Member-only story

Data Science in Marketing: Hands-on Uplift Modelling with Python

Learn how to use causal machine learning to maximise ROI for marketing campaigns

Rebecca Vickery
8 min read4 days ago
Customer segments generated by uplift models. Image by Author

This is the second article in my series on Data Science in Marketing. The first article can be found below.

  1. Data Science in Marketing: Hands-on Propensity Modelling with Python

Offers and promotions are widely used in marketing campaigns. Discount vouchers, incentives, and giveaways are common interventions designed to generate actions, like purchasing or renewing a service subscription.

Marketing teams usually have a limited budget, and for that reason may not want to provide every customer with an offer. Typically marketing teams will identify segments of customers to target and this is often achieved using models like Churn Prediction or Customer Lifetime Value (CLV).

These techniques are all useful but none of them address the true challenge. A customer with a high churn propensity or high CLV may very well not respond to an offer, and in some cases, sending an incentive to these customers may trigger them to leave you. To ensure the highest rate of success, marketing teams should focus on sending offers to those customers who are most receptive to the promotion.

--

--

Rebecca Vickery
Rebecca Vickery

Written by Rebecca Vickery

Data Scientist | Writer | Speaker

Responses (1)