Relative Comparison of Features Predicting Likelihood to Convert on Digital Advertising Platforms

Authors: Varun Chivukula

DOI: https://doi.org/10.5281/zenodo.14593233

Short DOI: https://doi.org/g8xmvh

Country: USA

Full-text Research PDF File:   View   |   Download

Abstract: Digital advertising has evolved into a sophisticated ecosystem that relies heavily on data-driven insights to optimize ad delivery. A critical aspect of this optimization involves predicting the likelihood of user conversion using diverse feature types, such as demographic, behavioral, and contextual data. This study empirically evaluates the relative value of these feature categories in predicting conversion likelihood. Using machine learning models and real-world advertising data, we quantify the predictive power of each feature type and assess their combined effectiveness. The findings highlight the varying contributions of different feature types, offering actionable insights for ad platform strategies. Furthermore, this work discusses the implications of these findings on personalization, data collection policies, and algorithm design.

Keywords: Real time bidding (RTB), Digital advertising, Propensity to convert

Paper Id: 231953

Published On: 2023-07-05

Published In: Volume 11, Issue 4, July-August 2023

Cite This: Relative Comparison of Features Predicting Likelihood to Convert on Digital Advertising Platforms - Varun Chivukula - IJIRMPS Volume 11, Issue 4, July-August 2023. DOI 10.5281/zenodo.14593233

Share this