Today, data science is not only an industry buzzword, but an emerging business need to drive profitability, as we will be exploring in our upcoming Summer 2018 Series. Data science is applied to pinpoint and quantify risk, identify opportunities to maximize customer value, and provide insight into product and service offering changes that will yield improved business results.
Challenges Facing Investment Banks in the Adoption of Data Science
May 17, 2018
Today, data science is not only an industry buzzword, but an emerging business need to drive profitability, as we will be exploring in our upcoming Summer 2018 Series. Data science is applied to pinpoint and quantify risk, identify opportunities to maximize customer value, and provide insight into product and service offering changes that will yield improved business results. With this being said, many financial services organizations are taking an aggressive approach to reclaim their missed opportunities to profit and grow further. Though often times, such endeavors end up falling short of the expectations. Why?
Failed to swiftly demonstrate the value-add
Many leaders seek innovation through data, analytics and technology. In fact, according to McKinsey Quarterly, more than 70 percent of the senior executives say innovation will be one of the top three drivers of growth for their companies in the next three to five years1. Specifically in the banking industry, automation, digitization, machine learning, etc. are some of the biggest goals in achieving innovation. Despite their desire for innovation and data science, we have seen the stakeholders’ expectation gap impede the speed and efficacy of next generation technology development. Why is this the case?
Often times, the stakeholders can become frustrated with what is perceived as slower than expected results. While it is not uncommon to see a SWAT team of advanced analytics professionals working around the clock, they may appear to be ineffective despite these efforts. What is unseen is the significant amount of time spent dealing with data acquisition, processing and transformation -- making it challenging to quickly translate the dollar value.
We have a great story to share: How we were able to help an investment bank build a research recommendation engine by demonstrating the incremental lift one step at a time -- by “Going Small”, coming in our next blog to be published in June, 2018.
Provided too much information (still)
With the support of high-level executives, some investment banks have already begun to pursue a long-term solution by using data science to generate insights, build applications, and develop platforms. However, the benefit to end users is not always emerging at the rate expected while their data science teams are becoming increasingly swamped behind the scenes. The management teams are left to wonder -- what is going wrong? We have seen and worked with such institutions that house an impressive repository of analytic apps that serve a variety of end users and stakeholders. These institutions noticed that they began to run into the aforementioned issue because the user stories weren’t thoroughly studied upfront and the influx of information (generated from all these apps) became confusing.
It can be easy to overlook the fact that a strong process is not only needed to design and develop apps, but also to curate, renovate and maintain them. As part of our Summer 2018 Series, we will be covering a story of how we assisted a major bank with the development of content recommendation process.
Rapid change in environment
Emerging RegTech solutions have started offering sustainable and seamless tools that monitor transactions and generate alerts for sanctions and anti-money laundering. With the help of such tools, the percentage of transactions and files that need to be manually reviewed have been reduced significantly. However, compliance workloads remain high due to an increase in cross-border transactions, heightened expectations from regulators, and the most recent updates in sanctions (i.e., Russia). In other words, the number of transactions being screened has grown tremendously which offsets the gained efficiency of screening.
Then what can be done further? What’s missing is the ability provide key insights to accelerate the investigative work. Generating the right insights (e.g., the reason code for flagging, case classification, etc.) can streamline the workflow and reduce the early steps of an investigation -- greatly aiding the efficiency of the compliance team. We will cover this topic in further detail in an upcoming post.
While there are numerous ways that data science can help to establish companies as industry leaders, it can be difficult to apply successfully. The good news is that there are only few large institutions that are marginally ahead of the competition as the complexity of an organization can sometimes make it slower to adopt and change -- whether it is driven by the expectations gap, lack of promotion, or change in trends. What’s important is that the innovative leaders need to stay proactive and vocal in pursuing data science.
Our Summer 2018 Series will highlight a success story a month, so stay tuned for our upcoming articles. In the meantime, find out how we helped our clients stay at the forefront of innovation.
Source:  Joanna Barsh, Marla M. Capozzi, Jonathan Davidson 2008 January, Leadership and Innovation.