Retail financial institutions looking to take advantage of big data techniques for social media monitoring should focus on clear business benefits — such as improving sales and marketing efforts, customer service and reducing churn — and architect solutions designed to achieve those goals. Ultimately, the aim is to marry traditional financial metrics — determining factors such as the profitability and credit worthiness of a particular client — with insights into individual customer behaviour and preferences, which can influence their propensity to buy certain products or the risk of churn.

Stay current on your favourite topics


Ideally, teams charged with social media monitoring should operate in an agile development mode, being able to spin up new environments and test out ideas quickly, so as to speed up the learning curve with regards to new sources of unstructured data from social media and other channels. Those data sets can then be used to supplement existing customer analytics to build richer customer profiles. With a more accurate profile of each customer, obtained by analysing richer data from a variety of sources, financial institutions will be able to better guide and personalise their customer interactions.

This paper provides:

  • An overview of the key data analysis techniques used to process and interpret unstructured data from social media channels
  • Some example use cases and an overview of the big data tools available to facilitate social media monitoring, including:
    • Tools for large dataset storage and analysis, including Hadoop and Amazon’s Elastic MapReduce; Apache Hive; Apache Pig and Apache Mahout
    • Natural language processing tools, including Apache Lucene, Apache OpenNLP and Stanford NLP, which each provide unique capabilities to pre-process, model and analyse natural language content
    • Big data visualisation tools, including Carrot2 Workbench; Gephi, NodeXL, yEd, GraphWiz, TikZ and CartoDB
  • Reference stacks to support specific use cases
  • Conclusion outlining a maturity curve for using Big Data techniques

Download this whitepaper now.

Do you want to know more about our capabilities in this space?

The author

Ian Tivey

Ian Tivey

Associate Partner, New York

Ian has a broad background across DevOps and Infrastructure disciplines in the design, build and operation of globally-distributed market data distribution and trading platforms. He currently leads Citihub Consulting’s Cloud Practice, having worked with clients in Europe, Asia, and North America to design and build hybrid cloud solutions in highly regulated banking environments.