Improving banner ad strategies through predictive modeling

Abstract

Purpose

The purpose of this exploratory, data-driven study is to identify the optimal banner advertising strategies for achieving different business metric goals, such as effective cost per activity, via unique predictive modelling methods.

Design/methodology/approach

The k-fold cross-validation method is used to build predictive models to analyze 18,956 online banner advertising records.

Findings

Banner ads with high visual complexity and attractive offers tend to draw users to participate in online activities, whereas voluntary banner ads with low visual complexity tend to draw user clicks. Further, banner ads with lower visual complexity tend to have lower costs. Finally, the third quarter of a year is the most important period for online advertising campaigns in terms of achieving the optimal effectiveness and cost for running internet banner ads.

Research limitations/implications

As only visual and temporal characteristics of internet banner ads are covered in this study, future research should concentrate on the specific language within each banner ad message. Further, this study does not specifically tie internet-specific metrics, such as activities, costs and clicks to business metrics, such as revenue and profit.

Practical implications

Advertisers can use the findings from this study to create an effective and cost-efficient banner advertising strategy. Specifically, firms should use larger banner ads with features and offers, advertise at the end of the year and use caution with rich media expandable banners and banner ad videos as they can significantly increase costs.

Originality/value

This is one of the first exploratory studies to use the k-fold cross-validation method to build predictive models to identify visual and temporal factors that significantly impact the effectiveness and cost of internet banner ads.

Citation

Obal, M.W. and Lv, W. (2017), "Improving banner ad strategies through predictive modeling", Journal of Research in Interactive Marketing, Vol. 11 No. 2, pp. 198-212. https://doi.org/10.1108/JRIM-08-2016-0092