Python Conference APAC 2024

Alternative A/B Testing: Using Causal Inference to Measure Impact of Your Experiment
2024-10-27 , CLASS #4 - 3A
Language: English

Discover powerful alternatives to A/B testing for measuring experiment impact using causal inference techniques. This talk will demonstrate how we can leverage the matching process methods, such as propensity score matching to reduce bias when comparing groups, providing reliable metrics even when traditional A/B testing isn't feasible. Learn practical applications, and new approaches, and gain actionable insights to broaden your data analysis toolkit and effectively measure the impact of your features or products.


In this talk, we will learn how Python can be used to solve the real-world problem of measuring the impact of experiments without relying solely on A/B testing. By focusing on causal inference, particularly the matching process and propensity score matching, the audience will learn how to reduce bias when comparing groups. This method is handy for situations where it was not possible to design a proper A/B test initially, making it a practical and valuable alternative.

This method is relevant based on my experience as a data person who wants to measure the impact on the company itself, offering deep and unique perspectives on measuring experiment impact without A/B testing.

The talk starts with an introduction to the limitations of A/B testing, followed by an in-depth explanation of the matching process, specifically propensity score matching. Practical demonstrations and case studies will illustrate the concepts, and a brief Q&A session will ensure clarity and address audience queries.

Attendees hopefully will gain actionable insights into alternative methods for measuring experiment impact, broadening their toolkit for data analysis and product evaluation.

See also:

A data professional with over five years of experience in the fintech, e-commerce, and banking industries. He has led projects using Python to enhance customer behavior models, improve product adoption, and increase operational efficiency, delivering measurable business outcomes.

Previously, Aris worked with major banks and one of the biggest e-commerce platforms in Indonesia, where he developed predictive models for cross-selling and operational improvements, driving gains in customer engagement and process optimization. He is passionate about leveraging Python and machine learning to transform data into actionable insights that drive innovation and growth.