This week I was fortunate to attend the 2nd Big Data & Analytics Conference, organised by IQPC and PEX Network in London. The following are my take-aways from what appears to be a relatively confused landscape of business intelligence professionals, data architects, technology providers, and system integrators:
1. Whole market solutions – Financial Services institutions are looking more to data to create ‘whole banking’ and ‘whole market’ solutions, away from historically silo-ed asset class or product-oriented data stores and applications – this suggests there is a strong case for big data solutions that allow business and IT to manage and analyse transactional data across the enterprise.
2. Resistance to public Cloud – An increase in analytical capacity, given the huge amounts of data that can now be brought into a big data solution, does not lead to bringing FS data into public clouds, like Amazon Web Services, any time soon. Bigger institutions will initially look at more effective sharing of capacity in their own data centres and might look at private analytics cloud as an alternative. Concerns around having sensitive customer and transaction data hosted on the cloud has not stopped smaller institutions from doing just that. It may take up to 5 years for bigger FSI’s to follow their example, despite the economic argument.
3. Retail and wealth management leading – Most use cases for exploiting big data and advanced analytics appear to be on the retail and wealth management side, less so in corporate and investment banking. However, a strong case was made by Bank of America to develop use cases for investment banking around real-time trade executions and post-trade analysis, extending the time horizon, looking for trends and patterns that inform the health of the portfolio, pricing competitiveness and product profitability.
4. Compliance gets the IT resources – Compliance seems to get in the way when it comes down to making business decisions about investments in analytics and big data – this is due to IT departments being tied up in risk and compliance work and the lack of use cases and clarity around ROI for big data projects. Compliance could be your ally if you decide to use a common analytical platform and tools that can be used to satisfy both the regulator and your heads of business.
5. Sophisticated labeling of data – The ultimate goal of getting closer to your customer and responding ever better and faster to their demands and market events, requires better insight into what individual customers want. This is less about segmentation, more about sophisticated labeling based on context (what clients do, what/how/when/where/why they transact). This again requires an advanced analytical capability that does not exist in most FSI’s.
6. Telling the story in data visualizations – Visualisation is a powerful enabler for advanced analytics and needs to be created as closely to the end-user as possible, for it to ‘tell the right story’. The way the data is prepared and delivered to the visualisation team needs to be flexible and configurable to give the end-user actionable data at their fingertips.
7. Role of the data science team – The role of the analytical teams (data scientists, quants, etc) was discussed as well. It is a competence/skill set that needs to be well organised, resourced, nurtured and protected from ineffective deployment and becoming a single point of failure in the organisation.
Many delegates keep an open mind to what big data will bring to their industry, but they clearly need better use cases and more convincing insights in order to fully appreciate the benefits of these new technologies.