top of page


Space Plan Solutions
Shopping basket with assortment products



  • Our solution groups stores using a bottom-up approach (user nominated sales mix)

  • It uses the kmeans method to determine which stores are grouped together

  • The solution allows for the creation of several clusters using user nominated product attribute drivers

  • It can identify the most accurate cluster based on how close stores are to the centre of the cluster

  • Users can name clusters, exclude or manually make changes

  • We import data covering store attributes, product attributes and performance

Space Plan Solutions clustering approach

Data Import


Store Results

  • Product drivers and 4 Cluster Scheme

Clustering_ Choose Product Drivers

Choose Product Drivers

Clustering stores in the cluster

Stores in Cluster

Clustering_ Stores List

Store List

Number of Iterations and Scheme Accuracy

Clustering_ Kmeans Iterations
Clustering_ Cluster Accuracy

Cluster Accuracy

Kmeans Iterations

Output Reports

Clustering_ Store Importance by Attribute

Stores Importance by Attribute

Clustering_ Product Importance by type of Attribute

Product Importance by type of Attribute

Clustering_ Demographic Importance by Type

Demographic Importance by Type

Output Reports (Report Writer)

  • Users can create their own reports for Stores or Products

  • An interactive Microsoft Excel workbook is delivered 

Clustering Report
Clustering Report

Project requirements

  • We complete a discovery process to find out your key objectives and goals for the Clustering project

  • To achieve the expected results of an accurate projection, data is key. All data is reviewed to ensure we have clean data to run the project

  • Any project delivery is designed around the available data that can be used. 

  • A full list of the data required to run a project can be found by clicking the data requirements button below

  • Alternatively, please contact us and we will be happy to discuss your requirements further

bottom of page