Once you upload or sync your orders, Metrixon automatically analyzes them and generates high-converting product bundle suggestions using AI and rule-based scoring models.
All suggestions are listed in the Bundles section of your dashboard, ranked by their Total Score.
Each row in the bundle table represents a unique product combination your customers frequently buy together, with AI-generated insights to help you decide which bundles are worth promoting to increase your AOV.
Column | Description |
---|---|
Products | The products included in the bundle |
Scores | A breakdown of score components like:Â FBT (Frequently Bought Together) count, Revenue impact, etc. |
Total Score | A final score (normalized to a range) that ranks the bundle overall. Higher scores = stronger recommendation. |
Discount | A suggested promotion strategy based on product margins and bundle strength. Example format: Type: Percentage Amount: 15% Confidence: Very High |
Reasons | Clear explanations of why this bundle was suggested. These help you understand buyer behavior. Example: • Frequently bought together • Combined price is above AOV • Both are top sellers |
Each time you import your orders, Metrixon analyzes your data using a mix of AI models, machine learning algorithms, and rule-based logic to identify high-potential product combinations. The system looks at how frequently products appear together in the same order, how much they contribute to larger order values (AOV), and how widely those combinations appear across different customers. The goal is to find bundles that not only make sense statistically but also align with how your customers actually shop.
On top of that, we apply business-safe rules to ensure suggestions are practical and profitable. This includes checking discount profitability, evaluating product margins, and filtering out bundles that are too rare or too risky. The result is a ranked list of product bundles — scored and sorted — that gives you confidence to take action without needing to guess or experiment manually.
👉 Continue to: