Download case study for the largest discount retailer in the Nordics
Store Network Optimizer showcases relevant nearby competitor presence as well as detailed population demographics within the estimated catchment area.
It generates credible sales forecasts for each map square. This information is to be used for evaluation site profitability calculations.
Which stores are performing up to their full potential? Where and in which categories performance can be improved, and what demographic or competitor presence explain store performance variations?
Comparing budgeted sales to actual sales leaves one key question unanswered: what is the sales potential and how well is that met?
Current network performance analyses uses forecasted sales as the baseline which actual sales is compared with. Total sales forecast is broken down on a category level to indicate over- and underperforming categories giving actionable insights on category allocation – and performance enhancements.
"Together with Houston Analytics we were able to model our 2020 store location network. Our co-operation was very smooth and the knowledge we gained thanks to analytics will be very useful for IPO and everyday decision-making."
Director, Business Development, Tokmanni Group Oyj
Store Network Optimizer leverages data from different data sources. The data typically includes existing service/store network area and offering ( customer data ), customer database (CRM), offering data (products/service), historical sales data, and Grid database for a selected market.
A mix of commercially available and open source technologies has been used. IBM Decision Optimization Studio (CPLEX) has been used for optimizations and IBM SPSS Modeler and IBM SPSS Collaboration & Deployment Services for predictive analytics. As the data warehouse IBM DB2 Enterprise Server has been used. Alternatively, Oracle Database or Microsoft Database can be used.