Jon Ander Beracoechea — Co-CEO at BBVA Data & Analytics
Jon holds a Ph.D. in Electrical Engineering and has published multiple papers in the field of adaptive signal processing. He has been applying advanced analytics and machine learning techniques in the banking industry for 9 years in áreas such as CCAR stress testing (Moody’s) or derivatives pricing (Banco Santander).
This is the complete interview that was made to Jon Ander by the newspaper Expansión in April 2017.
What do the data represent for BBVA?
The data is a fundamental part of BBVA’s DNA. Banks and the financial world in general is gradually understanding the importance that the data (and their treatment) have in providing the best possible service to our customers. BBVA is a pioneer on this road. The data and its exploitation are present in many facets of the day to day of the bank: in the creation of new products and services, automating processes, helping to make better decisions and, in general, making life easier for our customers. This is a path that has no turning back.
When was the BBVA Data & Analytics division created? What are their goals? How many people work there?
BBVA Data & Analytics was created in May 2014 (although it had been working since 2012) precisely to accelerate this process. Its goal is to transform the bank into a data-based organization. For this, we collaborate with both the Business Areas and the Engineering teams, developing projects and solutions of a very varied casuistry. Although our focus is on providing value in the near future, we also have a vocation of exploration in the longer term where we try to make “science” and transform it into Business. We try to imagine how the bank will be in the future and how we can anticipate these changes. The team is made up of about 50 people and we have offices in Madrid, Barcelona and Mexico City.
What volume of data does the bank manage? How can you extract value from so much information?
The volume of data that we handle easily is placed in the band of petabytes. Now, the real problem is not so much volume as variety. BBVA collects information from many channels (offices, web, mobile applications …) and for many verticals (private clients, SMEs, large companies …). Banking data is extraordinarily rich and varied, but the real challenge is how to tame that informational diversity to bring value to our customers. This implies having the right talent, the necessary technological platforms and an organizational design that facilitates the interactions between the different actors necessary to carry out the different initiatives.
It is estimated that only in the USA there is a potential demand for 4 to 5 million data analyst jobs in the next year. What new profiles is incorporating BBVA? Is it difficult to find and attract this talent?
The type of profile that is being demanded is not easy to find. They are professionals with a very multidisciplinary profile. We move at the intersection between mathematics and software development all mixed with a good deal of knowledge of how banking works. Most of the incorporations come from very technical fields (engineering, mathematics, physics …) but with a pragmatic mentality that seeks to develop concrete and actionable solutions. We put a lot of effort into attracting and retaining this kind of talent. This involves creating a culture that encourages curiosity and the ability to experience; A flexible, dynamic and elbow-to-elbow environment with pairs with which to contrast results. It is an interesting mix between the business environment and the academic.
Can you give me an everyday example of how a customer can benefit from a good big data analysis?
The truth is that we do not like to talk about “Big Data” since we feel it is a relatively empty word of content. We see ourselves more as a unit that intelligently uses the data to add value to the business. As I said before, volume is not the problem, but the generation of value in an environment with such diversity. Customers benefit in many ways: some are evident (Commerce 360 is a tool for shops that help them in their day to day) and others not so much: we study for example the way in which customers browse for bbva.es to improve and simplify processes in such a way that experience is as satisfying as possible. In such cases the customer is not necessarily aware, but “under the hood” many intelligent data-based processes help.