Harnessing Data & Eliminating Bias

The following examples highlight how critical it can be to use data to either validate or disprove previously held retail / customer assumptions.

Northern Italy

The data analysed is for a large Italian grocery retailer who operates 139 stores mainly concentrated in the Northern part of the country.

Two categories were reviewed where assumptions about customer shopping patterns existed :

  • Pet Food – Stores located in more urban areas will have greater strength in cat food compared to dog (due to less living space, more flats versus houses, less homes with gardens etc)
  • Sparkling Wine – Stores in more affluent areas will buy more expensive wines ie. champagne

 

Assumption

Stores located in more urban areas will have greater strength in cat food compared to dog (due to less living space, more flats versus houses, less homes with gardens etc)

CaseStudy: Italian Grocery Retailer, Store Clustering
CaseStudy: Italian Grocery Retailer, Store Clustering

 


Sparkling Wine

Assumption

Stores in more affluent areas will buy more expensive wines ie. champagne

CaseStudy: Italian Grocery Retailer, Store Clustering
CaseStudy: Italian Grocery Retailer, Store Clustering

Conclusion

Commercial teams will have an hypothesis about how customers shop their categories. These beliefs can be based on retail feedback, supplier input or a variety of data sources.

The RETAILIGENCE clustering tool generates insight which will help to either confirm or contradict existing category understanding. The tool makes no assumptions about the attributes that may be impacting shopper behaviour but identifies similar stores purely from analysis of shopping patterns.

Bias is eliminated !