We solved this problem with the following approaches:

  • Designed and developed a web portal for the users to upload all the transactional data along with their product catalog.  These data elements were captured in a relational database repository.
  • A unified product hierarchy was created to maintain consistent product categories across network. Machine learning techniques were used to classify the products in to product categories.
  • Applied advanced analytics to identify the most favorable vendor for each product category.
  • Reconciliation process of matching the transaction data from lending and buying group was carried out to compute rebate for each buying group member according to business rules set by the lending group.


  • A transparent one stop solution to identify the rebate overdue for all buying members
  • Actionable insights into missed rebate opportunity for buying group along with recommended favorable vendor


  • Our solution reduced the recurring manual efforts in computing the rebates for each buying group member
  • The rebate mismatch and sales mismatch reported by various buying and lending groups were resolved by reconciling the data elements from both parties
  • An overview of all the sales activity and rebate opportunity across the network is available to understand buying patterns across network