Advantage
Better decision on future branch locations
Data
Characteristics and KPIs of the branches, characteristics of the branch environment
Methods
Random Forest models, Bayes approaches
Challenge
A retailer is concerned with the question of which factors are responsible for the success of its stores. Determining the potential success factors is complex. In addition to the heterogeneous characteristics of the stores, the individual circumstances of the location and its surroundings are also important.
Goal
Against the backdrop of further expansion, it is important to find out which factors or combinations of factors significantly influence the success of a branch in order to be able to make location decisions on this basis.
Solution
eoda supported the client from data preparation through model development to deriving recommendations for action. The following information was incorporated as variables:
- Store characteristics (store space, number of parking spaces, equipment, etc.)
- Store KPIs (sales, average receipt, number of receipts, etc.)
- Store environmental characteristics (room type, rent index, distance to competitor stores, etc.)
A particular methodological challenge of this project: the relatively small number of cases faced many variables. To address the relatively small number of cases, eoda conducted resampling and used random forest models and Bayesian approaches. The latter are particularly robust in the area of small data.
Result
Based on the analysis conducted by eoda, it is now possible to empirically assess which locations are promising and how new stores should be equipped. This provides retailers with a reliable decision-making and knowledge base for location planning and the evaluation of existing stores.
Get started now:
We look forward to exchanging ideas with you.
Your expert on Data-Science-Projects:
Lutz Mastmayer
projects@eoda.de
Tel. +49 561 87948-370
