Who is this assistant for?

The month's food cost jumped two points and the costing cards still say what they said six months ago. You know the dish went up, but not whether it was the supplier or somebody changing the portion. The dish cost barely moved and you suspect plenty happened underneath. Six dishes went up at once and you cannot see why, since you buy different things for each.

A card's cost freezes the day it is written and the owner finds out it moved once the month's food cost has already jumped. This assistant sets two cuts of the same dish side by side and explains it peso by peso: how much was the supplier, how much was yield, how much was a portion somebody changed without telling anyone, and how much was one sub-recipe dragging half the menu.

AI assistant · N5 Complete system

◆ Included in the annual plan👨‍🍳 Standard Recipes, Spec Sheets and Productionv1.0.0 · 2026-08-26

📊 Works best with an AI that reads Excel or Sheets (Claude for Excel, ChatGPT, Copilot in Microsoft 365)

Recipe Cost Variance Analyzer for Restaurants

Compares the same recipe costed on two dates, splits every difference into price effect and quantity effect, and puts a cause and an owner on each movement, including the ones covering for each other.

Depth

9/23

Your data

Compatible AIs

Steps to use it

Depth of the resulttap a level to see what you get

◀ Put out today's fireRun the year ▶

🏛️

N5Complete system

Several chained pieces you install and operate over time, with their review cadence. It changes how the business runs, not one decision.

◆ This assistant 110 of the 164 published assistants are at this level

Is this for you?

What it does and what you get

A card's cost freezes the day it is written and the owner finds out it moved once the month's food cost has already jumped.
This assistant sets two cuts of the same dish side by side and explains it peso by peso: how much was the supplier, how much was yield, how much was a portion somebody changed without telling anyone, and how much was one sub-recipe dragging half the menu.

What you end up with

The comparison card with both dates and the threshold, the line-by-line variance table with price and quantity effects apart, the cause matrix with an owner, the sub-recipe drag table and the cut-off checklist with the next date.

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

Sazón de OrigenFictional contemporary Colombian restaurant in Medellín, 2 locations, 6 years in business. Every sample result in the library comes from this same case. Fictional restaurant

The owner asked this: «My sancocho cost 15,900 in March and now it comes out at 16,400. That is 500 pesos, almost nothing, but the month's food cost moved more than that explains. I want to know what happened underneath.»

Same recipe and same declared portion weight at the main Medellín location, two cuts: March 12 and August 12. Materiality threshold declared by the house before looking: 150 COP per line.

Line	March cost	August cost	Variance	Price effect	Quantity effect	Cause
Beef rib	2,640	3,480	+840	0	+840	Measured yield
Three-meat stock	2,905	3,400	+495	+495	0	Sub-recipe
Chicken leg	1,380	1,540	+160	+160	0	Supplier price
Avocado	1,596	1,330	−266	−266	0	Seasonal price
Corn on the cob	1,200	880	−320	−320	0	Supplier substitution
Pork loin	1,850	1,580	−270	0	−270	Portion weight
Remaining lines	4,329	4,190	−139	−139	0	Movements below threshold
Portion total	15,900	16,400	+500

Net variance 500 COP. Gross variance 2,490 COP. Five times more movement than the total let you see. Illustrative figures for this case.

Grouped by cause, with an owner. Across 880 sancochos a month:

Cause	Effect on the dish	Effect for the month	Who answers	Reading
Measured yield	+840	+739,200	Kitchen	Not an increase: the card used to cost at purchase weight
Sub-recipe (stock)	+495	+435,600	Purchasing	The meaty bone rose, and it drags 5 other dishes
Supplier price	−245	−215,600	Purchasing	Seasonal avocado and minor adjustments
Supplier substitution	−320	−281,600	Purchasing	Corn from another supplier, and quality needs testing
Portion weight	−270	−237,600	Kitchen	Pork portion trimmed from 100 g to 85 g without authorization

Two uncomfortable conclusions. First: the +840 is not inflation, it is the old card lying; that money had been leaving for months with no line to show it. Second, and this is tomorrow's job: the −270 is a cost drop nobody should celebrate. Somebody trimmed 15 g of pork off the dish most ordered on Sundays, and the guest who comes back does notice, however much better the paper looks.

And a concession: the corn dropped 320 COP through a supplier change. Before taking that saving as good the product needs two weeks of testing, because a purchase saving that lowers quality comes back through the till.

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 sits what turns one comparison into a watch system: the drag table —which other dishes each sub-recipe moves and what that adds up to for the month— and the cut-off checklist with an owner, the next cut date and the indicator that really matters, the distance between gross variance and net variance. As long as that distance grows, the dish total is hiding work.

See the full example — free account

Works with these AIs

Claude — the best for this assistant The best one here: it keeps price effect and quantity effect in separate columns and checks that they add up to the line's variance, instead of handing you a delta and a tidy story. Ask for gross variance before net, or it will give you the reassuring total first.

How to use it

Fill in My Restaurant: kitchen, city, number of locations, star dishes, price range, channels, monthly sales and declared food cost.

Gather two dated costings of the same dish, 90 days apart or more, with price and quantity per line.

Paste the assistant into your everyday AI and answer its questions one at a time, without running ahead.

Declare your materiality threshold before looking at the data, so you do not investigate noise.

Read gross variance beside net variance: that distance is what the total was hiding.

Which of your data it uses

of 23

This assistant works with 9 of the 23 fields in «My Restaurant», among them Cuisine or concept, City and country and Number of locations. 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.

Cuisine or conceptCity and countryNumber of locationsBest-selling dishesMenu price rangeActive sales channelsAverage monthly salesApproximate food cost (%)Main goal this year