A Peek Behind Credit Union Data Analytics At Lake Trust

“Big data” is long past its heyday as a buzzword in the financial services industry. For credit union data analytics, leveraging vast stores of internal and external data is more standard practice than emerging trend. The ability to analyze member behavior, predict market shifts, and optimize operations is now an essential function of a well-run cooperative.

Data management and analytics are core operational components at Lake Trust Credit Union ($2.6B, Brighton, MI), which uses analytics to improve branch operations, enhance product performance, and create predictive models that guide strategic decision-making.

Branch Dashboard Enhances Decision-Making

Using analytics, Lake Trust has developed a tool to track new member and account openings, teller transactions, and referrals across the credit union’s 23 branches. By analyzing key data points — such as ATM and teller transactions, new member acquisitions, and loan activity — the cooperative ensures its branches are optimally placed to efficiently serve more than 175,000 members.

“The branch dashboard has allowed us to make data-driven decisions about where to move or open branches,” says Razi Qadri, Lake Trust’s chief operating and information officer who joined the organization in July 2023 after spending 12 years in a similar role at another large Michigan credit union. “For example, we relocated our Centerline branch to Warren based on traffic and performance insights.”

The dashboard also allows for direct comparisons between branches to assess their relative performance and influence future decision-making.

Branch Forecast Analysis Optimizes Expansion

Before opening a new branch, Lake Trust conducts a thorough analysis of member activity in that specific market. This includes reviewing member transactions at its own ATMs along with general market demand for in-person banking services.

“This process helps us determine where to expand our physical presence,” says Shubhangi Pararha, who took on the role of business intelligence manager at Lake Trust in June 2022. “Rather than guessing, we use real data to forecast the impact of new branches on membership growth and financial performance.”

ATM/VTM traffic patterns and teller engagement rates provide valuable insight into the potential success of a branch location, the business intelligence manager says. Additionally, total deposits and loan performance in a given region contribute to decisions about whether to move forward with expansion.

A branch performance dashboard helps Lake Trust managers monitor performance across time.

Lake Trust managers use a comparison tool like this template to track and evaluate performance across the cooperative’s 23-branch network.

Performance Monitoring Allows For New Loans

Lending benefits from credit union data analytics, too.

Lake Trust launched its Easy Cash loan two years ago to offer small-dollar personal loans based on a member’s total relationship with the credit union rather than their credit score. Data analytics played a key role in forecasting how many members would qualify and how the loan would impact financial metrics.

“The product performed in line with expectations, but we did see a spike in charge-offs, particularly among newer members and repeat borrowers,” Qadri says. “To address this, we refined our qualification criteria, making small adjustments to deposit and membership requirements.”

The credit union continues to closely monitor Easy Cash, using ongoing data analysis to fine-tune eligibility requirements and maintain responsible lending practices while ensuring accessibility for members.

Predictive Analytics Enhances Member Service

Lake Trust has also leveraged data to create predictive models that improve operations and mitigate risks. These efforts include:

• Predicting loan defaults by analyzing historical member data.
• Identifying suspicious transactions to enhance fraud detection.
• Analyzing seasonal deposit and withdrawal trends to optimize liquidity.
• Adjusting staffing based on branch foot traffic trends.
• Identifying high-demand regions for future branch locations.

“Predictive analytics allows us to be proactive rather than reactive,” Pararha says. “For example, we use data to adjust our staffing levels so we are fully equipped to meet member demand in real time.”

Integrating additional datasets into its models has also enabled Lake Trust to identify members who might benefit from specific products or services, allowing for more targeted and personalized financial solutions.

Key Areas And Overcoming Challenges

For credit unions looking to improve their data strategies, Qadri and Pararha recommend focusing on two key areas — the holistic member relationship and the life of loans. The holistic member relationship with the credit union includes all loan and deposit accounts, balances, transaction history, member interaction channels — such as online, call center, and branch visits — referrals, and insights for recommending the next best product. The life of loans includes the full cycle, from application to servicing, denial rates, withdrawal rates, and loan abandonment to ensure continuous optimization.

That’s not to say these two areas won’t pose challenges.

“Credit unions often work with multiple CUSOs and more for services like mortgage processing and credit cards, making it critical to aggregate data from multiple sources to build a complete picture,” Pararha says.

Standardized data ingestion processes and structured data mapping have helped Lake Trust overcome the complexities of such data integration, which can be one of the biggest challenges in such work. Issues the Michigan shop has tackled include inconsistent data formats, delivery delays, and a lack of standardized data dictionaries.

But Lake Trust has made inroads and continues to power on.

“One of our biggest wins has been automating data validation to flag inconsistencies before they impact reporting,” Pararha says. “We’ve also standardized ingestion pipelines and improved vendor communication to ensure consistency.”

By proactively addressing these obstacles, Lake Trust has improved reporting accuracy and enhanced its ability to generate actionable insights. In cases where required data fields were unavailable from standard extracts, Lake Trust negotiated custom extracts or alternative data sources to close reporting gaps.

Lessons For Next Time

If Lake Trust could restart its analytics journey, Pararha and Qadri say they would prioritize structuring data correctly from the beginning. Early on, the credit union wrestled with duplicate data fields across multiple systems and no clear source of truth.

“We spent significant time rearchitecting our data warehouse,” Qadri says. “Now, we have a normalized data structure that ensures efficiency and accuracy.”

Another major improvement has been filtering out “noise” — redundant or inconsistent data that complicates analysis.

“For example, transaction data resided in multiple places, including ATM, VTM, and credit card systems,” Pararha says. “We are now consolidating this into a single source to simplify reporting and decision-making.”

Indeed, standardizing data definitions and automating data cleaning processes have played a significant role in enhancing data integrity and usability across Lake Trust’s analytics environment.

Meanwhile, its journey in data analytics showcases the power of using information to drive operational efficiency, enhance member services, and optimize financial performance. By investing in analytics now, it is ensuring smarter, more strategic decision-making moving forward.