Improving Ego-Cluster for Network Effect Measurement

TL;DR

This paper introduces an improved cluster-level experimentation method for accurately measuring creator-side metrics in social network A/B tests, addressing unit mismatch issues to enhance efficiency and flexibility.

Contribution

It presents a novel, improved clustering algorithm for network effect measurement that better handles unit mismatches in A/B testing, boosting experimental speed and accuracy.

Findings

Enhanced clustering algorithm increases measurement accuracy.

Method improves experiment velocity and flexibility.

Supports better assessment of network effects in social platforms.

Abstract

The network effect, wherein one user's activity impacts another user, is common in social network platforms. Many new features in social networks are specifically designed to create a network effect, enhancing user engagement. For instance, content creators tend to produce more when their articles and posts receive positive feedback from followers. This paper discusses a new cluster-level experimentation methodology for measuring creator-side metrics in the context of A/B experiments. The methodology is designed to address cases where the experiment randomization unit and the metric measurement unit differ. It is a crucial part of LinkedIn's overall strategy to foster a robust creator community and ecosystem. The method is developed based on widely-cited research at LinkedIn but significantly improves the efficiency and flexibility of the clustering algorithm. This improvement results in a stronger capability for measuring creator-side metrics and an increased velocity for creator-related experiments.