen, pick the most appealing photo, and tap. Optimize for that behavior.

Naming Items for Search and Recognition

A ghost kitchen menu lives or dies on platform search. When a guest searches "fried chicken sandwich" on Uber Eats, your listing either appears or doesn't, based on what's in your item names and descriptions.

The rule: be searchable first, clever second. Naming patterns that work:

[Specific dish] [+ unique modifier]. "Nashville Hot Chicken Sandwich — Maple Glaze" wins over "The Big Cluck." The first 2-3 words match what customers search for; the modifier is your differentiator.

[Cuisine adjective] [base item]. "Korean Fried Chicken Bowl," "Sicilian Pizza Slice," "Vietnamese Banh Mi." Strong cuisine signal helps both search and customer expectation.

Avoid food puns in primary names. Move them to descriptions if you must. The pun doesn't appear in search; the dish type does.

Front-load the keyword. "Chicken Tikka Masala Bowl" outperforms "Bowl of Chicken Tikka Masala" even though they describe the same dish.

You can still have brand personality. Put it in descriptions, in the brand name itself, in the listing's hero image. The item names should be findable.

Photo Coverage and Consistency

Three things consistently kill ghost kitchen menu performance more than anything else:

Items without photos. Direct conversion hit. Estimated 30-50% fewer orders versus the same item with a photo.
Inconsistent photos across items. When the burger looks like a magazine shot and the wings look like a phone snap, the guest assumes the wings are not as good as the burger. Trust collapses.
Photos that don't match what arrives. This generates negative reviews. The food doesn't have to look identical to the photo, but it has to come from the same visual family.

The implication: photograph the entire menu in one session, in one style. Re-shoot quarterly as the menu evolves. Don't add an item to the menu without a matching photo.

This is where the cost math used to be brutal — a 20-item shoot at $1,500-$3,500 and 2-3 weeks, then a re-shoot every quarter when the menu changes. For ghost kitchens running 2-5 brands, multiply that by 5. AI food photography has collapsed both — see the AI vs traditional food photography costs for the math.

The brand calibration angle matters too. A ghost kitchen running multiple virtual brands needs each brand to have its own consistent visual identity. Customers who order from two of your brands and notice the photos look related are losing trust. Tools that lock a visual identity per brand and apply it across every photo make this enforceable at scale. See the AI menu design guide.

Multi-Brand Menu Strategy

Many ghost kitchens run 2-5 virtual brands from a single kitchen. The menu strategy here is its own discipline:

Each brand needs a distinct cuisine or angle. "Spicy Wings Co." and "BBQ Brothers" can coexist. "Italian Bistro" and "Italian Trattoria" cannot — customers will notice they're the same kitchen.

Each brand needs its own photo style. Different lighting, different plating presentation, different background palette. Same kitchen, different visual identity. This is the highest-leverage signal of brand distinctness.

Menus should not overlap. If the same chicken wing appears under two different brand names, customers will figure it out. Differentiate the actual offerings — different sauces, different sides, different sizes.

Operationally, share what's invisible. Same fryer, same prep, same inventory. The customer never sees the kitchen; they only see the menu and the packaging.

Packaging matters for repeat orders. A customer who ordered from Brand A on Monday and Brand B on Wednesday should not get food in identical packaging. Branded sleeves and labels are cheap; the loss of trust if they figure it out is expensive.

Pricing for the Commission Reality

The 15-30% delivery app commission has to be priced in. The naive approach — "price the same as dine-in" — destroys margins. The aggressive approach — "mark everything up 30%" — kills volume by making your listing uncompetitive.

The framework that works:

Start with the price the dish needs to support a reasonable margin after commission. If the food cost is $4, the dish needs to sell for ~$14 to leave acceptable margin after a 25% commission and standard overhead.

Bundle for perceived value. A combo (sandwich + side + drink) at a $2 discount looks generous and shifts customers toward higher-ticket orders. The commission is a fixed percentage; the margin on a higher ticket is structurally better.

Use modifiers to raise effective price without raising listed price. The $13 base sandwich appears competitive; the 3-4 high-margin modifiers (premium cheese, bacon, avocado) push the average ticket to $17. Customers don't compare modifier prices across listings.

Avoid price-anchor traps. A $7 item that costs you $4 looks cheap to customers but loses money after commission. Cut these items rather than absorb the loss.

For the deeper framework, see menu engineering 101 — Stars, Puzzles, Plowhorses, Dogs all apply, with the added wrinkle that delivery commissions tighten every category.

What to Photograph for a Ghost Kitchen Menu

A few format-specific notes on photography:

Hero shots over flat lays. Delivery app listings show square or 4:3 photos in a grid. Hero shots (3/4 angle, plate centered, food filling the frame) outperform flat lays for this format.

Show portion size honestly. Photos that look like restaurant-quantity dishes when guests receive a delivery-portion serving generate negative reviews. Bowl shots should look like the bowl they'll actually get.

Avoid styling cues that suggest dine-in. Linen tablecloths, restaurant lighting, plated garnish that won't survive a delivery bag — all generate disappointment when food arrives in clamshell packaging.

Show the meal in a delivery-realistic context where possible. Open clamshell, branded packaging visible, ready-to-eat presentation. This sets expectations correctly.

For the underlying technique, see food photography for restaurants. For delivery-specific format specs, see delivery app photo optimization.

A 2-Week Ghost Kitchen Menu Launch Plan

For a new virtual brand launching on DoorDash, Uber Eats, and Grubhub:

Week 1

Day 1-2: Lock the brand concept. Cuisine, angle, name. Compare your concept against existing listings in your delivery radius.
Day 3-4: Lock the menu. 15-25 items. Hero item, bestseller deck, depth. Pricing model.
Day 5: Photograph the entire menu. AI food photography for full coverage in one session.
Day 6-7: Build the platform listings. Item names optimized for search, descriptions for differentiation, photos applied.

Week 2

Day 8-9: Test ordering. Order from your own listings. Time the delivery. Check the packaging. Review the photos in the customer flow.
Day 10-11: Soft launch. Light marketing. Watch first-order conversion and rating.
Day 12-14: Iterate. Adjust hero item, photo, name, or pricing based on what week-one data shows.

For a deeper end-to-end ghost kitchen launch playbook covering branding, operations, and platform setup, see the ghost kitchen complete guide and the DoorDash-specific launch guide.

Common Ghost Kitchen Menu Mistakes

Cloning a dine-in menu onto a delivery listing. Different format, different attention span, different customer behavior. Strip down, rephoto, rename.

Too many items. The instinct is to cover all bases. The result is decision paralysis and underperformance on every item.

Inconsistent photos. A magazine shot for the burger and a phone snap for the wings tells the customer the wings are not as good. Photograph in one session, in one style.

Generic brand names. "Tasty Eats" or "The Kitchen" rank for nothing and stand for nothing. Specific names with cuisine cues outperform.

Identical packaging across brands. Customers who order from two brands and receive food in the same boxes will figure it out. Cheap branded sleeves prevent the trust loss.

Set-and-forget operations. Ghost kitchens reward iteration. Quarterly menu refreshes, monthly photo refreshes for hero items, weekly review of ratings and performance.