Success story of Retail Analytics and Forecasting Center of Excellence

 Retsci has helped its customers establish a forecasting center of excellence through its senior data science leadership.  This entails helping set up forecasting models, tuning existing models and establishing best practices in their forecast management process.Retsci partners with its customers to evaluate and improve their forecasting solution.  We have worked with different customers on different platforms to help them improve their forecasting solutions.
 
  • At Scheels, Retsci provided senior forecasting management support to work with them as they had built their in-house forecasting solution.  The end result is a forecasting solution to support their inventory replenishment for their stores.  The Retsci experts worked with the Scheels data science team to:
 
    • Architect a forecasting model to support the business strategies at Scheels
    • Develop best practices for their forecast management process
    • Setup scorecarding and accuracy reporting to be able to assess improvements over time
    • Root cause analysis of forecasting issues and enhancement suggestions
 
  • At Lululemon, Retsci also provided senior forecast management support.  Lululemon has in house forecasting models that were causing issues with certain stores.  Retsci was able to dig into the algorithms and the code to find the source of the problem and provide tuning recommendations to solve the problem.  After changing the configurable parameters the forecasting issues disappeared.  Next, we continued to work with Lululemon to enhance their models.  This involved:
 
    • Learning and documenting the models used in their forecasting solution
    • Tuning assessments based on various configuration changes to provide the best setup
    • Suggestions on where configurable parameters should be updated to provide more strategic settings based on business areas
    • Development of Forecast accuracy scorecard and root cause investigation tooling
    • Development of framework to test and assess the model changes to determine what should be kept or what isn’t improving the overall system.