KDnuggets Home » News » 2016 » May » Opinions, Interviews, Reports » Trust and Analytics in the Banking Sector ( 16:n20 )

Trust and Analytics in the Banking Sector


This post explores the intricate relationship between customers, trust, and analytics in the banking sector, and offer actions that banks may need to take to assess the way they assure trust across the analytics lifecycle.



By John Hall, Partner, Data & Analytics KPMG (UK) & Mitch Siegel, Principal and National Financial Services Strategy Leader, KPMG (US).

Trust has always been central to the relationship between a bank and its customers. Since the global financial crisis of 2008, the banking sector has been highly focused on regaining public trust, knowing that a single error or the poor behavior of a few individuals could destroy confidence overnight.

Trust

Analytics have inherent potential to create value and help build trust. Analytics can help create tailored services to customers, detect fraud, assess risk exposure, ensure consistency of service and predict market risks.

Increasingly, as banks become more data-driven, the trustworthiness of their data and analytics (D&A) will underpin trust in relationships with consumers and regulators. As we witness algorithms moving from the back office to the frontline, banks must attend to help assure trust in their analytics.

Banks and their regulators need to have the proper frameworks in place to be able to trust the data and analytics that underpin their decisions and actions. As we proposed in the series’ first article, The Power of Trust in Analytics, enabling trusted analytics across an enterprise can be addressed across four key dimensions, or anchors: quality, effectiveness, integrity and operational control. Our experience suggests that it will take a consistent focus on all four anchors for banks to achieve a higher level of trust in their analytics.

Quality

Most banks are aware of the need for fresh and reliable data and the significant challenges this creates for ‘know your customer’ (KYC) programs, for example — new questions arise as analytics fulfills a greater role in the organization. Is the right mathematical model being applied? And how do you maintain data and analytical quality when working with a wide range of providers and internal talent? What capabilities, processes and controls should surround analytics in different business functions?

Effectiveness

Executives and regulators want to be sure that the analytics not only work in theory, but also in practice, that they achieve their intended purpose in the context in which they operate at any given time. Banks need to ensure that they are using the right analytics (as determined by experts in the field) to achieve their intended outcome in each situation. They will also need to ensure employees are using the analytics in an appropriate way.

Integrity

Legal compliance — for example, with data privacy laws — is only part of the story. In many ways, a clear framework for integrity increases the chances that the analytics are being used in the best interest of the consumer and will be key to creating trusted analytics. Not all enterprises will want to follow the same ethics strategy. Beyond a minimal compliance-only approach, there are options to take a specific, risk-focused approach or a more transparent values-based approach as part of a wider brand strategy, for example. Banks need to understand the implicit deal they have with their customers and will need to be careful not to overstep the bounds agreed in the relationship.

Operational Control

In a fast-changing, data-driven world, the management of analytics never stays still. Banks are familiar with cyber vigilance, technical progress and changing regulations, but are less accustomed to issues of data currency and algorithm ‘lag’, where (over time) analytics can perform differently and lag behind best current practice in human decision making. Banks need clear governance strategies — not only for emerging risks, but also to optimize performance and justify investments.

Ultimately, trust starts by knowing what data you have and understanding the often complex impact of your analytics. A key, initial question banking decision makers should consider as they apply the trust framework is whether they trust their analytics as much as they trust their staff. If the answer is no, the four anchors can help identify where the gaps are.

Authors' Bios:

John Hall is a Partner, Data & Analytics, at KPMG in the UK. John has 20 years’ experience in information technology with a focus on data management and analytics, technology related risk, project and program management. He leads the UK Financial Services Data & Analytics practice across Audit, Tax and Advisory.

Mitch Siegel is a Principal and National Financial Services Strategy Leader at KPMG in the US.  Mitch offers C-suite and Board level guidance to FS organizations on business and operating model strategies intended to drive revenue growth and cost efficiency, with an emphasis on rapid digital transformation through fintech and customer experience enabled strategies.  

Related:


Sign Up