Financial services analytics is typically focused on the collection and analysis of large datasets. The data might come from customer transactions, customer interactions, internal processes or other public information with respect to the customer. The analysis of the data can done using a variety of techniques, including machine learning, process mapping and data visualization.
The goal of the analysis might be to:
- Improve a process within the financial organization
- Understand customer behavior (ex. for improved risk management)
- Understand a customer’s interactions (ex. to improve the customer’s experience).
Below are some examples of projects within the Financial Services Analytics domain.
1) Process Optimization via Process Mining
The basic idea in process mining is to extract knowledge from event logs recorded by an information system. Process mining aims at improving a process by providing techniques and tools for discovering process, control, data, organizational, and social structures from event log. An example use of process mining would be to leverage process modeling, process mining, predictive modeling, and optimization techniques to streamline and improve a service process (customer facing or internal). Typical objectives might include:
- Document the current “as is” process via process modeling and process mining techniques
- Identify gaps / issues in current process - compare the expected process (from Subject Matter Experts) with the actual as-is process (obtained via process mining – analyzing system output across different key points within a process)
- Develop process-based predictive analytics for the service - based on mined models and historical logs
2) Risk Management Analytics
Analytics can be used to improve a banks credit and fraud exposure. One example would be to improve the identification of fraud in point-of-sale and online transactions. This is particularly challenging since it is a dynamic environment, where patterns within fraudulent transactions change with time. A project might aim, for example, to create advanced online and real-time learning algorithms and Artificial Intelligence methodologies that result in significant improvements in the ability to rapidly detect fraudulent transactions.
3) Customer Behavior Analytics
A framework/model to anticipate customer behaviors and interactions is often developed, and that knowledge might be used to design and implement proactive customer service treatment strategies. For example, one might use predictive analytics to anticipate when a customer will contact the financial services organization, how often it will occur (frequency), the channel or preferred medium, and the foundational information relative to why the customer contact is occurring. That insight and knowledge on the customer behavior may be applied for proactive customer service treatments that provide for a better customer experience. This analysis could focus on answering questions such as:
- What are the key drivers of customer interactions? Which key events and/or sequence of events influence customer behaviors?
- Which customer groupings are most sensitive to those drivers? How sensitive are the various customer groupings to the same drivers?