Second thoughts on digital first: Exploring the development of election campaigning among Swedish political parties, 2010–2022

Studies of the use of digital (particularly social) media during election campaigns have become a rapidly expanding research field in media and communication studies in recent decades.

When we make references to social media in this article, we refer to Web 2.0 tools that offer a higher level of interactivity than other forms of digital media (Dimitrova et al., 2014). Examples of such are, for example, Facebook, Instagram, Twitter (now known as X), and YouTube. When we use the broader term digital media, we thus include both social media and less interactive Web 1.0 platforms such as e-mail, SMS, MMS, and websites (cf. Carr & Hayes, 2015).

As the digital development of election campaigns is less analysed from an inside political-party perspective, we lack thorough knowledge about the extent to which political parties embrace the new opportunities offered by digital technology, how digital elements are integrated into existing campaign structures and strategies, and which factors are decisive when party campaigning goes digital.

The observations of political-party communication content are relevant as overviews of campaign developments, but they do not provide full information about party considerations for expressing these messages or themes at any given time. This requires a more comprehensive approach where strategies and practices are also analysed, or, as Jungherr (2016a: 374) described in his study of German campaign managers:

To fully understand the impact of digital tools on campaigns requires that scholars move away from simply analyzing the political content campaigns post online toward a focus on the embeddedness of digital tools in organizational structures and practices.

Previous research has also addressed the need to “further explore some of the systematic drivers and impediments to social media embeddedness in various campaign contexts” (Veneti et al., 2022: 62).

Against this backdrop, this article offers a longitudinal perspective on communications during election campaigns from a political-party perspective, where strategic considerations about digital communication tools and platforms are analysed and compared across time. The analysis is grounded on the concepts of hybridisation and data-driven campaigning, where digital technology tends to play a central role and where digital and traditional distinctive campaign features are assumed to be integrated in line with party strategic purposes.

Our aim is to examine political parties’ strategic considerations, including how they assess the importance of digital media in comparison with traditional campaign communication tools. Empirically, the study is based on a longitudinal analysis of four national election campaigns in Sweden during 2010–2022, to investigate to what extent and how digital media play an integral part in political parties’ strategies to be as electorally successful as possible. Both pre- and post-election party surveys and personal interviews with campaign managers from all parties in the parliament have been collected to map how and why traditional and digital campaign communications have been implemented and blended.

The selection of Sweden as a case was motivated by technological, political, and pragmatic reasons. Technologically, Sweden is one of the most digitalised countries in the world in terms of broadband penetration and level of digital and social media use (Internetstiftelsen, 2022). Politically, Sweden is a suitable case because it provides a well-established party system with old and institutionalised parties. Finally, political parties in Sweden are, in comparison with many other countries, relatively open and transparent about campaign strategies and objectives, which allows this study, unlike most other previous research, to be based on empirical analysis of the strategic considerations political parties make rather than a study of what and how parties communicate on social media.

The article is structured as follows: In the next section, we discuss the transformation of election campaigns and introduce the theoretical concepts of hybridisation of campaigning and data-driven campaigning. We then discuss methods and data and the results from the qualitative and quantitative analyses. In the final section, we answer the research questions and offer a discussion about how digitalisation and hybrid, data-driven election campaigns tend to develop over time.

Election campaigns in most democracies are in a process of transformation, and previous research largely confirms that they are gradually becoming more digitalised and professionalised (Bolin et al., 2022; Davis, 2019; Grusell & Nord, 2020). Both political parties and individual candidates use a variety of digital tools to target segments of the electorate and reach out more efficiently. The digital trend is further emphasised by the fact that news media election reporting is available on mobile and digital platforms, and social media play a significant role in digital political communications (Tambini, 2018; Taras & Davis, 2022).

Technological development is an important driver of the evolution of how political parties design and execute their election campaigns. From this perspective, there are reasons to expect changes in political parties’ campaign strategies to take place both more quickly and more comprehensively (Chadwick & Stromer-Galley, 2016; Russmann, 2022). However, from a party organisational perspective, there are reasons to expect that changes are implemented incrementally and stepwise. It has even famously been argued that parties are simply conservative organisations that are resistant to change (Panebianco, 1988).

In this cross-pressure between rapid digital developments and resilient political-party structures, contemporary election campaigns emerge. It is plausible to believe that campaigning is open to – and sometimes driven by – technological innovations intended to result in electoral successes, but at the same time also careful to preserve previous campaign features that have proven to be important. Thus, it is reasonable to argue that election campaigns today are hybridised to varying degrees of mixtures of older and newer communication tools. Such an idea of hybridisation of campaign practices has grown stronger over time (Chadwick, 2013; Plasser & Plasser, 2002; Schmitt-Beck, 2007).

In the hybridisation model, digital and social media are expected to interact with existing campaign practices rather than replace them (Karlsen, 2009; Schmitt-Beck, 2007). The adaptation of new campaign practices depends on a range of contextual factors, such as the political system, political culture, election laws, media system, and electoral behaviour, just to mention a few. From this perspective, the question is not so much whether digital media replace previous communicative platforms, but what role different digital media play in relation to a wide range of new and old media platforms in the greater political communicative system (Steffan & Venema, 2020).

The entry into such a hybrid and fluid communication environment has also been characterised as a more fundamental structural transformation, or a “new era” of election campaigning, where a key component for understanding contemporary campaigns is that they have become more driven by data and platform convergence considerations (Bossetta & Schmøkel, 2023; Kefford et al., 2022; Römmele & Gibson, 2020). In this perspective, digital technology plays a central role in campaign organisation and operations, voter communications have a more networked approach, there is a more personality-based understanding of the electorate, and campaigns are more internationalised and open for external influence and propaganda (Römmele & Gibson, 2020).

Empirical observations support the idea of data-driven campaigning: A recent comparative study of six countries (Australia, Canada, Germany, the Netherlands, the US, and the UK) found that digital technology was of central importance in all countries for collecting data and tailoring messages to target voter groups. However, true data-driven campaigning was the most distinctive feature only in the American context, while in the other countries, political parties often lacked the capacity to execute data-driven campaigning on a regular basis. Consequently, day-to-day campaign decisions and practices relied on other principles (Kefford et al., 2022). An important recommendation from this comparative study is to move away from a media-centric account of data-driven campaigning and focus less on the impact of new media developments and instead observe more carefully how campaign organisations are adapting to the developments they are confronted with (Kefford et al., 2022: 456).

Another study addressing country-specific conditions for data-driven campaigning concluded that canvassing in German election campaigns seemed to be less data-driven than in the US, mainly because of diverging political culture, privacy laws and regulations, as well as limited party resources in terms of money and number of staff and volunteers (Kruschinski & Haller, 2017).

Longitudinal studies of digital campaigning in Sweden from a political-party perspective are rare; however, studies of single election campaigns indicate that the probability of data-driven campaigning depends more on the resources and size of a party than on party ideology. In the 2014 Swedish elections, the three biggest parties (the Social Democrats, the Moderate Party, and the Sweden Democrats) were most present on social media and paid more attention to new digital technology (Larsson, 2017). A cross-national study of right-wing populist platform strategies in the Nordic countries – examining elections in Denmark, Finland, Norway, and Sweden during 2017–2019 – found that the Swedish party in the comparison (the Sweden Democrats) held the strongest position on all analysed platforms (Schwartz et al., 2022).

To conclude, hybridisation theory and data-driven campaign theory indicate that digital media have become an integrated part of election campaigns, but it remains largely unclear to what extent they are used, which factors decide their adaptation, and how they relate to the overall communicative context (cf. Karlsen & Enjolras, 2016). While there is widespread knowledge about the existence of hybrid and data-driven campaigning during contemporary elections, less is known about how the two models develop in a longitudinal perspective, and what strategic considerations political parties make in relation to this over time.

With this study, we intend to fill these gaps in two ways. First, we map how hybrid and data-driven campaign processes are related to political parties’ strategic considerations in terms of the planning and implementation of election campaigns. Second, we examine how hybrid campaigning and data-driven campaigning have developed over time. By following hybrid and data-driven campaign processes across time and space, this article contributes new knowledge to our understanding of how digitalisation processes evolve during contemporary election campaigns. Our empirical analysis of the implementation and integration of digital campaign communications by Swedish political parties in general election campaigns during 2010–2022 is guided by two overarching research questions:

RQ1. How can political parties’ strategic considerations pertaining to digital communication be described?

RQ2. How do political parties perceive the importance of digital communication in election campaigns?

In the first part of the results section, we report the thematic analysis of the interviews to capture political parties’ strategic considerations about digital communication during election campaigns. In the second part, we move on to the analysis of the survey and how political parties perceive the importance of digital communications in elections.

Currently, Swedish political parties are using most of the digital channels available, and in hindsight, the process for this has evolved rapidly, though not without obstacles. Within merely a decade of the introduction of digital media such as blogs, YouTube, Facebook, Twitter, Instagram, Snapchat, and TikTok, the range of channels to communicate on has increased dramatically. From the beginning, political parties have been interested in how to incorporate these digital media in their campaign practices, but it took time to find the right tone and way of using them. Over time, however, digital communications, and especially social media, became less complicated to work with and was conducted on many platforms – traditional as well as new. The journey was not distinct from the beginning and met several challenges, so we start by describing how the digital journey has evolved. In the thematic analysis, we identified two overarching themes: the political parties’ digital journey and what digital lessons were learned.

In the second empirical part, we examine the importance of digital communication channels. To assess the extent to which there are general trends, we use the survey responses and compare the mean scores of the digital communication channels with the mean scores of the remaining communication channels in the survey. Positive values should be interpreted as above average importance.

As shown in Figure 1, digital media are on average less important than the other communication channels included in the survey. Even though the difference in importance is quite small, digital media are perceived as less important than other forms of communication in each of the four election campaigns. Moreover, there is no clear trend. The difference in importance decreases from 2010 to 2014 but increases again in 2018. In the most recent election campaign, digital communication channels continue to be significantly less important than the other communication channels included in the survey. Thus, in contrast to recent research that assumes that the Internet and social media are becoming more important for political communication (Chadwick & Stromer-Galley, 2016; Taras & Davis, 2022), our data do not suggest a relative increase in the importance of digital media compared with traditional means of communication. Importantly, however, this is the picture that emerges from an analysis of aggregate data. To further assess the role of digital media, we also need to delve deeper to see what patterns can be revealed when individual communication channels are compared.

FIGURE 1

Table 2 shows the average level of importance for all individual communication channels in the survey for the four election years, as well as the average level of importance for all years together and the change in importance from 2010 to 2022. Positive values indicate that the individual channel is considered above average in significance, while negative numbers indicate the opposite.

The picture that emerges is quite ambiguous. While some digital media are indeed among the more important channels, others are among the less important, according to party informants. Moreover, while social media generally are considered somewhat more important than other forms of digital media, the picture is not clear-cut.

Roughly, digital media can be divided into four groups in terms of their relative importance. Facebook and party websites are consistently important, having been considered relatively important by party informants in all four election campaigns. YouTube and Instagram are increasingly important, as they were relatively unimportant at first but gained in importance. Together with Facebook, these two social media were among the most important among all items covered in the survey in the most recent election of 2022. A third group consists of blogs and Twitter and can be described as decreasingly important, as they were initially considered relatively important but have gradually lost importance over time. Finally, communication via SMS, MMS, and e-mail during the election campaign can be described as consistently unimportant, as these were considered relatively unimportant in all four surveyed election campaigns.

Since Snapchat and TikTok have only been included in two surveys and one survey, respectively, it is too early to place them into any of the four groups; thus far, however, both have been considered relatively unimportant by the parties. There is, however, indications that the parties have misjudged the importance of these platforms, as one prominent topic in the postelection debate was about how TikTok has become one of the most important sources of political information for young voters (Bolin et al., 2022).

Diverging trends can also be observed among the other communication channels included in the surveys. Traditional media channels, such as political shows on television, television news, radio news, and newspapers, are all among the most important communication channels on average. At the other end of the spectrum, we find other forms of media; party informants ranked, for example, both magazines and television entertainment as among the least important. Direct forms of communication – election posters, election cabins, and door-to-door canvassing – are relatively unimportant, as they are all below 0 on average. There is, however, an increasing trend in the level of importance from 2010 to 2022 for election posters, door-to-door canvassing, and direct marketing.

To conclude, what emerges is thus a mixed picture, where parties seem to combine various forms of communication on different platforms and outlets rather than focusing solely on digital media, traditional media, or direct communication. Moreover, while social media are generally considered more important than less interactive “web 1.0” digital media channels, this trend is not without exceptions. Party websites are still relatively important whereas Twitter, for example, is among the least important. In line with Chadwick’s (2013) concept of a “hybrid media system”, parties prioritise a range of different media platforms to create, steer, and respond to a flow of communication. No clear general trends can be easily identified, and digital media do not seem to replace older forms of communication. Our results thus align with recent research on platform convergence and suggest that political actors run successful campaigns by creating a synergy between traditional media channels and social media (Karlsen & Enjolras, 2016).

This study of the digitalisation of Swedish election campaigns from a political-party perspective was guided by two main research questions. First, we examined the political parties’ strategic considerations pertaining to digital communications. The results from the interviews showed that party adaption to new communication channels has been an evolutionary process, and that the political potentials of new communication tools have generally been overestimated. Party perceptions of digital and social media varied in the campaigns examined. An initial curiosity and enthusiasm have been gradually replaced by a more pragmatic approach and an increased awareness of the complexity of communication in a more fragmented media landscape.

There has been a widespread consensus among the political parties about the need to think strategically when approaching the new communication tools. The interactive and agenda-setting potentials, as well as the new abilities to target specific segments of the electorate, have encouraged the parties to integrate the new tools into their campaign strategies. At the same time, the parties have realised that effective campaign work on digital platforms demands significant resources, and that recurrent public debates about the risks with disinformation and hate speech may undermine the credibility of the new platforms. In line with results from previous studies of the implementation of data-driven campaigning (Kefford et al., 2022; Kruschinksi & Haller, 2017; Larsson, 2017), Swedish political parties do not, so far, let their campaigns be completely decided by existing digital technology opportunities.

The second research question asked how the political parties perceived the importance of digital communications during election campaigns. The results of the party surveys indicate that digital media were generally perceived as less important than other communication channels during the analysed period. In contrast to previous research (Russmann, 2022), no clear general trend of higher appreciation of digital communication channels could be observed across time, even though parties’ use of digital media had become more strategically oriented over time.

Furthermore, the analysis of the survey results suggests that digital media channels could be roughly divided into four groups in terms of their relative importance. Some outlets, like Facebook and party websites, were consistently important during the twelve years. Others, like YouTube and Instagram, were increasingly important. Finally, Twitter and blogs were considered decreasingly important, while e-mail, SMS, and MMS were perceived as more or less unimportant in every election campaign analysed. Overall, the results confirm that the political parties continuously try to create a successful synergy based on instant evaluations of the efficiency of the different communication channels at hand.

The results of our study shed some new light on the digitalisation of contemporary election campaigns in a longitudinal perspective. Swedish political parties do not simply replace traditional communications with digital channels as they become available, but instead gradually integrate them into their campaign structure and strategy if they have the resources to do so and if it works well within the overall campaign context. Some channels introduced become permanent important tools, while others never achieve an important place in the party campaign toolbox. The process of adaptation of digital campaigning is not linear, but rather back and forth, as party perceptions of the importance of communication channels vary across time.

The findings of Swedish election campaign developments can be related to international perspectives. The most strategically developed and innovative election campaigns have so far taken place in the US, and many studies are focusing on the American campaigning environment. However, the situation in the US differs from national contexts in many other parts of the world. In countries with multiparty political systems, party-centred campaigns, media systems with stronger public service broadcasting, and less liberal campaign financing regulations, campaigns can be expected to develop in less predictable ways.

The fact that Swedish political parties, regardless of size and ideology and in contrast to developments in other countries, do not thus far prioritise digital over traditional communication may have different underlying explanations. First, party approaches to the new communication channels have been characterised by both unrealistic expectations and uncertainty about their possible political potentials. Digitalisation has not yet proved to be a key to electoral success. Second, the national context of election campaigns in Sweden matters. News media, and particularly television, have remained central in importance despite the rapid digital transformation. Traditional direct communications such as canvassing and advertising are also considered relatively important over time. Thus, all Swedish political parties must try to find a reasonable balance between the assumed usefulness of new campaign communications and the existing experiences of traditional communications. In such deliberative processes, digital does not always come first.