{"id":868,"date":"2020-04-06T19:43:12","date_gmt":"2020-04-06T19:43:12","guid":{"rendered":"https:\/\/nilg.ai\/?p=868"},"modified":"2022-10-06T15:00:56","modified_gmt":"2022-10-06T15:00:56","slug":"embedding-domain-knowledge-for-estimating-customer-lifetime-value","status":"publish","type":"post","link":"https:\/\/nilg.ai\/pt\/202004\/embedding-domain-knowledge-for-estimating-customer-lifetime-value\/","title":{"rendered":"Embedding Domain Knowledge for Estimating Customer Lifetime Value"},"content":{"rendered":"

\"\"As part of the rise of Deep Neural Networks in the ML community, we have observed an increasing fit-predict approach, where AI practitioners don\u2019t take the time to think about the domain knowledge that is already available and how to embed that knowledge in the models. In this blogpost, we will cover how we created custom-made deep neural networks that combine domain knowledge for estimating customer lifetime value in multiple timesteps, in a project developed together with a major Telco in Portugal. We will start by explaining the problem and the different ways it has been approached in the literature, followed by our solution and the way we incrementally built it.<\/p>\n

While this post reflects the ideas applied in the Telecommunications industry, it can be easily extrapolated to any other subscription-based industry.<\/p>\n

What is Customer Lifetime Value and how to estimate it?<\/h2>\n

The customer lifetime value is the net present value of customers calculated profit over a certain number of months<\/b> [1]. Specifically, in the telecommunications industry, where price transitions are limited, customer monthly margin and customer survival curve are the two major components of this term.<\/p>\n

There are several approaches in the literature for estimating Customer Lifetime Value (CLTV)<\/b>. The most classical approaches are based on statistical models (e.g. Buy \u2019Til You Die Models, Pareto\/NBD model, Recency-Frequency-Monetary value [2] ) and consider features such as purchase frequency and most recent purchases, fitting them to a certain statistical distribution. Most recently, machine-learning-based approaches are reported in the literature, which consider hand-crafted features [3, 4].<\/p>\n

Based on a previously presented idea<\/a> of a model for recommending the best package for a given customer, we set out to build a model that could estimate customer lifetime value in a time window of N months, <\/b>given such recommendation. We intended to build a tool for understanding if CLTV could be used to help our stakeholders make decisions in the best package to offer in an outbound call. By maximizing CLTV for a set of possible package transitions, it is expected that client satisfaction and loyalty will also increase.<\/p>\n

Business Rules and important dataset features<\/b><\/h2>\n

After accepting an offer, clients agree to a fidelization\/contract period of 24 months. If the offer is rejected, the fidelization period remains the same, and the risk of the client leaving the company (churn<\/i>) increases.<\/p>\n

Our dataset contains two groups of features:<\/p>\n