An Algorithm for Allocating Sponsored Recommendations and Content: Unifying Programmatic Advertising and Recommender Systems

Edward Carl Malthouse*, Yasaman Kamyab Hessary, Khadija Ali Vakeel, Robin Burke, Morana Fudurić

*Corresponding author for this work

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

Retailing and media platforms recommend two types of items to their users: sponsored items that generate ad revenue and nonsponsored ones that do not. The platform selects sponsored items to maximize ad revenue, often through programmatic auctions, and nonsponsored items to maximize user utility with a recommender system (RS). We develop a binary integer programming model to allocate sponsored recommendations considering dual objectives of maximizing ad revenue and user utility. We propose an algorithm to solve it in a computationally efficient way. Our method is a form of postfiltering to a traditional RS, making it widely applicable in two-sided markets. We apply the algorithm to data from an online grocery retailer and show that user utility for the recommended items can be improved while reducing ad revenue by a small amount. This multiobjective approach unifies programmatic advertising and RS and opens a new frontier for advertising and RS research. We provide an extended discussion of future research topics.

Original languageEnglish (US)
JournalJournal of Advertising
DOIs
StateAccepted/In press - Jan 1 2019

Fingerprint

Recommender systems
Marketing
revenue
systems research
auction
Integer programming
programming
Revenue
market

ASJC Scopus subject areas

  • Business and International Management
  • Communication
  • Marketing

Cite this

@article{f6901369493c4f30b865349eee9a0c33,
title = "An Algorithm for Allocating Sponsored Recommendations and Content: Unifying Programmatic Advertising and Recommender Systems",
abstract = "Retailing and media platforms recommend two types of items to their users: sponsored items that generate ad revenue and nonsponsored ones that do not. The platform selects sponsored items to maximize ad revenue, often through programmatic auctions, and nonsponsored items to maximize user utility with a recommender system (RS). We develop a binary integer programming model to allocate sponsored recommendations considering dual objectives of maximizing ad revenue and user utility. We propose an algorithm to solve it in a computationally efficient way. Our method is a form of postfiltering to a traditional RS, making it widely applicable in two-sided markets. We apply the algorithm to data from an online grocery retailer and show that user utility for the recommended items can be improved while reducing ad revenue by a small amount. This multiobjective approach unifies programmatic advertising and RS and opens a new frontier for advertising and RS research. We provide an extended discussion of future research topics.",
author = "Malthouse, {Edward Carl} and Hessary, {Yasaman Kamyab} and Vakeel, {Khadija Ali} and Robin Burke and Morana Fudurić",
year = "2019",
month = "1",
day = "1",
doi = "10.1080/00913367.2019.1652123",
language = "English (US)",
journal = "Journal of Advertising",
issn = "0091-3367",
publisher = "M.E. Sharpe Inc.",

}

An Algorithm for Allocating Sponsored Recommendations and Content : Unifying Programmatic Advertising and Recommender Systems. / Malthouse, Edward Carl; Hessary, Yasaman Kamyab; Vakeel, Khadija Ali; Burke, Robin; Fudurić, Morana.

In: Journal of Advertising, 01.01.2019.

Research output: Contribution to journalArticle

TY - JOUR

T1 - An Algorithm for Allocating Sponsored Recommendations and Content

T2 - Unifying Programmatic Advertising and Recommender Systems

AU - Malthouse, Edward Carl

AU - Hessary, Yasaman Kamyab

AU - Vakeel, Khadija Ali

AU - Burke, Robin

AU - Fudurić, Morana

PY - 2019/1/1

Y1 - 2019/1/1

N2 - Retailing and media platforms recommend two types of items to their users: sponsored items that generate ad revenue and nonsponsored ones that do not. The platform selects sponsored items to maximize ad revenue, often through programmatic auctions, and nonsponsored items to maximize user utility with a recommender system (RS). We develop a binary integer programming model to allocate sponsored recommendations considering dual objectives of maximizing ad revenue and user utility. We propose an algorithm to solve it in a computationally efficient way. Our method is a form of postfiltering to a traditional RS, making it widely applicable in two-sided markets. We apply the algorithm to data from an online grocery retailer and show that user utility for the recommended items can be improved while reducing ad revenue by a small amount. This multiobjective approach unifies programmatic advertising and RS and opens a new frontier for advertising and RS research. We provide an extended discussion of future research topics.

AB - Retailing and media platforms recommend two types of items to their users: sponsored items that generate ad revenue and nonsponsored ones that do not. The platform selects sponsored items to maximize ad revenue, often through programmatic auctions, and nonsponsored items to maximize user utility with a recommender system (RS). We develop a binary integer programming model to allocate sponsored recommendations considering dual objectives of maximizing ad revenue and user utility. We propose an algorithm to solve it in a computationally efficient way. Our method is a form of postfiltering to a traditional RS, making it widely applicable in two-sided markets. We apply the algorithm to data from an online grocery retailer and show that user utility for the recommended items can be improved while reducing ad revenue by a small amount. This multiobjective approach unifies programmatic advertising and RS and opens a new frontier for advertising and RS research. We provide an extended discussion of future research topics.

UR - http://www.scopus.com/inward/record.url?scp=85071339277&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85071339277&partnerID=8YFLogxK

U2 - 10.1080/00913367.2019.1652123

DO - 10.1080/00913367.2019.1652123

M3 - Article

AN - SCOPUS:85071339277

JO - Journal of Advertising

JF - Journal of Advertising

SN - 0091-3367

ER -