Sara receives an email from her bank about a home loan offer. That Sara is a privileged customer for long and uses the bank’s credit card lends no insight into this offer. On its part, the bank had done its homework in terms of creating the product propensity index – Behavioral segmentation propped up by Customer Lifetime Value (CLV) and attrition model. This let the bank to cross-sell its product to customers who were more likely to buy the recommended bank product, home loan in this case.
Banks turn proactive in using propensity-to-buy models to link customer characteristics and product needs to the right solution. In gaining high propensity customers as leads, banks drive future marketing and sales initiatives to enhance marketing effectiveness and increase sales productivity.
How do banks keep propensity modelling at the forefront to nurture customer delight and increase profitability?
Look before the Product Propensity Leap
A customer from a bank gets a call for this reason – Transfer the home loan sanctioned by another bank to the very same bank where the customer holds the account. Another customer receives an email about using seamless investing and multiple trading platform facilities of the bank to seek better investment fortunes. The same bank targets another customer with an offer for car insurance.
This personalized communication, marketing on the part of the bank is spurred by ‘propensity-to-buy’ metric. In hitting the bull’s eye in terms of promoting the next best product, the bank extracted insights across key features including:
- Is the customer the right target to sell this product?
- What’s the customer’s propensity to buy this product?
- What’s the predicted lifetime value of the customer?
- Would this next best product enhance customer profitability?
- Would the risk score of the customer take a hit if the customer buys this proposed next best product?
With customer intelligence acquired though these queries, the bank kept customer-centricity and profitability in balance to recommend the next best product for its customers.
Data and Analytical Push to Propensity Modelling
It starts with historical data revealing past purchases as well as customer data. Data and the analytical push is inclined to predict customers who will buy a product. As to data surrounding the customer, social networks, demographics, transaction data and online behavior offer the promise of better customer intelligence. Random forests, Logistic regression and decision trees offer a good choice of algorithms to predict customers’ propensity to buy a bank product.
For unearthing the ‘Propensity to buy’ through propensity modelling, every tiny bit of customer information matters in the way it can lead to better results in predicting a product of interest for a customer segment. And with banks dealing with more products including mortgages, loans and credit cards, and customers, Single Customer view allows banks to get a comprehensive view of customer interactions and strengthens propensity modelling.
‘Everything Customer’ Ring to Propensity Model
Banks also want to tap into the potential of targeting ‘customer segments’ that show high propensity for balance-transfer offer – This is about acquiring credit card customers attached to competitors. To make this happen, the bank builds a propensity model augmented by customer segmentation and predictive modelling to taste the twin benefits of more customers at low acquisition costs.
Customer behavioral segmentation nurtured through predictive models empower banks to predict and pinpoint customers who would be needing a specific bank product. From a marketing perspective, micro-segmentation and product propensity model make up for a winning combo to create the right product and offers for the right customers.
As a core component of the product propensity index, Customer Lifetime Value adds value by predicting the future potential customer revenue taking cue from product purchase propensity and historical behaviors. In effect, lifetime value adds a critical dimension to Product Propensity Model in the way it guides banks in targeting customers towards generating more revenue. In making the Product Propensity Model promote a sync between customer delight and bank profitability, more dimensions in the form of customer preferences and product usefulness are infused to ensure right customers become the target for right bank products.
With one of the top-performing banks in Southeast Asia envisioning a solution to address key challenges of identifying customers likely to buy specific products and designing promotional campaigns for the right customers to enhance customer loyalty, Saksoft created a propensity model leveraging ML algorithms and predictive modelling techniques to guide the bank in identifying high propensity customers, optimizing qualified lead generation, running targeted campaigns and increasing sales productivity.