Tackling transformative change: Investing in data, talent and technology

While it’s true that artificial intelligence and machine learning are not new concepts, the accumulation of huge stockpiles of data, coupled with rapid advances in distributed computing technology now widely accessible on cloud platforms at modest cost, have raised the field’s profile and rendered it a persistent, top-of-mind object of concern for senior financial executives.

The good news is that in the midst of relentless media attention concerning AI and machine learning, the “natural laws” of analytics remain intact.

In an era of accelerated progress and instability, how are best-in-class organizations positioning themselves to reap the benefits? The answer: They are shrewdly investing in three particular areas without losing sight of the fundamental, natural laws, as reinforced by the Aite Group’s October 2019 study, entitled Current State Assessment: Global Analytics Ecosystem, which focuses on the needs of the financial services industry.

The first area of investment is data, which is and always has been, the lifeblood of AI/ML and analytic solutions. An organization’s mission-critical, often undervalued, data assets should be actively managed as a portfolio, typically composed of a blend of owned and licensed content.

Each asset in a well-managed portfolio should contribute a unique signal that contributes to materially better decisions, as demonstrated by a rigorous validation framework. Robust internal processes must be built to preserve the high and consistent data quality necessary to serve as a stable platform for analytically driven decisions. More than half of the Aite Group survey respondents indicated plans to increase spending on data across nearly every category.

The second area of investment is talent in the data science and machine learning disciplines. In a period of rapid change, it’s unrealistic to expect good things to result from a monolithic strategy that depends solely on hiring elusive “unicorns.”

Best-in-class institutions have adopted a learning mindset and portfolio approach to managing this precious, yet constrained, resource. They are nurturing relationships with leading university programs in targeted fields. They are strategically selecting and engaging collaboratively with commercial partners who contribute proven, complementary expertise.

Above all, they are creating internal environments that promote knowledge exchange and retention, enabling their existing domain experts and analytic professionals to steadily upgrade and enhance their capabilities.

The Aite Group study found that 12 percent of financial institutions already augment their capabilities through external partnerships, while another 45 percent might benefit from such arrangements because of staff shortages, whether overall or in particular areas of expertise. Furthermore, the survey found that of the 90 percent of respondents who leverage data science talent, they rely, on average, upon more than two different methods of sourcing it, including external partnerships.

The third noteworthy area of investment is technology, whose steady advance is rendering complex and unfamiliar AI/ML solutions ever easier to digest and understand.

A growing array of tools aims to automate and reduce the substantial effort required to control, interpret, and explain ML algorithms, with particular benefits for highly regulated industries. Meanwhile, automated modeling platforms have emerged that streamline the construction and documentation of advanced analytic solutions. So-called “low-code/no-code” options like these naturally invite more diverse expertise and experience into the construction effort, potentially yielding solutions of higher overall quality and relevance.

Aite found that 39 percent of financial institutions already have an ML-ready platform, while an additional 35 percent are considering taking the plunge within the next 24 months. Clearly, advanced analytic technologies are already making an impact for some, and will soon become commonplace for most.

There is tremendous hype swirling around the topic of AI/ML at this moment in history, but the opportunities are undoubtedly real. Ultimately, AI/ML, when used properly, promises to improve human decision-making.

Our bank executive and analytics professional should rest assured: The fundamental and time-tested processes used to validate and monitor analytic solutions have new facets, but remain essentially intact. Applying these processes rigorously, while adopting a flexible, learning posture toward new sources of data, talent, and technology will go a long way towards addressing today’s top AI and ML challenges. Striking the right balance will help increase their abilities to capitalize on an era of rapid, transformative change.

For reprint and licensing requests for this article, click here.