investment banks vs retail banks and the data science race

These are certainly interesting times for all involved in the global sphere of data science and none more so than within the world of banking. Within such a complex and diverse industry there are many moving parts across multiple sectors, making it more challenging for experts to spot trends. With this not only comes uncertainty but anticipation about the potential that lies within for many top-tier banks. However there is one trend in particular that has elevated to the forefront of data science. Investment banks are playing a game of catch-up whilst the candidate job market continues to gather pace.

Today most retail banks have mastered the art of leveraging the use of data science, by analyzing customer behavior and predicating which customers are most likely to default on credit cards; as well as identifying potentially fraudulent transactions. As a result, this provides a data science workload, driving the need for a retail bank to invest heavily in modern technology and recruit personnel.

Many of the top-tier retail banks are mirroring the technological advancements of the global insurance and management consultancy sectors, by creating data science groups housing predictive modelers and machine learning experts. These groups then act as an internal resource of intelligence; aiding data-driven decision making across the organization.

It is one thing identifying the need to invest aggressively in modern technology and personnel, although there are only a select few top-tier banks who have successfully executed such a strategy. There are only a select few top-tier banks who have successfully executed such a strategy. It is only a matter of time before others follow suit and begin to gather deeper customer intelligence, and realize the tangible, quantifiable benefits at stake for the entire organization going forward.

Growing Importance of Data Science

As for investment banks, substantial consideration must be given to fact that they currently struggle to determine what beneficial data science use-cases exist within their business model. Undoubtedly this issue will continue going forward into the next quarter.

The demise of the Wall Street boom-era left many investment banks in a perilous condition and those still standing have done remarkably well to stabilize. However only a fool would underestimate the beast that lies within investment banks, in the form of huge potential relevant to spending power and intelligent strategists. Once valid use-cases are identified and streamlined, then the hard-push will begin to catch up with retail banks.

During this state of industry flux, we at Harnham have witnessed notable developments taking place within the data science candidate market. As the demand for us to supply quality personnel intensifies, the job market has become more competitive than ever for candidates and employers alike.

This has led Harnham to open an office in New York City, as it is undoubtedly the banking & financial services stronghold of the United States. With this, comes the powerful status of being the most competitive data science job market in the world. Equally this creates a number of problems, yet a number of advantages for those operating within this space.

Banking on Life Long Learning

One trend in particular has seen quants, reared in the traditional banking world of Wall Street, who then seek a career path towards modern data science falter; as such a move means utilizing unknown technology and software packages.  These candidates now face increasing difficulty making this transition internally due to a lack of modern skills.

Their woes are further compounded when entering the job market, as most are leaving roles at investment banks with the same technology skill-set they acquired many years ago. With that said, all is not lost as there are an abundance of online data science courses now available at prodigious universities such as Harvard, allowing skill-short candidates an opportunity to get to grips with modern programming languages such as Python or R etc. and big data packages such as the Hadoop platform.

Furthermore the emergence of data science boot-camps such as the Metis Boot-Camp (an immersive Data Science boot-camp in New York City) and Columbia University offer a Master’s Degree in Statistics, with focus on data science and machine learning; provide an alternative option. Many job seekers have gone as far as ceasing employment to join such boot-camps. Some have gone a step further still, by going back to university and undertaking a Masters in a data science related field.

As a result of this movement in the candidate market, there is a tug-of-war taking place between those coming from the top universities with Masters or PhD qualifications and those leaving the investment banks searching for data science roles on the retail side.

The Winner Takes It All

Who gets hired, all boils down to a question of each individual banks requirements and which skills they deem as the most important - domain expertise and practical experience verses a highly accomplished modern data science skill-set. Those who combine the two will naturally be more competitive in their search and overall more enticing to potential employers.

From a positive perspective, despite the barriers to entry mentioned, we are right in the midst of a data science market that is actively evolving all the time. It is fast paced, dynamic, exciting and full of opportunity on all fronts.

Given the lean years we have experienced, this is an exciting market to be part of and one that will be at the forefront of a data and analytics evolution, which will grow exponentially over the next five years and beyond.

	Allister Duncan