Recommendation Model

recommendation-model-image


  Case Study

The client, a fortune 500 financial services client, wanted to identify cross-sell opportunities for existing clients and direct sales reps on products to pitch for all clients.


  Solution

  • Identify variables used to describe accounts and segment them based on past purchases
  • Developed algorithms to identify similar accounts and similar items
  • Use supervised learning methods to formulate customer likelihood to purchase each item using information from similar accounts and items
  • Create a simple tool that helps reps and managers identify top recommendations for each account that captures ~90% of cross sell possibilities


  Value Addition

Identify top 10 items that are likely to be purchased by each account from a universe of more than 500 items with high accuracy

Interactive tool to help reps and managers customize product pitches for each account

Calculate scenario based cross sell opportunity for each account

Benchmark account performance amongst other similar accounts

 

→ Related Service Offering: Private Equity Analytics as a Service