AI assistant · N2 Ready-to-use piece
Food Cost Analyzer for Restaurants
Computes your actual food cost from inventories and purchases, faces it against the theoretical from your recipes, and prices the monthly gap with the three most likely leaks.
N2Ready-to-use piece
Produces a piece you can use as-is: a menu description, a service script, a supplier email. Copy and go.
◆ This assistant 7 of the 164 published assistants are at this levelIs this for you?
What it does and what you get
- Most owners repeat a food cost somebody computed once and that no longer describes their kitchen.
- This assistant computes it properly, separates actual from theoretical and prices the monthly difference, because an uninvestigated three-point gap is money walking out the back door.
What you end up with
Your actual and theoretical food cost for the month, the gap between them in points and in money, and a short plan with the three most likely leaks and one action each.
EXAMPLE This is what you getAn example built with Sazón de Origen, the house's fictional restaurant.Not your data: your result is built with YOUR restaurant's.See the full example →Sample result
The owner asked this: «I have been repeating a 34 % food cost for two years and the bank does not back it up. I have purchases, sales and inventories for both locations.»
The 34 % the house kept repeating was not added up wrong: it was split wrong. It was food and beverage consumption divided by total sales.
| Line for the month (2 locations, COP) | Food | Beverage |
|---|---|---|
| Opening inventory | 18,400,000 | 6,900,000 |
| + Purchases in the period | 61,000,000 | 10,200,000 |
| − Closing inventory | 17,100,000 | 6,600,000 |
| = Cost of goods sold | 62,300,000 | 10,500,000 |
| Sales excluding tax | 178,000,000 | 36,000,000 |
| Actual cost of the family | 35.0 % | 29.2 % |
Added together and divided by the 214,000,000 of total sales they give exactly the declared 34.0 %. Split apart, the kitchen sits at 35.0 % and the bar at 29.2 %: two different businesses that had been averaging each other out for two years.
What went in and what stayed out, so next month compares against the same thing: comps (1,240,000) and staff meals (2,100,000) go in, because they come out of the same storeroom, and they are declared separately. Delivery packaging stays out, because it is not food. And it was confirmed with the administrator that the 214,000,000 are reported excluding consumption tax: if the tax were inside, this 35.0 % would be artificially low and the whole reading would be worthless.
The theoretical is built from each dish's current costing card multiplied by its units sold. It is not an opinion: it is what the kitchen should cost if everything came out as written.
| Dish | Units for the month | Price | Sales | Card cost | Total cost |
|---|---|---|---|---|---|
| House bandeja | 1,420 | 42,000 | 59,640,000 | 13,700 | 19,454,000 |
| Three-meat sancocho | 880 | 46,000 | 40,480,000 | 15,900 | 13,992,000 |
| Garlic trout | 610 | 52,000 | 31,720,000 | 17,800 | 10,858,000 |
| Rest of the menu | — | — | 46,160,000 | — | 12,300,000 |
| Total food | 178,000,000 | 56,604,000 |
Theoretical food cost 31.8 % · actual 35.0 % · gap 3.2 points.
In this month's money: 5,696,000 COP. The house sets its own threshold at 3 sustained points and this one crossed it, so it gets investigated this week and not next month.
The translator that makes the figure land: 1 point of food cost on food sales is 1,780,000 COP a month. And against the declared goal —lifting operating margin from 8 % to 14 % in 12 months without touching menu prices— those 5,696,000 are worth 2.7 of the 6 points that have to be recovered. Almost half of the year's goal sits inside this single gap.
One honest condition before going hunting: the theoretical is only worth as much as the costing card feeding it. If the cards are light, part of the "leak" is arithmetic and not theft.
SUPUESTO: the food cost target used here is not an industry yardstick, and no single yardstick fits everyone. It is the 31.8 % that Sazón de Origen's own recipes state, and the 35.0 % it is compared against comes from its own inventories. Source: the sales, food cost and goal the house declares in its context card. If the house agreed to move menu prices, or if the goal were different, the target changes and this whole table gets recalculated.
Illustrative example generated with a fictional restaurant. Not a promise of results, not a client case.
This is 2 of 3 parts. Behind the wall are **the three leaks by name**: which one takes the largest share of the 5,696,000, what evidence confirms each, the 15-day test with owner and date, and how to rule out that the gap comes from the costing card rather than from a leak.
See the full example — free accountWorks with these AIs
How to use it
Which of your data it uses
This assistant works with 9 of the 23 fields in «My Restaurant», among them Business type, Cuisine or concept and Active sales channels. It does not ask out of curiosity: these are what make the answer speak about YOUR scale instead of an industry average — the same calculation on a business of another size returns a number you cannot decide with. You fill them once and they apply across the library, so the second assistant you open already starts with them in place. Whatever you leave blank, the assistant asks for one thing at a time instead of inventing it, and says so before answering, so you know what it is working from.
Always up to date
Every assistant carries a visible version and date. When an AI changes how it works, the assistant's version goes up and you see what changed.
v1.2.12026-08-26The body stops contradicting the answer ladder. It used to say «do not move on until it arrives», the very order the ladder was born to replace: the assistant asked the owner for what the owner does not have and stalled. It now says what the house doctrine says — ask one at a time, and if they do not know, continue with a marked assumption. FORMAT OF THE RESULT was not touched, so the example still corresponds and its seal moves up with the card.v1.2.02026-08-20Grounding: the assistant stops costing in the abstract. The task now works against this house's monthly sales and declared food cost, costs the star dishes instead of a generic plate, ranks causes by cuisine type and sales channels, and drops any test that does not move the main goal; money at stake reads in covers of the average check. Duel still 100.0; from zero client fields in the working sections to seven.v1.1.12026-08-14Spelling: Spanish accents restored in the prose. No field marker, link or meaning changed.v1.1.02026-08-13Rewritten under the margin mandate: 6 numbered steps, 5 named output blocks, 8 business fields, named cautions, verification checklist and figures with units. Duel 40.0 -> 100.v1.0.02026-08-11The assistant is born.Same method, another variant
Inside the app this is a single control: you press it and the assistant reframes itself, including the variants that don't appear here because they don't have their own page yet.
More assistants in this category
Computes your full labor cost with employer charges and overtime, translates it into productivity per shift an…
N3 · Analysis with a verdictMeasures your bar cost split by family, bottled drinks, beer, wine, cocktails and coffee, shows which family l…
N3 · Analysis with a verdictRebuilds each dish's real cost at today's purchase prices, computes cash margin and price floor per item, and …
Included in the always-growing library
Access to every published assistant, adapted to your AI and personalised with your restaurant's data, updates included.