From Home Base to Swing States: The Evolution of Digital Advertising Strategies during the 2020 US Presidential Primary

• Author: NaLette M Brodnax,Piotr Sapiezynski
• DOI: 10.1177/10659129221078046
• Published date: 01 June 2022
• Date: 01 June 2022
• Subject Matter: Mini-Symposium: America in the 2020 Elections

Mini-Symposium: America in the 2020 Elections

Political Research Quarterly

2022, Vol. 75(2) 460–478

© The Author(s) 2022

Article reuse guidelines:

sagepub.com/journals-permissions

DOI: 10.1177/10659129221078046

journals.sagepub.com/home/prq

From Home Base to Swing States: The

Evolution of Digital Advertising Strategies

during the 2020 US Presidential Primary

NaLette M Brodnax

1

and Piotr Sapiezynski

2

Abstract

Political advertising on digital platforms has grown dramatically in recent years as cam paigns embrace new ways of

targeting supporters and potential voters. We examine how political campaign dynamics have evolved in re sponse to the

growth of digital media by analyzing the advertising strategies of US presidential election campaigns during the 2020

primary cycle. To identify geographic and temporal trends, we employ regression analyses of campaign spending across

nearly 600,000 advertisements published on Facebook. We show that campaigns employed a new strategy of targeting

voters in candidates’home states during the “invisible primary.”In contrast to earlier studies, we f‌ind that home state

targeting is a key strategy for all campaigns, rather than just for politicians with existing political and f‌inancial networks.

While all candidates advertised to their home state, those who dropped out during the invisible primary tended to spend

disproportionately more than the candidates who outlasted them. We also f‌ind that as the f‌irst wave of state caucuses

and primary elections approach, campaigns shift digital ad expenditures to states with early primaries such as Iowa and

New Hampshire and, to a lesser extent, swing states.

Keywords

political advertising, digital media, campaign strategies, Facebook

Introduction

In recent years, presidential campaigns in the United States

have shifted from traditional forms of advertising—such as

television, radio, and print—to digital media. Historically,

campaigns have spent the largest proportion of their ad-

vertising budgets on television commercials, as they are

most effective in reaching a large number of voters.

However, the proportion of ad spending devoted to tele-

vision has declined in each election year since 2008

(Cassino, 2017). Over the same period, the proportion spent

on digital advertising increased from 0.2% to 14.4% in 2016

and was projected to account for 28% of political ad

spending in 2020 (see Figure 1). In fact, the campaign of Joe

Biden, the winner of the Democratic nomination and general

election, spent 30.7% of its ad budget on Facebook and

Google ads alone (Center for Responsive Politics, 2020).

The strategic decisions that campaigns make about

advertising—where, when, and how to deploy ads—are

critical to understand given their inf‌luence on electoral

outcomes (West, Kern, Alger, and Goggin, 1995).

1

Digital advertising has a number of advantages over

television and other traditional media. Most importantly, it

allows campaigns to precisely target voters using a range

of tools made available to them by ad platforms. Cam-

paigns can choose the target audience based on their

location, age, and gender, as well as a number of interest-

based parameters such as inferred political alignment.

They can also target particular users based on their per-

sonally identif‌iable information (PII), such as email ad-

dresses, phone numbers or names, as well as promote ads

to other users similar to those whose PII they’ve obtained

(Martinez, 2018). By clicking, commenting, or sharing

ads, social media users provide campaigns with imme-

diate feedback on ads’ability to engage potential voters;

1

McCourt School of Public Policy, Georgetown University, Washington,

DC, USA

2

Khoury College of Computer Sciences, Northeastern University,

Boston, MA, USA

Corresponding Author:

NaLette M Brodnax, McCourt School of Public Policy, Georgetown

University, 37th & O St NW, Old North 406, Washington, DC 20057-

0004, USA.

Email: nalette.brodnax@georgetown.edu

the campaigns, in turn, can use it to more eff‌iciently al-

locate resources (Erdody, 2018;Kreiss, Lawrence, and

McGregor, 2018). Campaigns can advertise on digital

platforms with relatively small budgets, in contrast to

television advertising where budgets can run from hun-

dreds of thousands to millions of dollars. Due to their

lower cost, digital ads have served as an equalizing force

for long-shot candidates (Christenson, Smidt, and

Panagopoulos, 2014;Paolino and Shaw, 2003).

Despite dramatic growth in digital advertising, scholars

know little about how political campaigns leverage its

unique features. To address this gap, we consider two

questions. First, we investigate whether campaigns target

voters by location in a manner consistent with traditional

forms of advertising. Second, we investigate whether

campaigns’advertising strategies shift over time as they

consider the next immediate need. To answer these

questions, we examine Facebook advertising by US

presidential campaigns during the 2020 primary election

cycle. Facebook accounts for the largest share of digital

advertising due its ease of use and the size of its user base

(Erdody, 2018), and experiments demonstrate that ads

hosted on the platform can lead to increased political

participation (Haenschen and Jennings, 2019). Using the

Facebook Political Advertising Library (Facebook, 2019),

we analyze nearly 600,000 advertisements published by

26 Democratic presidential primary campaigns from 1

January 2019 through “Super Tuesday,”3 March 2020.

After that date only f‌ive candidates remained in the race

for the nomination. For each ad, the library reports the

estimated number of impressions—the number of times

that the ad appeared in users’feeds. In addition to

estimated impressions, the library reports the approximate

cost, as well as the locations and demographic charac-

teristics, such as age and gender, of users who viewed

each ad.

Research on campaign advertising strategies has uti-

lized a range of data sources, the most prominent being

television advertising archives (Fowler, Franz, and

Ridout, 2017;Goldstein, Niebler, Neilheisel, and

Holleque, 2011). These archives collect and encode

data on television ads across 210 designated market areas

(DMAs), including location, cost, and content. Since

some DMAs span multiple states, this data has limited

utility for analyzing advertising strategies by state. Be-

cause states award both primary delegates and general

electors, state-level outcomes are important for answering

questions about presidential campaign strategies. The

Facebook Ad Library provides the same level of detail,

but also allows us to compare state-level advertising by

each campaign over time.

We f‌ind that campaigns introduced a new dynamic

strategy: home state advertising. Early in the primary

season, campaigns spend a larger share of their budget in

the candidate’s home state where digital advertising helps

to raise money, signal viability, and build momentum. In

contrast to earlier studies showing that candidates’home

states allow them to tap into existing political and f‌inancial

networks, we f‌ind that all candidates advertised heavily in

their home states, regardless of the size of their existing

constituency or network. Candidates who dropped out

during the invisible primary invested much more heavily

in home state advertising than the candidates who out-

lasted them—despite an attempt, we argue, to elevate their

Figure 1. Online advertising has become more central to political advertising budgets, from below 1% in 2008 to 14.4% in 2016. It was

projected to reach 28% in 2020 and surpass the combined spending on political ads in press, radio, direct mail, telemarketing, and

others (Cassino, 2017).

Brodnax and Sapiezynski 461