Articles in this section

Lesson 28: Recommendations

Machine learning recommendations in Incentivio help increase order revenue by delivering smarter, more personalized item suggestions to guests. Our new recommendations model intelligently analyzes what a guest has in their cart, past purchase behavior, and recent ordering trends to recommend complementary items like drinks, sides, or desserts, at the perfect moment. By surfacing relevant add-ons in real time, restaurants can drive higher average check values while creating a more seamless and tailored ordering experience. This guide walks through how to analyze your recommendation's performance.

To manage recommendations, please see Dashboard - Managing Recommendations 

 

Summary Metrics

The Summary section provides a quick snapshot of how recommendations are performing over the selected date range.

% of Orders that Displayed a Recommendation
Shows how often recommendations appeared during checkout. A higher percentage means recommendations are consistently being surfaced to guests.

Orders with Purchased Recommendation
The number of orders where at least one recommended item was added to the cart. This helps measure guest engagement with recommendations.

% of Transactions where a Recommendation was Purchased
Indicates conversion rate: how often recommendations led to an actual purchase.

% of Revenue Lift for Orders with a Purchased Recommendation
Shows the average increase in order value when guests purchase a recommended item compared to orders without one.

Total Revenue Added through Recommendations
Revenue generated directly from recommended items being purchased.

Revenue Added after Showing Recommendations
Represents additional revenue associated with orders where recommendations were shown, helping illustrate overall impact beyond direct item sales.

Detailed Metrics

Filters & Controls

Use the filters to refine what you’re viewing:

  • Date Range – Adjust the timeframe to analyze performance trends.

  • Locations – View results across all locations or focus on specific stores.

  • Refresh – Update the report with the latest available data.

Top Recommended Items

This section highlights the items most frequently purchased through recommendations.

Each item card shows:

  • Quantity sold through recommendations

  • Revenue generated

  • Relative performance compared to other items

Use this to identify:

  • Strong upsell performers

  • Items that pair well with your core menu

  • Opportunities to adjust overrides or categories

Average Basket Value Chart

The chart compares average order values across different scenarios:

  • Orders where recommendations were purchased

  • Orders where recommendations were shown but not purchased

  • Orders with no recommendations displayed

This helps you understand the true impact recommendations have on guest spend behavior.

Top 15 Recommended Items Table

This table provides a deeper breakdown of recommendation performance.

Item Name
The recommended menu item.

Quantity Sold
How many times the item was purchased through recommendations.

Orders
Number of unique orders containing the item as a recommendation.

Revenue
Total revenue generated from recommended purchases.

Take Rate
The percentage of recommendation impressions that resulted in a purchase.
Higher take rates often indicate strong item pairings or effective recommendation placement.

How to Use This Report

Use this report to:

  • Identify which items drive the most incremental revenue

  • Spot low-performing recommendations that may need adjustments

  • Evaluate the effectiveness of custom overrides or category rules

  • Understand how thew new intelligent recommendations model is influencing order behavior and basket size

Regularly reviewing this report helps ensure your recommendations stay relevant, balanced across categories, and aligned with your business goals.

Next lesson - Lesson 29: Loyalty Offers

Was this article helpful?
0 out of 0 found this helpful