Complete AI Training
Sign inGet my AI kit

Your job's AI kit

Get your AI kit

Tell us who you are and what you do. We show you your kit right away and email you the link: skills, prompts, AI agents, MCP servers and courses for your job.

500+ jobs ready, and we make a kit for any other job. No payment needed to look.

Share

AI agent for chefs

Menu Engineering Sales Mix Agent

Menu decisions are based on popularity, margin and prep time for each dish, with the effect of changes tested.

DecisionsAnalysisOperationsHospitality and eventsManagement
See how it worksGet access
Menu Engineering Sales Mix Agent: what goes in, what the agent does and what you get

What it does

A chef keeps a dish because guests love it, not noticing that it loses money, while a quiet item makes the profit. Each month this agent combines sales counts, plate cost and prep time for every dish. It classes each dish as a star, plowhorse, puzzle or dog, based on popularity and margin. It tests the effect of proposed price or placement changes against past data, for example how sales moved last time a price rose. Dishes with data gaps, such as missing recipe costs, are listed and held back from conclusions until filled. It suggests menu changes with the expected effect. The chef approves menu changes. Edge case: a seasonal dish has only 3 weeks of sales, so the agent marks it as too early.

How it works

Follow the arrows from top to bottom. The orange dashed arrow is the loop: when a check fails, the agent goes back and tries again.

Start and resultWhat it doesA check on its own workWaits for your OKGoes back and retries
Yes, continueYes, continueApprovedNoNo 1 STARTS WHEN Monthly menu review 2 USES A TOOL Pull sales counts, plate costs and prep times 3 CHECKS THE RESULT Does every dish have a current cost and enough saleshistory? If not: list the gaps, request updated recipe costs andhold those dishes out. Back to step 2. 4 DOES Calculate margin and popularity for each dish 5 DOES Class each dish as star, plowhorse, puzzle or dog 6 DOES Draft changes such as price, placement or a recipechange 7 USES A TOOL Test each change against past sales after similarchanges 8 CHECKS THE RESULT Does the test show a net gain in profit without alarge drop in sales? If not: adjust the proposal or drop the change andretest. Back to step 6. 9 DOES Write the recommendations with the expected effect 10 YOU APPROVE Chef approves menu changes 11 RESULT Menu report filed
Read the steps as a list
  1. Monthly menu review
  2. Pull sales counts, plate costs and prep times
  3. Does every dish have a current cost and enough sales history?If not: list the gaps, request updated recipe costs and hold those dishes out. Back to step 2.
  4. Calculate margin and popularity for each dish
  5. Class each dish as star, plowhorse, puzzle or dog
  6. Draft changes such as price, placement or a recipe change
  7. Test each change against past sales after similar changes
  8. Does the test show a net gain in profit without a large drop in sales?If not: adjust the proposal or drop the change and retest. Back to step 6.
  9. Write the recommendations with the expected effect
  10. Chef approves menu changesThe agent waits here for your OK.
  11. Menu report filed

How it decides

It classes a dish by comparing its popularity and margin to the menu average, and trusts a result only with at least 8 weeks of sales and a current plate cost.

  • Require 8 weeks of sales to class a dish.
  • Class popularity against the menu average mix.
  • Reject a price change if past data shows sales falling over 15 percent.
  • Count prep time as a cost for slow dishes.

Make it yours

Every agent is a starting point. You choose these settings for your own situation.

  • Minimum sales history (default 8 weeks)
  • Sales drop limit for changes
  • Dish classes and cutoffs
  • Review frequency

What keeps you in control

It always asks you first

  • Chef approves menu changes

Hard limits

  • Do not recommend changes from dishes with missing data
  • Show the data behind each class

It stops when

  • Done: recommendations approved.
  • Stop: more than 25 percent of dishes lack cost data.

Set it up

We guide you through the set-up, step by step

Members get the full set-up guide for this agent. No technical skills needed: you copy, paste and upload.

10 minto set it up in your AI
5 AIsChatGPT, Claude, Copilot, Gemini, Grok
  • One set of instructions to paste into your AI, with the clicks for ChatGPT, Claude, Microsoft 365 Copilot, Gemini and Grok
  • The agent then walks you through connecting your own data, one source at a time
  • A downloadable copy with the flow chart, the rules and the full guide
Get access to this agent

An example run

What happensThe October review covered 34 dishes. Three had old costs, so the data check failed. After the chef updated them, the agent classed a short rib as a plowhorse: 1,100 sold at a 22 percent margin. It tested a 1.50 dollar increase against last spring's change, when sales fell 6 percent, and projected a profit gain of 900 dollars a month. The chef approved the change.

More agents for chefs

Risk management · Healthcare

Food Label and Allergen Verification Agent

Every menu item and product is checked for allergens and substitutes are verified.

10 steps1 check3 jobs
Decisions · Automation

Final Headcount Lock and Order Adjustment Agent

Have orders, staffing and rentals match the final headcount without waste or shortages.

11 steps2 checks3 jobs
Creating · Decisions

Tasting Feedback to Menu Revision Agent

Turn tasting comments into a revised menu that is costed, safe and doable at event size.

11 steps1 check2 jobs
Compliance · Monitoring

Food Handler Certification Expiry Agent

Keep every employee's required certificates valid and filed before any inspection.

11 steps2 checks5 jobs