The R package ProfitBoost contains the algorithm developed by Lemmens and Gupta (2020), Managing Churn to Maximize Profits, Marketing Science. The package is available free of charge and can be downloaded on GitHub: https://github.com/ProfitBoost-Lab/ProfitBoost.
ProfitBoost predicts which customers firms should target during their proactive retention campaigns in order to maximize the profitability of their campaigns. The code requires as inputs a customer database with customer-specific characteristics (e.g., customer tenure, spending, demographics) as well as a treatment variable (intervention dummy indicating whether a customer received a retention offer). Importantly, the algorithm relies on an A/B test, so it is important that the intervention is randomly allocated.
The package outputs the expected profit of the retention campaign and the predicted profit lift for each customer in the dataset. Last but not least, it also outputs the optimal number of customers to target during the campaign.
Several companies are currently experimenting with the package, and an update will be posted soon on this website. If you have any questions about the code, please do not hesitate to contact me. A tutorial is currently under development.
If you are interested in learning how to estimate latent moderation, have a look at our paper called Six Methods For Latent Moderation Analysis In Marketing Research: A Comparison And Guidelines. Constant Pieters sets together an amazing repository on OSF that contains all the codes and explanations: https://osf.io/py7jx/.
The codes allow the user to compare six estimation methods for latent moderation, which are very common in marketing research and experimental work. Think about an experiment that manipulates the context a consumer operates in, and we want to understand the effect of a psychological trait (latent) on the consumer’s behavior. These situations occur in companies’ daily practices. For instance, how does consumers’ thinking style impact the effectiveness of a particular marketing intervention (a social media campaign)? How does a customer’s relationship with a company affect the success of a proactive retention offer?
If you have any questions about the code, please do not hesitate to contact me or Constant Pieters. A tutorial is currently under development.
If you are interested in learning how to estimate conditional average treatment effects using causal forests and, most importantly, use these insights to optimize a targeted (fundraising) campaign, have a look at our paper called Enhancing Donor Agency to Improve Charitable Giving: Strategies and Heterogeneity. We provide access to all codes and datasets used in the paper via the following OSF repository: https://osf.io/4nzsw/.
If you have any questions about the code, please do not hesitate to contact me. A tutorial is currently under development.
