Unlimited Testing: Let’s Test Your Emails with AI

Nguyen Nguyen (nnguyenlethanh@mbs.miami.edu), Joseph Johnson (jjohnson@bus.miami.edu) and Michael Tsiros (tsiros@miami.edu)
Additional contact information
Nguyen Nguyen: Miami Herbert Business School, University of Miami, Coral Gables, Florida 33124
Joseph Johnson: Miami Herbert Business School, University of Miami, Coral Gables, Florida 33124
Michael Tsiros: Miami Herbert Business School, University of Miami, Coral Gables, Florida 33124

Marketing Science, 2024, vol. 43, issue 2, 419-439

Abstract: Testing email marketing effectiveness is an active research area because email remains an important channel for customer acquisition and retention. Email open rates are a key measure of campaign effectiveness. Scholars identify three predictors of open rates: recipients’ characteristics, headline characteristics, and sending time. The industry-favored A/B testing has three drawbacks: it takes hours, depletes lists available for main campaigns, and limits testable email versions because of sample size and power requirements. These limitations continue to motivate researchers to build and improve open rate prediction models. Although they reduce testing time, models developed in marketing use only recipients’ past open rates as predictors. By contrast, models in computer science typically use only email headline characteristics as predictors. Consequently, current models’ open rate prediction errors are high. The authors address the limitations of both literature streams and use all three predictors and machine learning to build an email open rate predictor (EMOP) based on their universal emotion detector (UED). They test EMOP on data from four brands and set state-of-the-art prediction results. Experimental validation shows that EMOP can pick the best headline from a set of professionally generated headlines. Also, UED ranked second at the SemEval 2018 Task 1 E-c competition as of January 5, 2023.

Keywords: email marketing; machine learning; text mining; emotion detection; NLP; deep learning (search for similar items in EconPapers)
Date: 2024
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
http://dx.doi.org/10.1287/mksc.2021.0126 (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:inm:ormksc:v:43:y:2024:i:2:p:419-439

Access Statistics for this article

More articles in Marketing Science from INFORMS Contact information at EDIRC.
Bibliographic data for series maintained by Chris Asher (casher@informs.org).