skip to Main Content
Strategy For Applying Analytics To A Demand Signal Repository

The Opportunity

  • A consumer packaged goods company had invested heavily in building a so-called demand signal repository (DSR) – a collection of retail point-of-sale data from most of their customers and syndicated data from their marketplace data supplier.
  • While the company had begun to use the DSR for basic status reporting, they felt the value of their investment would not be realized until they started to apply advanced analytics to the data.

Our Approach

  • We were engaged to assess the data and suggest top use cases. We evaluated the data in terms of suitability for purpose, and the ability to build analytical models with the data. We strived to integrate with current company process around sales, marketing and customer relationship management.

The Impact

While we identified many possible cases to recommend, four cases rose to the top:

  • The identification and profiling of out-of-stock and lost sales conditions. While this is the original intent in the design of all demand signal repositories, without the advanced analytics applied to the data, this functionality had been limited.
  • A “Predictive Executive Overview.” The company had already developed an executive overview report. We recommended that forecasting methods be applied to turn this report into a forward-looking dashboard.
  • Store clustering based on shopper behavior, rather than trade area and store attributes. We built store-level price elasticity models to show that stores could be treated differently based on the response of shoppers in those stores.
  • Scoring the potential value of adding additional product to existing assortments to improve their retail customers’ performance.
Back To Top