A ghost kitchen operator managing four virtual restaurant brands out of one commissary facility has one food distributor account, one receiving dock, one buyer handling purchasing — and four separate brand P&Ls, four separate investor groups, and four separate recipe books, some of which share the same raw ingredients.
Tuesday's produce order covers roma tomatoes for the pizza concept's marinara, for the burger concept's fresh tomato garnish, and for the salad concept's daily prep. One line item on the broadliner invoice. Three concept budgets. The buyer cannot afford to wait until month-end to figure out who owes what.
That is the structural procurement problem for ghost kitchens and commissary-based multi-brand operations: the purchasing happens at the facility level, but the cost accountability has to land at the concept level — in real time, before the invoice hits accounts payable.
Quick answer: ghost kitchens need living POs
Ghost kitchen procurement works best when the purchase order stays alive from order placement through supplier confirmation, commissary receiving, and concept cost allocation — updating as supplier replies arrive, capturing the actual quantities and prices received, and feeding each brand's food cost record without a month-end reconciliation sprint.
That is closed-loop procurement at commissary scale: the demand signal (brand-level recipe consumption), the purchase order (facility-level consolidated ingredients), the supplier reply (confirmations, substitutions, price changes), and the receiving record (actual quantities delivered per line item) stay connected in a single live object. The order does not freeze at send. Substitutions do not live in an email thread. Short shipments do not appear as a surprise at the weekly food cost review.
See Living Purchase Order for the full model.
The shared-vs-dedicated cost split
Every commissary facility carries three categories of purchasing, and the cost allocation logic is different for each.
Category A — Shared operating supplies: Cooking oil, gloves, cleaning supplies, packaging film, aluminum pans. These are facility expenses rather than concept-specific food costs. They belong in a facility overhead budget, not in any brand's COGS line.
Category B — Brand-dedicated ingredients: Lobster for the premium seafood concept, specialty cheese blend for the pizza concept, house-made hot sauce produced exclusively for the wing brand. These go 100% to one brand's cost the moment they enter the facility. No allocation math required.
Category C — Cross-concept raw ingredients: Roma tomatoes, ground beef, eggs, butter, cooking wine. These are ordered in bulk at facility pricing, but consumed by multiple brands through separate production runs and recipes. This is where the allocation problem lives.
For Category C, the allocation has to follow recipe usage — not purchase weight, not invoice line position, not an estimate. If 60 lbs of roma tomatoes are received and the pizza concept's marinara production that week consumed 36 lbs while the burger concept used 18 lbs and the salad concept used 6 lbs, the cost allocation is 60%, 30%, and 10% — period. The moment the facility tries to split it by headcount or by rough estimate, every brand's food cost becomes a fiction.
Multi-brand BOM: cost flows through production
The bill of materials in a ghost kitchen context has two layers that a single-concept restaurant does not face: the raw ingredient BOM (what the facility purchases) and the production BOM (what each brand's dish actually uses, after prep, production, and yield loss).
Consider a commissary that runs a beef prep production line every Monday for the week's service:
- Raw purchase: 80 lbs ground beef at $6.50/lb = $520 facility cost
- Production yield: After portioning, trim, and cook loss, usable yield is 78% = 62.4 lbs usable product
- Effective cost per usable lb: $520 ÷ 62.4 = $8.33/lb
- Allocation: Burger concept consumes 48 lbs, wrap concept consumes 14.4 lbs = 77% / 23%
- Concept cost impact: Burger brand carries $434, wrap brand carries $120
When the broadliner raises ground beef prices 14% on the next invoice, that yield-adjusted cost-per-lb changes immediately. Every dish that uses ground beef in either concept gets more expensive the moment the new price lands — not three weeks later when someone notices the food cost report is off. A system that re-costs recipes dynamically against supplier-confirmed prices surfaces that cross-concept margin hit before production runs start, not after.
The purchase price variance calculation is the same at commissary scale: PPV = (standard cost − actual cost) × quantity received. But the number has to surface by brand, not by facility. A facility-level PPV summary tells you your costs are up. A brand-level PPV tells you which concept is exposed, on which ingredient, and how much margin has drifted this week.
Ordering at facility scale from one supplier account
The facility buyer places one consolidated order to the broadliner — typically Monday or Tuesday for a Thursday delivery. One purchase order, one supplier contact, one delivery window.
The consolidation is the financial advantage: volume pricing, consistent delivery scheduling, one account rep. A facility buying across four concepts may hit $15,000–$30,000 per month in broadliner spend, qualifying for pricing tiers that no individual concept could reach on its own.
But the order still needs to be segmented at entry, not reconstructed from the invoice after the fact:
- 50 lbs chicken breast → 35 lbs wing concept, 15 lbs wrap concept
- 30 lbs salmon fillet → 100% seafood concept
- 12 lbs mozzarella → 100% pizza concept
- 40 lbs roma tomatoes → 22 lbs pizza, 12 lbs burger, 6 lbs salad
That segmentation needs to live in the purchase order as created — not as a manual allocation note on the invoice. If a line item gets substituted by the supplier, the system needs to know which concepts are affected and propagate the update to each brand's pending food cost.
Demand classification matters here too. The facility's aggregate demand for staple proteins — chicken, beef, fish — is typically smooth week to week, which makes straightforward reorder logic work well. But individual concept-level demand from delivery platforms can be lumpy. A viral social media moment can spike one brand's orders 3x on a Friday, creating a mid-cycle replenishment need that the weekly Monday order did not anticipate. The procurement system needs to handle both the planned facility cycle and the unplanned concept spike without conflating them in the weekly purchasing report.
See How Restaurants Order From Food Distributors for the baseline on broadliner ordering, cutoffs, portal ordering, and credit chasing.
Receiving at the commissary dock
One delivery arrives Thursday morning. The receiving manager checks 40+ line items against the broadliner PO. This is where the cost record for every brand's food cost this week is either built correctly or begins to drift.
Ghost kitchen receiving has three failure modes that single-concept restaurant receiving does not:
1. Substitutions that propagate incorrectly. The avocados were shorted and substituted with a different pack size at a higher per-unit cost. If the commissary records the substitution at the facility level without updating each concept's recipe cost, the pizza brand's guacamole appetizer will be costed at last week's avocado price until someone catches it.
2. Short shipments that create concept planning gaps. The seafood concept ordered 25 lbs of halibut for Friday service. The delivery was 17 lbs — the supplier confirmed the short via email Tuesday but the message did not reach the chef who ordered it. A living purchase order surfaces that 8-lb gap before the executive chef begins the prep schedule, not when the Friday night line is short by two covers' worth of protein.
3. Receiving that captures weight but not allocation. Standard commissary receiving might record "15 lbs ribeye received at $22/lb." The allocation of that ribeye to the steakhouse concept versus the premium brunch concept has to happen at receiving, not via a post-month allocation worksheet. Every day the allocation is deferred is a day the brand P&Ls show incorrect food cost.
The receiving step is where the gap between a closed-loop procurement system and a manual commissary process becomes visible at month-end. Closed-loop means the receiving event captures quantity, actual price, and concept allocation — and that data reaches accounting before the payable is approved, not as a reconciliation effort after.
Per-concept food cost reporting
The financial deliverable for a multi-brand commissary is not a facility food cost percentage. It is a food cost percentage per brand, per week, per period — that each concept's owner or operator can use to make pricing, portioning, and sourcing decisions.
Industry benchmarks by concept type:
- Quick-service virtual concept (burgers, wraps, wings): 28–33%
- Full-service virtual concept (premium proteins, plated dishes): 35–42%
- High-margin beverage concept: 18–25%
A facility-level food cost of 34% can mask a burger concept running at 29% and a seafood concept running at 43%. Those two numbers require entirely different operational responses. The burger concept is healthy; the seafood concept needs either a price increase, a portion adjustment, or a supplier negotiation on protein pricing.
The only way to generate an accurate per-concept food cost is if the cost allocation happened correctly at every upstream step: at purchase (which line items belong to which concept), at receiving (what was actually delivered at what price), and at production (how much of the received inventory was consumed by which brand's production run). Miss any of those steps and the concept P&L is an estimate, not a measurement.
Cost of goods sold is a procurement metric before it reaches accounting. A commissary operator who waits for the monthly accounting close to see food cost by concept has been operating on a 30-day information lag that compounds into meaningful margin loss.
What to look for in ghost kitchen procurement software
Four requirements distinguish software built for commissary-scale multi-concept operations from tools designed for single-concept restaurants:
1. Multi-concept cost allocation at the line-item level, at order entry. Not post-purchase invoice splitting. The concept assignment happens when the order is built, so substitutions and receiving variances can propagate correctly to each concept's food cost.
2. Recipe builder with yield ratios that flows through production. Each brand's dish recipes need to trace back to raw ingredients with yield factors applied — so that a 20% trim loss on beef is built into the cost-per-serving for every dish that uses it, and a supplier price change re-costs every affected recipe in every affected concept automatically.
3. Supplier reply parsing that updates the living PO before receiving. When the broadliner confirms the Tuesday order on Wednesday morning with two substitutions and one short shipment, that information should reach the facility buyer as a reviewable update to the existing PO — not as an email buried in the facility inbox that someone will remember to forward to the prep team.
4. Receiving at the facility level with per-concept line assignment. The commissary manager receives one delivery. The system captures quantities, actual prices, and concept allocation in a single receiving event — so the weekly food cost report by brand can be generated by end of business Thursday, not by end of the month.
See Best Food Cost Management Software for a broader comparison of tools by feature area.
Where LineNow fits
LineNow is a closed-loop procurement platform that connects POS-level or recipe-level demand, supplier purchasing, supplier communication, receiving, and accounting handoff. For ghost kitchen and commissary operations managing multiple virtual concepts from one facility, the fit is:
- Multi-concept ordering with line-item concept assignment — build one facility-level PO to the broadliner with each line item tagged to one or more concepts at the quantities consumed by each brand.
- Recipe builder with ingredient costing by concept — map each virtual brand's menu to its ingredients with quantities, yield ratios, and per-unit costs. Supplier price changes from confirmed invoices update every recipe that uses that ingredient, across every concept, before the next production run.
- Supplier communication across all channels — send purchase orders via email, WhatsApp, EDI, or supplier portal. AI reads supplier replies — price changes, substitutions, short-shipment notices, ETA updates — into reviewable order changes attached to the living PO so the facility buyer and each concept chef see the same current state.
- Commissary receiving with variance capture — structured receiving against the confirmed PO state, with short-shipment capture by line item and concept allocation at point of receipt. When the halibut order arrives 8 lbs short, that variance is in the seafood concept's food cost record before service begins.
- Accounting handoff — purchase data flows to QuickBooks Online or Xero with COGS classification by concept, so each brand's accountant or P&L owner sees food cost from a clean upstream record rather than an allocation worksheet built after the invoices arrive.
$100/month flat. No per-concept or per-location fees. 90-day free trial at linenow.co.
A quick diagnostic
Three questions:
- Can you generate an accurate food cost percentage per brand, this week, without waiting for month-end? No = your per-concept P&L is running on estimates, and margin drift is invisible until it compounds.
- When the broadliner substitutes an ingredient and confirms in an email, does that substitution reach each affected concept's recipe cost before the production run? No = the loop is open between supplier reply and the kitchen.
- Does your receiving event capture both the actual price delivered and which concept budget each line item belongs to? No = your facility-level invoice allocation is a spreadsheet exercise that happens after the fact, not a live procurement record.
If any answer is no, the commissary procurement loop is open. Closed-loop procurement closes it: facility-level purchasing discipline, supplier reply tracking, concept-level cost allocation at every step, and a clean food cost record that reaches each brand's P&L before accounting needs it.