Case Study: Cannibalization & Affinity - Prescriptive Analytics
Background: Client houses multiple products. Sales of some products are impacted due to certain changes in other products. For example - Introduction of newer version of a product hurts the sales of older version (cannibalization). Decreasing the price of substitute product hurts the sales (cannibalization). Decreasing the price of a main product results in the increased sales of attachment products(complimentary). The objective of this project was 2 fold. Firstly identify substitutes & attachments products. Secondly understand the impact of price change on substitute & attachments products such that the price models can be built at a category or sub category level.
Approach: Sophisticated statistical techniques like Cross Price Elasticity, Attribute based product classification techniques were used to identify substitutes and attachments. Historical data points were studied to measure the impact on price change on substitutes & attachments. Category level price models were built to measure overall impact on the category due to a price change on an individual product.
Impact: Client is able to take better pricing decisions of products by understanding the impacts on substitutes & attachments. Client also understood that the pricing decisions need to be measured more at a category level rather than at a product level.