Data Mining

Data mining is the process of automatically analysing databases with the aim of extracting valuable information.

Nowadays, in any organisation, significant amounts of data are collected. With the application of data mining tools, it is possible to obtain significant competitive advantages using the information hidden in the raw data.

Through a thoughtful understanding of advanced statistics, information theory and mathematical modelling, we apply data mining tools to solve a wide range of practical problems.

For instance, a clustering algorithm can split the set of customers, based on demographical and behavioural attributes, into lightly overlapping groups. Clustering allows therefore the development of a predictive model that labels a new instance (in our case, a new customer) as a member of a group of similar records (a cluster). Cluster dynamics can be then analysed via Markov Chains (or Transition Matrices).

 Once we have a “sales history” database, we can apply a second technique, affinity modelling, to predict which products and services sell best together.  

Decision Trees or Neural Networks can help identifying  "churning" customers, or customers that are going to leave. Their  impact on the bottom line can be estimated with CLTV methodology, and actions to maintain them go under the "umbrella" of "customer retention".