Propensity Modelling - Using h2o and DALEX to Estimate the Likelihood of Purchasing a Financial Product - Part 3 of 3 - Optimise Profit With the Expected Value Framework ·

In this day and age, a business that leverages data to understand the drivers of its customers' behaviour has a true competitive advantage. Organisations can dramatically improve their performance in the market by analysing customer level data in an effective way and focus their efforts towards those that are more likely to engage. One trialled and tested approach to tease out this type of insight is Propensity Modelling, which combines information such as a customers' demographics (age, race, religion, gender, family size, ethnicity, income, education level), psycho-graphic (social class, lifestyle and personality characteristics), engagement (emails opened, emails clicked, searches on mobile app, webpage dwell time, etc.), user experience (customer service phone and email wait times, number of refunds, average shipping times), and user behaviour (purchase value on different time-scales, number of days since most recent purchase, time between offer and conversion, etc.) to estimate the likelihood of a certain customer profile to performing a certain type of behaviour (e.g. the purchase of a product). Once you understand the probability of a certain customer to interact with your brand, buy a product or a sign up for a service, you can use this information to create scenarios, be it minimising marketing expenditure, maximising acquisition targets, and optimise email send frequency or depth of discount. In this project I'm analysing the results of a bank direct marketing campaign to sell term deposits in order to identify what type of customer is more likely to respond. The marketing campaigns were based on phone calls and more than one contact to the same client was required at times.