Efficient Allocation of USD 25 MM/year Capex through Spatial Data Analytics for Optimal Store Site Selection

Case Study

Efficient Allocation of USD 25 MM/year Capex through Spatial Data Analytics for Optimal Store Site Selection

Business Objective

Our client is a specialty retail business in the US. They wanted to leverage advanced analytics to identify optimal locations for setting up new stores. The high-level objective was to pick locations with high consumer spends in the category of focus and adjust potential sales for cannibalization effects due to existing stores and competition. The real estate team and executive management also wanted to understand key influencing factors, both internal and external, that impact store sales.

Challenges

  • The focus was pan-US, but data was sparse in certain states/regions
  • Information about competitors was limited to location, company/brand name – posing challenges in demand allocation
  • Integrating multiple internal and external data sources at various granularities

Solution Methodology 

  • Two sets of models were developed
  • Market Serviceable Area (MSA) for a store was defined based on the population density in the region
  • Identified and collated 2000+ features from external data sources comprising housing/construction, macro-economic indicators, socio-economic data, weather and other indicators that could explain the demand for plumbing products
    • A spatial model that apportions demand to own stores and competitor stores based on their attractiveness (location, competitor density, customer preference)
    • A regression model that identifies the key drivers and quantifies their impact on store sales
  • Finally, the two models were integrated to obtain:
    • Sales estimates for a prospective location
    • Incremental sales after considering cannibalization
    • Influence of competitors on sales
    • Contribution from key sales drivers including various external factors

Business Impact

 

  • Identified multiple high-value location alternatives, representing an average incremental of USD 5 – 7.5 million per location compared to the base plan
  • Replaced the current heuristics-based approach with a data-driven approach
  • Built a geographical visualization application to identify future optimal store locations at state and city level
Download Case Study

©2020 Tiger Analytics. All rights reserved.

Log in with your credentials

Forgot your details?