The world is changing. One of the most important trends is that things are beginning to run on data, and we’re getting more and more data as a result of this. We have a data-rich society, where we can clearly see patterns of behaviour and habits on humans (or customers). Businesses commonly refer to these type of analysis as “Big Data”, as we can now collect very rich (e.g. speech, text and images) and very dense (i.e. frequent) information. Today we have the means to organize all this data, store it and extrapolate valuable information from it.
Through the implementation of knowledge graphs, we were able to connect and link as a graph detailed customer data from a number of sources: transactional customer data (what they buy and where they spend), geolocation (where they are and how they move), financial indicators, browsing & call data.
By using and combining enriched and heterogeneous data at very granular level, available from digital channels, smartphones, mobile apps, marketing platforms, communication channels, it was possible to analyze, explore and predict customer behavior with greater accuracy compared to traditional methods. The richness of the customer information (as well as the surrounding environment data) which could be collected in real time from these multiple channels, was stored in a single analytical environment. It becomes a magnifying lens through which analyst can easily query single customer data as well as higher-level cluster information to detect more general patterns. It is an organized asset of all customer related information at granular and at higher level.
The process of building a comprehensive Customer Insight Graph, requires the use of both Machine Learning algorithms (i.e. Entity Recognition, Speech-to-text, Sentiment Analysis, Image recognition, Pattern Analysis) and Big Data Architecture to store and promptly extract information.