Skip to content

The Opportunity

  • A business-to-business company with many high volume, high-quantity orders from customers was concerned about declining sales for a sizable number of accounts. They wanted to use their invoicing data to diagnose the causes and to seek remedies.

Our Approach

  • Our analysis of the invoice data indicated that the issue was not with order quantities, but rather, with timing;  many customers would slow down or delay their ordering patterns.​
  • We developed a model of the purchase timing cadence of each customer based on past sales patterns.  The model produced a “likelihood score” of the odds that a purchase should occur each day.  As days would pass without a purchase, eventually a probability threshold would be crossed, and the model would produce an alert that a purchase should have been expected by that time.​

The Impact

  • The alerting system was able to reveal potential lost or deferred sales situations.  Though the sales force has regular contact with a handful key accounts, the middle tier of customers, large in number, that are responsible for a significant volume of sales, are harder to monitor manually.  This system alerted the sales executives where to direct their calls to help increase the buying rates of those customers.​
Back To Top