Online Program Home
My Program

Abstract Details

Activity Number: 663 - Topics in Large-Scale Online Experimentation
Type: Topic Contributed
Date/Time: Thursday, August 2, 2018 : 10:30 AM to 12:20 PM
Sponsor: Section on Statistical Consulting
Abstract #330588
Title: Advances in Measuring User Learning
Author(s): Niall Cardin* and Henning Hohnhold
Companies: Google Inc. and Waymo
Keywords: User Learning; Long Term; Dosage; Ads; Blindness

Online services need to understand how product decisions affect their users. Product changes will often cause an immediate change in user behavior and this is what most A/B tests will measure. However, there is often longer term effect as users learn new behaviors in reaction to the change. Accurately measuring these long term effects is critical to properly understanding how to make the best products. We present an extension to earlier work that further explores how Google search ads affect our users. We compare user learning at different dosage levels, over very long time periods, and user product usage learning.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2018 program