Customer Segmentation in banking is now more about gaining nuanced insights by using different viewpoints to understand the customer. And as personalization and enriched customer experience have turned into a banking requisite, banking customer segmentation looks beyond the standard demography and geography with the scope now stretching across attitudes, behaviors, lifestyle, interests and preferences.
Generic ads and marketing are out of sync with the times. Banks need to delve into granular details around the customer to know the customer better. Using renewed customer segmentation in banking, opportunities to improve sales and revenue can be unearthed among customer segments, and new customer segments that offer growth opportunities can also be unearthed.
How to extract value from Customer Segmentation in Banking?
Customer segmentation in banking leveraging Micro-segmentation approach
The traditional banking customer segmentation is now being overtaken by the ‘micro-segmentation’ approach. It is done by creating smaller customer segments with finer customer–related details used for customer segmentation in banking – it is about creating micro-segments by looking at banking customer segmentation from various ‘customer contexts’.
John is a prolific user of a bank’s credit card. He shops extensively during the weekends. He uses his credit card to buy branded shirts and trousers; dine at some of the most exquisite restaurants during the weekends. He also uses his credit card on some of the e-commerce sites to avail some of the fabulous offers promoted by brands (shirts).
There are more customers like John. That’s a cue as well to form a micro-segment in this exercise around Customer segmentation in banking. The bank has captured specific customer behaviors to tap into targeted marketing – Say offering discounts on purchase of branded shirts (choice of 3 brands) for a specific amount – Identifying the high-potential card user to increase wallet share through can be accomplished through this banking customer segmentation exercise.
Banking customer segmentation has evolved in terms of analysing customer behaviour based on different perspectives.
Our experience with Customer segmentation in banking
A leading Fintech company aspired to automate loan approval process and increase approved loan applications. For this, the Fintech company took the banking customer segmentation route to accelerate the process based on priority.
Saksoft helped them adopt a robust approach – Customer segmentation had to be achieved through a classification approach, where the target column held information to understand if a loan can be approved or not (binary).
Going beyond predicting Y/N with respect to a loan application for approval, it was more about predicting the probabilities of the classification model. The model had to provide probabilities as output, which would answer how likely a customer is positioned to get the loan approved or not instead of Yes or a No. This customer segmentation in banking exercise encompassed the following.
- Predicting probabilities from 0 to 1
- Fixing thresholds for the predicted probabilities
- For instance, let’s say that we considered 0 to 0.3 as very bad, above 0.3 and less than 0.5 as bad, above 0.5 to 0.7 as good, and above 0.7 as very good
The Threshold served to create segments as part of banking customer segmentation as per company’s needs in order to prioritize and accelerate loan approval process. The Fintech Company not only used customer segmentation in banking to automate loan approval process but also set the tone for increasing approved loan applications.
