Stay tuned – Receive JSM-news !

Join the JSM mailing list to receive our latest updates.
Email address
All too often companies have only the vaguest idea about what kind of data they’re holding; because such data is very often hidden deeply away in a variety of databases and fragmented across different departments. We identify this data and bring it to light, making it visible, cohesive, comparable and easy to understand so that it really does support YOU in making the right decisions. And if need be, we can also identify any lacking data and define a concept to fill in the gap.

Gauging what works best for a brand using correspondence analysis and hierarchical clustering

Posted by on Nov 3, 2017 in Case Studies | No Comments


Marketers have always had to manage goals such as making their brands distinctive, making them central in their category, and understand competitors. The objective of the study was to understand how brands and a list of attributes/features are perceived by consumers. While understanding why a customer chooses a particular brand repeatedly over time, the aim was to ascertain the degree of differentiation in the market place and identify opportunities for the respective brands.


Correspondence analysis is a graphical display of the rows and columns (here, for example, brands and attributes) for cross tabulation. However, examining a cross-tab table and identifying relationships between rows and columns can be difficult and time consuming.

With the correspondence analysis map, we can arrive at more apparent conclusions and it allows us to quickly identify the association between the two or more categorical variables, but it fails to identify the order. From this, we could determine the list of attributes that contributed to the success of the respective brands.

Therefore, to strengthen the analysis further, we deployed a hybrid technique combining Correspondence Analysis and Hierarchical Clustering which helped us to understand the distance at which the clusters combines indicating the differentiation between them. It determined which attribute had a greater positive impact for the brand in comparison to the others.


The graphical outcome of the above mentioned methods helped the client to gauge the relationship between the brands and their association with various products and service attributes/features as well as the order of their impact. Simply put, three results we achieved were – determining the attributes helpful for a brand’s success, understanding which attributes/list of attributes impacted more than others, and the comparison between the brands themselves.

Based on the analysis, client might change its views on the messages used in communications and on the website, highlighting how its brands are different and better.


A post graduate in Statistics from Osmania University, I have worked on Data Processing and Analytics for over 10 years now. I am always passionate about learning new techniques in analytics and more.

More Posts

Leave a Reply