March 11, 2022
In the first installment of our Data-First blog series, we made the case for Data-First banks or credit unions, and why becoming one will be key to thriving in a digital economy. “Data-First” refers to banks that proactively use data to make business decisions. Rather than regarding it as a byproduct of the business cycle, these banks treat data as a raw material with inherent value—and strategically manage it as a business asset.
Many financial institutions recognize the need to adopt this model, but don’t know where to start. Here, we outline the steps organizations should take—and milestones they must meet—on their journey to becoming a Data-First bank.
Becoming a Data-First bank starts with strong data governance. This means adopting bank-wide policies and practices for managing your data assets, with the goal of keeping data trustworthy, accessible and actionable. To begin, your institution should focus their efforts on three pillars: ensuring data quality, collecting and storing data effectively, and understanding the information they have.
For data to be trustworthy, it must be high-quality. One way to ensure data quality is to establish a data stewardship team, made up of people from across your organization. These employees serve as domain experts who are intimately familiar with the data related to their business area, and are responsible for ensuring that data remains accurate, consistent, usable and compliant. This includes monitoring data to identify duplicates and inconsistencies, and reconciling these so the data can be used confidently by other parts of the organization.
Having the right data visualization tools can help data stewards achieve this. With tools like Tableau or Microsoft Power BI, banks can use dashboards to visualize which data exists in the system and run exception reports to flag potential discrepancies.
To achieve strong data governance, your bank should also have a data warehousing strategy. Data warehouses store, integrate and standardize data from across the organization, making it easy to find and use for analytics and reporting. In this way, your warehouse serves as a single source of truth for your bank or credit union’s data.
When data is organized and structured properly, banks can gain an accurate, holistic view of their business and make informed decisions that wouldn’t have been possible in a siloed environment. With the right analytics, data warehousing can help your credit union identify trends, patterns and correlations in customer behavior and market conditions. You can then use these insights to support your customers more proactively and market to them more effectively.
Establishing a data warehouse doesn’t have to be a daunting task. Start small with the few use cases or business problems you’re trying to solve. Then, align on the Minimum Viable Product (MVP) that will put the actionable insights into the decision-makers’ hands. For most community banks, this would be a customer-centric data mart—also known as a Customer Data Platform.
From there, understand which data sources are the data coming from and plan appropriately to build out a flexible data model to service your needs. Recognize that the data warehouse will incur technical debt or require rework in the future; however the ROI you see from the warehouse will allow you to continually evolve it.
Data governance also means having methods for understanding your data—through documentation as well as analytics.
Every Data-First financial institution should be armed with two essential pieces of documentation. The first is a data dictionary—a reference guide that describes the data in your warehouse. This includes metadata for each asset, such as its source, type, format, meaning and relationship to other data. The second is a data catalog, which is an inventory of all data assets and their locations. Much like the Dewey Decimal system, it helps users locate the data they need for analytics and reporting.
To derive meaning from its data, a bank must also have mature analytics capabilities. At minimum, this means being able to use descriptive analytics to describe the current state of business, and then diagnostic analytics to understand why this outcome occurred. Ideally, it also means using predictive analytics to forecast what will happen, and finally prescriptive analytics to determine what should be done to achieve a desired outcome.
While most banks have mastered the first two stages of analytical maturity, few have mastered the second two. That’s because many banks still don’t have a solid foundation of data governance and warehousing, which is critical to achieving predictive and prescriptive analytics. Banks must establish this and master descriptive and diagnostic analytics before they can be ready for predictive and prescriptive analytics. Naturally, a Data-First bank or credit union should be capable of all four.
AI and ML also are vital to analytical maturity. Once your bank has a strong data warehousing foundation, you can apply Al and ML to power the back end of your predictive and prescriptive analytics. This makes them key to staying ahead of customer needs and reducing risk.
When applied to your customer data, AI and ML can help you predict customer behavior with speed and accuracy and take preemptive action to address customer needs. For example, an AI model can automatically flag early indicators that a customer is at risk for default—and provide possible reasons why—even if the customer hasn’t defaulted yet. With this insight, your bank can offer tailored support to help the customer avoid default before it occurs.
In this way, AI and ML can help you gain a deeper understanding of your customers and provide them with higher-value products and services.
To treat data as a true business asset, your bank must have a way to measure ROI from the strategic use of that asset. Here are five pillars you can use to assess the value your data strategy is creating for your organization:
If you’re leading your bank’s Data-First efforts, your ability to calculate this ROI will be critical. Ultimately, you’ll need to report these metrics to stakeholders in order to justify your bank’s investment in its data initiatives.
Data-First banks or credit unions aren’t built in a day. Becoming one is a journey marked by small wins and incremental change. These milestones can help you track your success along the way, and help your bank stay on course to its digital transformation goals.