Financial Services

We help financial institutions contain costs, increase revenue, and mitigate risks, using innovative solutions to improve regulatory compliance, automate business processes, and enhance customer experience. We provide support throughout the journey — from assessing your analytic strategy to managing the challenges of integrating analytics into your operational financial systems and banking processes. Making analytic insight accessible to the right people at the right time is critical to maximizing value derived from data. Examples of our experience in financial services analytics include:

Financial Fraud and Money Laundering Detection

We apply advanced analytics to study risky networks of financial actors, providing greater insight into dynamic financial behavior across high-volume transactional data.  Using the latest advancements in behavioral modeling, our prediction engines can identify fraud far sooner than the typical programmatic solutions, and reveal complex, emerging trends to bank administrators before a loss occurs.

Investment & Asset Management Modeling

We partner with investment clients to optimize and verify original and third-party investment strategies, manage assets, develop proprietary investment modeling software, and provide trusted advice, rapid prototyping, and investment strategy consulting.

Banking Customer Churn Modeling & Marketing Analytics

We provide predictive and text analytics to help banks and financial services corporations acquire new customers, reduce customer attrition, and personalize customer experience through targeted products and services to improve customer loyalty and profitability.

Request a Consultation

Contact Us

Case Studies


Banking Customer Churn Modeling

We partnered with a diversified bank to predict account closures (churn), prioritize marketing interventions, and understand precursors to customer churn. The goal was to improve the predictive performance of an existing account churn model that was based on heuristics in order to reduce account churn rates by at least 10% using only internal data.

Results: We built a predictive model that was 20% more effective at predicting customer churn than the existing techniques.

Read the Case Study

Asset Management Modeling

The project goal was to use text mining and machine learning to extract economic sentiment indicators from millions of disparate documents. We built a weakly-supervised text sentiment classifier for economic indicators using the latest Natural Language Processing tools.

Results: The models make valuable new sources of data available for the client to inform decision-making such as rapid portfolio rebalancing based on continuous market signaling.

Read the Case Study

Improving Credit Card Risk Scoring

The client’s risk model had been developed and deployed by dozens of expert statisticians over many years. The client was interested in whether any new insights would be produced by using modern machine learning techniques. To be adopted, the new model needed to have significantly better predictive performance and be suitable to run in the client’s production scoring environment.

Results: The new model ensemble reduced the number of credit card accounts that defaulted on the client’s evaluation data set by more than 10% when compared to their production model.

Read the Case Study

Compliance Monitoring

We designed and implemented sequence matching patterns for a regulatory agency to identify risks in brokerage firm compliance, anti-money laundering, and equity trading and market-making applications.

Results: These technologies were incorporated into a product that operates on large data marts to uncover patterns of suspicious behavior and provide actionable alerts to financial institutions.