Sales Analytics: Maximizing Upselling and Cross-selling Goals, 4 Week Proof of Concept

This 4-week proof of concept aims to explore any one business scenario at your organization and leverage Data Science, as well as Microsoft Data Suite, with the explicit purpose of maximizing your revenue and margin. We propose this proof of concept as the best starting point for evaluating the potential of Analytics in your business.

Market Basket Analysis is an Advanced Analytics project that allows us to identify Association Rules between different products and customers; we use data mining to identify the most frequent combination of transactions, we then calculate a series of robust statistic tests which validate the probability of selling said combination of products. Market Basket Analysis has various commercial use cases such as up-selling and cross-selling recommendations in Points of Sale, promotions building for e-commerce, and merchandise display.

We will implement the solutions by making use of the Azure Suite for Advanced Analytics: • Logs, data files and structured business apps will be ingested and administered using Azure Data Factory. • Data storage will be mounted in the best possible instance in Azure Data Lake Storage. • We will, develop, prep, and train the models in Azure Data Bricks. Modelling may be developed and supported in Python, which will then be plugged into ADB. • Finally results of the model will be reported using Power BI. Any and all steps will be evaluated and adjusted according to the needs of the customer.

Note: The 4-week Proof of Concept is a first stage diagnosis to evaluate the viability of the project. The estimate is based on the use case 2 different data sets with medium complexity. Firm time and price to be provided after further review of the use case and data sources.