Analytics can help with forecasts on customer behaviour, the purchase rate of a specific product or service at a certain tarif….
Optimising Customer Segmentation
One of the most lucrative and profitable businesses for many banks is wealth management or portfolio management. Many banks focus their resources on building their wealth management business. High net worth clients are also the most demanding of the lot, their expectations are growing by the day so as their level of involvement making decisions related to the way their wealth is deployed. Analytics can play an important role is helping banks understand their clients better and allow them to modify their basic products and services to suit the needs of their special clients. Using clients past data analytics can be used to better understand behavioural patterns of clients and then use these patterns to customize service offerings leading to building sustainable client relationships.
Fraud Detection using unstructured data
Standard fraud detection techniques used in the financial services sector are built for structured data and rely on building patterns of transactions or more aptly fraudulent transactions. Even though this is a widely accepted and proven technique the philosophy of current solutions is based on corrective action which means that it cannot detect the first fraud but can only ensure that subsequent similar transactions can be identified and prevented.
Using unstructured data analytics can help in identifying or predicting and maybe in the future the type of fraud as well with the maturity of technology. Sentiment analysis using unstructured data analytics in the social media can help in identifying or predict the chances of fraud happening and to some extent even the region where the probability of occurrence of fraud is maximum thus optimising in surveillance effectiveness and costs.
Managing customer attrition in the credit card business
Predictive analytics can be used to avoid customer attrition by building attrition models. Attrition models are complex mathematical algorithms which are derived from historical data associated directly or indirectly with the customer.
These models help the card company in designing incentive packages that can aid in customer retention. Analytics can also help companies in identifying possible defaults well in advance which gives these card companies sufficient time to counsel the customer and work out schemes by which default can be avoided. This not only helps in avoiding attrition but also ensures in building long term relationships with customers.
In some cases past history combined with unstructured data can be used to build analytics models that help in customer segregation. This can help in identifying high maintenance low margin customers as well as high margin low maintenance customers. This segregation can then be used to build a model by which costs associated to risks and finance can be spread over the customer base so as to get optimal returns.
Buying pattern optimisation for credit card users
Credit card companies not only have access to data and information about their customer’s financial profile but also have information about their spending and or card usage. By using this information card companies can profile their customers and get vital information about their spend preferences as well and their spend criteria. Card companies because of their access to client information can also understand the affordability criteria and resulting buying decisions patterns.
Using this data effectively credit card companies can work with retail merchants and provide their customers incentivized deals with attractive payment options. They can also leverage bulk discounts with the merchants and pass on the benefit to their customers.