What sales and cost data do I need before building a menu engineering matrix?

Before building a menu engineering matrix, you need clean item-level sales and cost data for the same time period. In most restaurants, the matrix is only reliable when each menu item has consistent naming, accurate recipe cost, and enough sales volume to compare popularity and profitability fairly.

Core data you should collect first

• Menu item name (standardized, no duplicates under different names)
• Units sold per item in the analysis period
• Selling price per item
• Net item revenue (after discounts/voids if possible)
• Recipe or plate cost per item (current ingredient cost basis)
• Contribution margin per item (selling price minus item cost)

Supporting data that improves matrix quality

• Date range and season context (for example, high season vs low season)
• Branch or outlet identifier if you run multiple locations
• Dine-in vs delivery channel split when mix differs significantly
• Temporary stockout or unavailable-item notes
• Major promotion flags that could distort normal demand

How it is typically prepared in operations

A common workflow is to export item sales from POS, pull recipe costs from purchasing or costing sheets, then merge both datasets by a single item ID or standardized item name. After that, teams calculate average popularity and average contribution margin thresholds to classify items into the usual matrix groups.

For a café, this may mean separating espresso-based drinks from food items first. For a restaurant group, many operators build one matrix per branch and one consolidated matrix to avoid hiding location-specific performance.

Common data issues to fix before analysis

• Same item sold under multiple names
• Outdated recipe costs after supplier price changes
• Bundles or combos not mapped back to component items
• Large one-off events included in a normal period review

Menuviel provides cleaner item consistency for matrix preparation

With Menuviel’s centralized menu management and Single-Point Item Management features, operators can keep item names, structures, and availability states aligned across menus and branches. That makes it easier to match sales exports with costing records and reduces classification errors before building the menu engineering matrix.