The client was seeking innovative data-driven solutions to identify clients’ trade interests through the use of predictive models and a machine learning feedback loop. In particular the client sought to:
- Assess and prioritize opportunities for business development
- Maximize the deal values by anticipating the end-clients’ needs more efficiently
- Provide customized offers to strengthen end-client relationships
- We combined internal data (e.g., transaction history, inventory lists, client holdings, etc.) and market data (e.g., interest rates, currency fluctuations, market volatility, etc.)
- We applied predictive analytic techniques to identify clients most likely to be interested in trading specific products in the bank’s portfolio
- We built a user feedback loop to continually update the process to identify better opportunities
Fulcrum effectively prioritized end-clients and provided a matching engine to allow the client to stay competitive in the market.
A regional bank’s Human Resources (HR) group was seeking a solution to incorporate market and competitive data to improve personnel evaluation and optimize resource planning.
- We integrated external data (e.g., publicly available competitor data, footprint demographics, etc.) with internal data (e.g., staff experience, branch performance, etc.) and developed new data elements on branch-level market share
- We created a deposit growth segmentation to benchmark branch growth and market penetration, controlling as many factors as available, excluding manager talent
- Lastly, we performed branch performance forecasting for each segment
Fulcrum streamlined the ability to identify exceptional managers, which impacted compensation formulas and recruiting efforts, and also enabled optimized resource planning through forecasting.
The client was looking to improve customer engagement among advisory customers with a process to deliver personalized email driven by customers’ digital behavior and advisory content consumption. The content was not classified which made personalized recommendations difficult.
- We scraped PDF and web pages from its microsites
- We categorized the content using Natural Language Processing (NLP) based on terms and phrases found in the subject matter
- We modeled topic relevancy for each customer and mapped it to the content library
- We scored the content so that the most relevant articles could be recommended to each customer via personalized email campaigns using the matching algorithms
Fulcrum enabled the deployment of a customized content delivery system to increase engagement with high value customers.
The client needed help flagging fraudulent medical insurance claims. The goal was to:
- Determine logic to flag suspicious behavior and identify providers of interest
- Leverage billing data to analyze suspicious providers
- We performed fuzzy matching to link multiple claim records to the same provider(s)
- We used geocoding to create physical distance calculations between providers and between providers and claimants
- We applied statistical methods (e.g., logistic ridge regression with sparse matrices, T-Score calculation, bootstrapping, etc.) to billing data to build propensity scoring models to rule claims as valid or invalid
- Based on our modeling results, we created the fraud flagging logic to identify suspicious behavior and providers of interest
- The team collaborated with the investigative unit, billing unit, and legal unit to create a provider fraud action plan
Fulcrum's data preparation procedures and predictive analytical models provided the quantitative basis for the provider fraud action plan.
The client sought to retain an at-risk multi-million dollar portfolio. In an effort to find a solution, the client needed help with insurance claims data analysis identifying risk drivers and preventing losses. In particular the client wanted to utilize their massive amount of unstructured adjusters’ notes to bring more detail-rich information to the risk consulting team aiming to prevent loss and reduce costs, and to identify additional risk drivers to support ongoing pricing strategy.
The key challenge was the inability to process massive amount of unstructured data (i.e., adjusters’ notes), limiting opportunities to quickly generate insights.
- We performed Natural Language Processing (NLP) on their unstructured adjusters' notes data including Term Frequency-Inverse Document Frequency (TF-IDF) and topic modeling (LDA)
- We enriched the input data to the pricing models through text mining that led to increased prediction accuracy
- We developed an interactive risk analytics tool to deliver loss and cost insights
Fulcrum helped the client with enhanced coverage insights and pricing inputs through the introduction Natural Language Processing on unstructured data.
The client was seeking to apply predictive models to improve its rating structure, moving away from subjective rating. Key challenges included:
- Highly subjective rating process based on individual underwriter's experience, without data-driven guidelines
- Lack of experience and resources evaluating external data sources
Furthermore, the client wanted to explore options to incorporate and build machine learning into the rating structure
- We identified pricing key drivers through LASSO regression
- We developed a statistical model (using GLM) that was implemented for predictive ratings scoring
- We used advanced techniques (including GBM, Random Forest, etc.) to improve model performance
- We created benchmarking models with SVM and NN to ensure model performance (i.e., how close the model can perform when compared with a simulated one without restrictions)
Fulcrum’s data-driven rating structure standardized future underwriting and increased pricing transparency.
The client needed to accomplish faster and more accurate decisions in making personalized offers. Fulcrum applied its experience in digital coupon operations to solve the business problem.
The client partnered with Fulcrum to utilize our hosted computing platform, Digital Fusion.
- Digital Fusion collected data across all channels and sources (POS, display ad, email campaign, web traffic) and created hundreds of micro-segments of customers with similar buying behavior
- The platform deployed stochastic frontier models for each segment to identify winnable shares for each product category
- Then Digital Fusion computed the optimal personalized offers based on purchase, coupon redemption, and digital behavior data
Fulcrum's unification of behavioral and marketing data with predictive models produced granular personalized offers to drive more trips, bigger baskets, and higher margins.