Verity for cloud kitchens
A cloud kitchen shares staff, stock and equipment across brands that report revenue separately. Verity attributes the shared cost so per-brand contribution stops being an estimate.
Verity manages kitchen operations. Aggregator platforms stay where they are.
Verity / Kitchen
LiveOrders today
412
Median dispatch
11m 40s
Items unavailable
17
Packaging cost
₹9.40
Needs attention
Illustrative figures. Verity shows your own kitchen in this shape.
How the business runs
Several brands run out of one kitchen using the same staff, the same stock, the same equipment and the same rent. Each brand reports its own revenue through the platforms it sells on. Almost nothing reports its own cost, because the cost is shared. So the fundamental question — which brands are actually contributing and which are being carried — is answered by allocating overheads on a percentage somebody chose.
The second structural difference from a restaurant is that there is no dining room, which means there is no direct customer relationship and no recovery from a bad experience. The entire quality signal is dispatch time and item availability, both of which are operational facts the kitchen controls and neither of which is usually measured against the brand or the item that caused them.
The third is packaging. A cloud kitchen ships every order in packaging that costs real money per order, and that cost moves. In a business where the platform takes a commission on the gross, a rupee of packaging is a meaningful share of what is left.
The fourth is capacity. Brands compete for the same stations at the same peak, and a menu decision taken for one brand slows the others. That interaction is invisible unless production is recorded against station and time.
Verity records the order with its brand, items, station, dispatch time and packaging, and attributes shared stock and labour to the brand that consumed them.
What Verity calls these things
What gets in the way
Every cloud kitchen difficulty comes from a cost base that is pooled and a revenue line that is not.
Per-brand contribution is an allocation, not a measurement
Stock, labour and packaging are shared, so brand profitability is calculated by splitting overheads on a percentage rather than on what each brand consumed.
In Verity Consumption is recorded against the order that caused it, so contribution per brand is built from actual usage.
Dispatch time is the only quality signal and nobody attributes it
A slow order damages the brand’s platform rating, and the kitchen knows the average without knowing which brand or which item is causing it.
In Verity Dispatch time is recorded per order with its brand, items and station, so the cause is identifiable.
Items stay listed when they cannot be made
An item runs out and remains live on the platform, producing cancellations that cost both the order and the rating.
In Verity Stock availability is connected to the items it supports, so unavailable items surface immediately rather than at the next cancellation.
Brands collide at peak
Two brands need the same station between half past seven and nine, and both slow down without anyone attributing the interference.
In Verity Production is recorded against station and time, so capacity conflicts are visible in the same view as the delays they cause.
Packaging cost is treated as overhead
Packaging is a per-order cost that moves with supplier prices, and it is usually absorbed into a general expense line.
In Verity Packaging is stock consumed per order, so its cost per order and its movement are visible.
Commission is netted before anyone sees the gross
Platform revenue arrives net of commission and adjustments, and reconciling what was ordered against what was received is a monthly chore.
In Verity Orders are recorded as they occur, so platform settlements can be reconciled against the kitchen’s own record.
The complete system
One system across brands, production, cost attribution and dispatch.
Orders
Every order records its brand, platform, items, station, preparation and dispatch times, packaging used and any cancellation or refund.
Why it matters here The order is the only place brand revenue and kitchen cost meet, which makes it the only place contribution can be measured.
In practice Four hundred and twelve orders across five brands with a median dispatch time, and the outliers attributable to specific items.
Inventory
Stock is held with supplier and cost and is consumed by the orders that use it, including packaging per order.
Why it matters here Shared stock is exactly what makes brand contribution unmeasurable, and recording consumption per order is what fixes it.
In practice Packaging at nine rupees forty an order, up one rupee twenty, applied to every order rather than buried in overhead.
Work
Production is work with a station, an owner, a timestamp and the order it belongs to.
Why it matters here Dispatch time is the product in a cloud kitchen, and it is made of recorded steps rather than of a single number.
In practice A brand running forty percent above the kitchen’s median, traced to three items at one station.
Records
Brands, listed items and their recipes are records with components, so each item connects to the stock it consumes.
Why it matters here Several brands frequently share ingredients, and only the recipe link makes shared consumption attributable.
In practice Two brands drawing on the same base preparation, with consumption attributed to each by orders sold.
People
Staff are modelled once, and every production step, dispatch and wastage record carries who handled it.
Why it matters here Labour is the largest shared cost, and attributing it to stations and periods is what makes brand contribution honest.
In practice Hours worked by station against orders produced at that station.
Workforce
Assignment, attendance and availability stay connected to the periods and stations they covered.
Why it matters here Cloud kitchen demand is sharply peaked and entirely predictable from history, which makes understaffing avoidable.
In practice Staffing set against orders per fifteen-minute interval rather than against a flat shift.
Suppliers
Suppliers and platforms are relationships with their terms, commissions, settlements, delivery reliability and balances.
Why it matters here A platform is a supplier of demand with commercial terms, and its settlements need reconciling like any other account.
In practice Platform settlement reconciled against the kitchen’s own order records for the period.
Workflows
Item availability changes, refunds, comps and write-offs move through defined steps with recorded reasons.
Why it matters here Taking an item down is a commercial decision with a rating consequence, and it should be recorded rather than shouted across a kitchen.
In practice An item marked unavailable carries the stock reason and the time it was taken down.
Reports and analytics
Contribution by brand, dispatch time by brand and item, cancellation rate, packaging cost per order and station utilisation come from the operational records.
Why it matters here These are the numbers a cloud kitchen operator actually needs and the ones no platform dashboard provides.
In practice Contribution per brand after attributed stock, labour and packaging, rather than after a percentage allocation.
Verity AI
Verity AI answers from your own order, production, stock and platform records, respects permissions, and can create assigned follow-ups.
Why it matters here The questions worth asking cross brand, item, station and time at once, which is exactly what separate platform dashboards cannot do.
In practice "Which items drive dispatch time above target?" returns three, with a prep review assigned to the station lead.
Locations
Locations roll into the business, with stock, production and reporting following the same structure.
Why it matters here Operators run several kitchens with different brand mixes, and comparison requires identical recording.
In practice Dispatch time and contribution by kitchen and by brand within it.
Communication
Notes, notifications and activity attach to the order, item or brand they concern.
Why it matters here There is no dining room, so the only record of what went wrong is what the kitchen writes down.
In practice A delay caused by a late ingredient delivery recorded on the orders it affected.
Control
One permission model and one audit trail across every record.
Why it matters here Taking items down and issuing refunds both have direct revenue and rating consequences.
In practice Every availability change and refund carries the person, the reason and the time.
Work in motion
These already happen. Recorded per order and per station, they produce the attribution the business is missing.
Dispatch time is the product. Recording its steps is what makes it improvable.
An item live on a platform that the kitchen cannot make is the most expensive failure available.
This replaces a percentage allocation with a measurement, which frequently changes which brand looks worth keeping.
Two brands needing one station at the same time is a planning problem, not a staffing failure.
Reconciling against the kitchen’s own records is the only way to catch adjustments you did not agree to.
Packaging is a per-order cost, not an overhead, and treating it as one hides real erosion.
Verity AI
Verity AI reads the same order, production, stock and platform records the kitchen creates as it runs. It answers from your own operation, only shows what the person asking can see, and can turn the answer into work assigned to a station lead.
Verity / Ask
Verity AI only returns what the person asking has permission to see.
Without chasing
Each runs from the order and stock records at the point the condition is met.
When
Stock falls below what a listed item requires
When
Dispatch time exceeds target on an item
When
Order volume is forecast above staffed capacity
When
A platform settlement is received
When
Packaging cost per order moves beyond tolerance
When
A refund or comp exceeds the threshold
What you can understand
Attribution across brands from records created per order.
Platform dashboards report each brand’s revenue. This is the cost side, attributed to the brand that caused it.
One system, different ways of seeing it
Everyone works from the same records, and each opens on what they need.
Operator
Which brands are worth running?
Contribution per brand after attributed cost, packaging movement, commission, kitchen comparison.
Kitchen manager
What is slowing us down?
Dispatch times by item and station, availability, peak conflicts, staffing against volume.
Station lead
What is on my station and what is short?
Orders queued by station, prep outstanding, stock supporting listed items, wastage to record.
Accounts
Does the settlement match what we made?
Platform settlements against recorded orders, refunds and adjustments, supplier payables.
Where it is used
Stock, labour and packaging attributed to the orders that consumed them, replacing a percentage allocation with a measurement.
Preparation and dispatch recorded per order with brand, item and station, so the cause of a slow median is identifiable.
Stock connected through recipes to listed items, so an item the kitchen cannot make comes down before it produces cancellations.
Order volume by interval against station demand by brand, so conflicts are planned around rather than discovered during service.
Packaging consumed and costed per order rather than absorbed into overhead, where its movement is invisible.
Settlements matched against the kitchen’s own order records, so commission and adjustments are checkable.
Kitchens as locations recording identically, so dispatch time and contribution are comparable across sites.
Plain-language questions across brand, item, station and time at once, with follow-ups assigned in the same step.
Getting there
Your aggregator arrangements and billing continue as they are and are mapped during implementation. Menus, recipes, suppliers and packaging costs are brought across, and Verity is introduced as the operational and cost-attribution layer over them.
Implementation runs about four weeks: discovery and mapping, configuration, migration, then an ongoing operations partnership.
Questions
Verity AI answers questions from your own order, production, stock and platform records: contribution per brand after attributed cost, which items drive dispatch time above target and at which station, which listed items are currently unavailable, where two brands compete for one station at peak. Each answer can become work assigned to a station lead.
It attributes shared cost rather than allocating it. Ingredient consumption is attributed through recipes to the orders that used it, packaging is costed per order, and labour is attributed by station and period, so contribution per brand is built from measurement rather than from a chosen percentage.
No. The platforms remain where the orders come from and are mapped during implementation. Verity is the kitchen-side operational layer — production, stock, dispatch times, availability, cost attribution and reconciliation against platform settlements.
Preparation and dispatch are recorded per order with the brand, items and station, so a slow median resolves to specific items at specific stations rather than remaining a kitchen-wide average.
Stock is connected through recipes to the items each brand lists, so when an ingredient falls short every affected item across every brand is identified and an availability change is raised before cancellations start.
Orders are recorded as they occur with gross value, so a settlement can be matched against the kitchen’s own records and commission, refunds and adjustments become checkable rather than accepted.
Kitchens and stations are locations rolling into the business, so dispatch time, contribution and availability are directly comparable across sites with different brand mixes.
About four weeks: discovery and mapping of how the kitchen runs its brands, configuration, migration of menus, recipes and suppliers, then an ongoing operations partnership.
Most operators are carrying at least one brand and cannot prove which. Tell us how costs are split today and we will show you what measuring them looks like.