Discount recommendation (Under the hood)

If your company is using or wants to use a variable discounting structure for your customers, we can help you to established or optimize it. We developed a discount recommendation algorithm that will recommend the right discount for your specific customer and product which he/she would like to purchase. The algorithm will take your customer’s purchase history, customer behavioral data and list of products based on which minimal, optimal and maximal discounts will be offered. The algorithm is trying to tract customers with slightly higher discounts and over time reduce them as more regular purchases will be made.

Configure the algorithm based on your needs

Discounting policies can be complex; therefore you are able to configure:

Per product & per product category minimal margin
Per customer & customer group minimal margin
Set some customers as VIP

This configuration enables you to start with a simple approach and make it more granular over time.

Data Sources

Our Discount recommendation algorithm is taking these 3 input files:

Sales history: History of purchases done by each customer
Customer demographics data: You can capture multiple demographics information for each consumer
Product data: Information about products

Machine learning in the background

To estimate the right discount for the right customer and product, we build and trained behavioral model based on our data. This model is already distributed and available as part of iERP studio platform.

In case your business needs to have a different approach for discounts over time, the model can be retrained.

For you, AI enthusiasts, we are using the CNN network to categorize customers into groups and dense layers to train and forecast results.