Digital Campaigning: The Rise of Facebook and Satellite Campaigns

Studies of digital campaigning have revealed substantial change in the nature of political campaigns. Tracing the rise of email, party websites, social media, online videos and gamification, scholars have shown how, since the 1990s, parties have become heavily dependent on digital technology (Gibson, 2015). In this chapter we focus on two elements of the 2017 digital campaign: Facebook advertising and what we term ‘satellite campaigns’. Whilst resisting claims of revolution and transformational change (Kreiss, 2010, Williamson, Miller and Fallon, 2010) we nevertheless argue that these digital practices have important implications for parties’ organisational structures, practices and behaviour, as well as for public expectations of campaigning. Through this analysis we contend that the 2017 general election provides further evidence that ‘digital media are reconfiguring party-related engagement’ (Vaccari and Valeriani, 2016, p. 295), and agree with Gibson (2015, p. 191) that by ‘chaf[ing] against embedded organisational routines and norms’ these developments challenge established understandings of parties’ campaign strategies.

In examining digital, we adopt an expansive definition of the term. Alongside an interest in social media and party websites, we also examine the organisational digital infrastructure on which parties rely. Including digital databases, canvassing systems, online phone banks, and email lists, digital infrastructure is pivotal to parties’ diverse campaign activities by enabling participation through the reduction of resource costs. In the analysis that follows, we employ this expansive conception of digital to consider developments within the Labour and Conservative parties, using these examples to illustrate wider emerging trends.

By April 2017, few were predicting an early general election, but in the days that elapsed between Theresa May’s surprise announcement on 18 May and the vote on 8 June, parties across the spectrum exhibited formidable online and offline campaigns. In the digital realm, Twitter, Facebook, YouTube, Snapchat, and Instagram all played a part in the campaigns (Dutceac Segesten and Bossetta, 2017) with numerous graphics, videos and messages shared online. While all parties were active on these platforms, emerging analysis has demonstrated the degree to which Labour, and particularly Jeremy Corbyn, dominated support on these platforms (Cram et al., 2017; Dean, 2017; Shephard, 2017). Organisationally, digital proved key to volunteer mobilisation, voter identification activities, and message targeting. Particularly prominent within the wider picture of social and digital media use were two elements of the digital campaign: parties’ use of Facebook advertising and the role of what we term ‘satellite campaigns’ facilitated by non-party intermediary organisations such as Campaign Together and CrowdPac. Considering these two developments in detail, we argue that such changes represent important new evolutions in political campaigning, and raise interesting questions for parties, the public and our expectations of political campaigning.

1. Facebook advertising—the new normal?

Parties’ use of Facebook advertising was heralded as a key component of the 2017 campaign (Bakir and McStay, 2017; Walsh, 2017; see also Bond, 2017; Ward, 2017; Waterson, 2017). Using this tool, parties across the spectrum targeted content at specific groups of voters. Drawing on demographic data such as age, postcode, religion, and gender, combined with indicators of users’ interests, parties were able to identify those with, say, a passion for cycling, international travel, or beer, and use these interests to filter messages about the environment, foreign affairs or taxation. Parties were therefore able to identify electorally significant voter groups, such as women over 65 in marginal constituencies, and tailor messages to their interests and ideas in attempts to win appeal. The uptake of this tool was especially notable in the two main parties. Labour, in particular, embraced Facebook advertising (Waterson, 2017), investing heavily at a national level in adverts designed to promote electoral registration, but also creating a new organisational tool, Promote, which allowed local parties to target their own Facebook adverts. Similarly, the Conservatives invested in adverts promoting Theresa May and questioning the leadership credentials of Jeremy Corbyn. Reportedly, these campaigns saw Labour and the Conservatives spend over £1 million each on the platform, although formal electoral commission figures have yet to be released (Bond, 2017). Due to their targeted nature, capturing the range of adverts is exceedingly challenging. Initiatives such as Who Targets Me allow users to track the adverts targeted at them (via web-browser extension) to offer some insight into the number, form, and focus, of party adverts; however, these data are yet to be analysed in full.

Given the evidence that seeing a political message on a friend’s page can affect voting behaviour (Bond et al., 2012), Facebook provides parties with a range of new capacities that can enhance their campaigns. As a social media platform, Facebook allows parties to connect with voters where they are, building on existing networks through sharing, comments, and reactions. In this way, an advert targeted to one voter may be liked and shared, signalling to friends and acquaintances that voter’s views and affiliation. This kind of sharing enables messages to be organically disseminated (Dutceac Segesten and Bossetta, 2017) and, as previous studies have shown, voters are increasingly comfortable sharing election related information online (Aldrich et al., 2016, p. 174).

Whilst the prominence of Facebook in 2017 may suggest the emergence of a new campaign tool, it is important to recognise that the use of targeted social media advertising in the UK is part of a developing trend. This technique was prevalent during the EU referendum the previous year (Cookson and Gordon, 2016) and at the 2015 general election the Conservatives embraced Facebook advertising, declaring a spend of £1.2 million to the Electoral Commission on this platform alone (Electoral Commission, 2016, p. 29). Reflecting on the success of the 2015 digital campaign, interviews with Conservative Party strategists have demonstrated that Facebook in particular was viewed as ‘the best place to advertise’ (Interview with Conservative Party official, February 2017),¹ an idea that has infused the strategies of other parties in 2017. Moreover, despite the increased prominence of Facebook as a campaigning space, the targeting it enables is by no means new. Parties have long focused on identifying and targeting their vote with the aim of refining communication strategies and identifying where the vote needs to be mobilised on election day (Whiteley and Seyd, 2003). However, historically, parties have been restricted by data protection laws to using the electoral roll (a list of everyone registered to vote), the marked register (a list of each elector's voting history), commercially available data (from private companies) and their own canvassing databases to target voters. Facebook is distinctive in offering a new source of voter information to parties that reflects voters’ interests and social preferences, whilst also providing the platform for communication.

From this perspective, we argue that rather than signalling radical change, the use of Facebook in 2017 represented a ‘normalized revolution’ (Wright, 2012), as it has adapted and extended party activities, whilst not radically changing what it is that parties do. In reaching this conclusion, we nevertheless argue that parties’ use of Facebook has important implications for public perceptions, control, and resource, that warrant further discussion.

First, in regard to public perceptions, Facebook allows parties to target content at highly localised audiences with greater precision than was previously possible (Aldrich et al., 2016; O’Brien, 2015; Council of Europe, 2017, p. 11). This ensures that voters hear about topics they are likely to be interested in/receptive to, which, given that social media is an increasingly important source of news and information for many voters (Miller, 2016), has the potential to yield significant effects (Marengo, 2013). Whilst companies such as Amazon and Google have been utilising targeting techniques to filter desirable content to users for years, parties’ use of this data is less familiar,² and the consequences of such targeting are unknown. Some scholars have already theorised that ‘unsolicited messages are likely to be regarded as more intrusive than a “cold call” to a landline or flyer posted through the mailbox’ (Aldrich et al., 2016, p. 166). Whilst the use of Facebook data is often not as sophisticated as may be presumed, the degree to which parties should be able to access and, indeed, purchase additional information about voters raises potential concerns. Asides from issues of resource inequality (discussed further below), the idea that private information is being used by actors in the public realm to further their own electoral success raises issues of transparency and appropriate democratic behaviour. The norms here are not absolute, but research indicates that politicians are held to higher standards than other people (Allen and Birch, 2015, p. 71), suggesting that commercially-accepted practices may not be tolerated to the same degree in the political realm. In this context, questions emerge around how parties can and should use data to connect with voters, questions that the Electoral Commission need to consider when re-examining existing regulations.

Second, we argue that whilst the unique attributes of Facebook offer parties advantages in terms of connectivity and reach, this platform also raises issues of control. While Facebook can be used to target official party adverts, it is also a forum in which unofficial campaign interventions can be made in an untargeted manner. While the Conservatives developed videos comparing their position on taxation to Labour, and Labour made videos citing 10 reasons to vote Labour, many other videos from ‘unofficial’ sources could also be found. From Cassetteboy’s remixes of political speeches, to memes mocking parties’ manifesto positions (see for example The Metro, 28 July 2017), Facebook provides a platform for a range of different political interventions. This raises a series of questions about the degree to which parties can exercise control and maintain campaign coherence, but also about how targeted campaigns intersect with other content. To take one example, during the 2017 election, Momentum generated high-profile Tory attack advertisements, one of which, set in 2030, depicts a young girl asking her Conservative-voting father if he hates her, to which he replies, ‘Obviously!’ Shared entirely on digital platforms including Facebook, Momentum activists claim the video was watched over 5.4 million times in two days (Peggs, 2017).³ This content had huge reach, but the stylistic approach is unlikely to have ever been sanctioned as part of an official Labour campaign, and it may be that many voters being targeted by official Labour advertisements were put off by the tone. The relationship between official and unofficial material, and the interactions between targeted and untargeted social media material are therefore far from clear, but they suggest a tension between parties’ desire to execute targeted campaigns and their capacity to do so on an open platform such as Facebook.

Third, Facebook raises issues of resource. On one level, the implications are organisational. Facebook demonstrates that digital has the potential to greatly reduce resource costs in terms of the efficiency of disseminating political campaign material, but also shows that there is a trade-off with the organisational capacity required to utilise digital tools effectively. As Labour’s experience in 2017 shows, parties’ use of digital requires investment not only in the cost of adverts, but also in the skills base of activists and organisational software. The tool Promote, developed by Labour’s Digital Transformation Team, allows local parties to identify their own target voters and deploy appropriate adverts, yet parties’ ability to design text and graphics likely to win attention and be shared is by no means guaranteed. As one digital consultancy company indicated, social media ‘has to give people something they cannot get elsewhere, and it needs to be designed for the environment it is appearing in. Otherwise, you’ve just made another trivial but terrible contribution to Sharemageddon’ (DigitalsLBi, 2015, p. 4). Ensuring that party activists have the capacity to generate attractive media poses a significant challenge to parties and may lead to a divide between ‘digitally native’ activists (Nielsen, 2013) and those lacking digital skills. As this divide is closely linked to generational profile, the Conservatives could be at a natural disadvantage when it comes to both their activist pool and their wider support base (Bale, 2017). This suggests that while parties face common challenges, these will be manifested in different ways depending on party context and culture.

At another level, Facebook also reveals issues of resource inequality. Whilst parties’ financial capacities have often been unequal—restricting their capacity to produce leaflets or commission campaign billboards—on Facebook, parties’ differing ability to buy advertising space raises important concerns about fairness, equality and transparency (see Norris, 2012). Whilst Labour and the Conservative Party used Facebook adverts extensively, other parties such as the Greens did not have the financial capacity to devote extensive resource. Moreover, it appears that, in certain marginal constituencies, only those parties willing to pay inflated prices for advertising space were able to promote their messages to voters (Cadwalladr, 2017). The importance of money for success in this realm, and the lack of transparency around how and why content appears in people’s Facebook feeds, raises issues about the fairness of elections.

These issues have begun to be noted by campaign regulators but, as yet, regulations have not been fully adapted to reflect the realities of digital campaigns. The Council of Europe therefore recently noted that ‘The Internet and new communications technologies undermine the ability of existing regulation to maintain a level playing field in electoral communication between new and established, rich and poor, corporate and civil society campaigns’ (2017, p. 2), and yet the Electoral Commission in the UK has only partially adapted for online campaigns. Hence, while recommending that organisations include an imprint on their online materials (Electoral Commission, 2017, p. 11) formal regulations have not been fully adapted in the UK to take account of issues such as campaign funding and political transparency.⁴

Parties’ use of Facebook therefore raises a series of questions about the acceptable use of personal data, parties’ control of election campaigns and the role of money and resource in elections. These questions have important consequences for the perceived fairness of elections, regulation and public tolerance of different kinds of campaigning intervention. As Facebook becomes a permanent part of the electoral landscape, these issues will only grow in pertinence.

2. Satellite campaigns

A second development in the digital sphere concerns the increased visibility of digital infrastructure offered by non-party organisations to encourage voting and campaigning. Though evident to different degrees, with greater activity around the Labour Party as opposed to the Conservatives, these organisations were seen to mobilise new activists and campaigners to parties’ causes. Innovations such as Momentum’s ‘My Nearest Marginal’ App, fundraising sites such as CrowdPac and campaigning hubs like the Progressive Alliance or Campaign Together were seen to empower and connect individuals to contribute to electoral campaigns via non-traditional routes. Organisations beyond parties were identifying, mobilising and organising citizens to deliver leaflets, canvass voters, and organise on- and offline. This development represents a distinctive and important shift in campaigning. It suggests that, in addition to Whiteley and Seyd’s categories of the central party campaign, centrally coordinated local campaigns, and purely locally directed campaigns (2003, p. 638), we can also identify campaigns originating beyond party structures and control—those termed here ‘satellite’ campaigns.

Satellite campaigns can be supported by a range of different organisations, making it useful to refer to Edwards’ (2006, pp. 8-9) notion of ‘democratic intermediaries’. Edwards outlines three types of intermediary that we apply to describe non-party organisations operating during the 2017 election. First, there are preference intermediaries, organizations that articulate and aggregate political demands and in 2017 were evident in the form of Momentum and Grime4Corbyn. Second, information intermediaries are seen to provide users with political information and details on voter registration; at the latest general election platforms like GE2017, Rize Up, and Turn Up fitted these criteria. Third, interactional intermediaries facilitate political participation, capturing tactical voting platforms such as Swap My Vote, tactical canvassing networks such as Campaign Together, and the crowdfunding and campaign-match tool CrowdPac. Whilst some organisations exhibit functions in multiple categories—Momentum, for example, could also be classified as informational and interactional—this framework demonstrates the different ways in which campaigning initiated beyond the official party campaign can occur.

Whilst affiliate organisations such as trade unions, business organisations, and community groups, have long provided an additional resource for parties’ electoral campaigns, the capacities of digital appear to have altered previous practice. Rachel Gibson has highlighted the capacity of digital technology, and specifically social media, to alter the power relations between citizens and central party headquarters. Tracing the rise of ‘citizen-initiated campaigning’, she describes the emergence of a ‘more devolved or “citizen-initiated” approach to campaign organization’ (Gibson, 2015, p. 183). The creation of tools by party candidates and teams that enable citizens to canvass voters on remote phone bank applications, raise money online, organise campaign events or disseminate party materials on social media are seen to enable ‘autonomous action and tactical control of campaign operations at the local level on a scale that was not possible in the pre-digital era’ (Gibson, 2015, p. 187). Numerous other scholars have picked up on this theme: Vaccari and Valeriani (2016, p. 306) have argued that social media are helping ‘new digital foot soldiers to emerge and allow existing members to expand their repertoires’, whilst Lilleker and Jackson (2010, pp. 74-75) have discussed the internet's ability to encourage ‘individual production and user-generated content’, creating supportive material, endorsing campaigns, and sharing campaigns through online networks (see also Chadwick and Stromer-Galley, 2016).

The idea of satellite campaigns mirrors these themes, but extends them by recognising the increasing importance of intermediary, unofficial, organisations beyond parties that facilitate and promote campaigning activities. A campaign can be classed as satellite when vote-seeking activism is primarily driven by intermediary organisations without the control of a party. The rise of digital media platforms does not determine that satellite campaigning will take place; however, it does greatly facilitate it. Satellite campaigns have the capacity to challenge ‘the professionalized top-down approach that has dominated post-war elections, particularly over the past three decades’ (Gibson, 2015, p. 183). However, these organisations should not be seen as a threat to parties as institutions because, primarily, they are not vote-seeking themselves, but also because they remain reliant on party infrastructure and activity. For instance, Campaign Together lacks its own canvassing system and instead organises by directing new activists towards existing party-led campaigns they identify in key marginal seats—the organisation brands itself as a part of a progressive alliance united by an aim to ‘stop the Tories’. Utilising digital media, intermediary organisations help to bring together, train, mobilise, and inspire individuals who may not engage through traditional (and often staid) party structures and have the potential to provide a considerable additional resource for parties. As a regional organiser of Campaign Together reflected:

I think why people got involved with campaign tools like Campaign Together and what Momentum were offering is because a lot of people, and I heard this from talking to people, were similar to me in that they were intimidated to go to [party] meetings or they didn’t enjoy them and they found it hard to get involved and they wanted to—this felt like something autonomous or indirect. (Interview with regional organiser, July 2017)

These bodies innovatively utilise important resources in the form of email lists, digital media presence, and organisational tools (such as Slack, WhatsApp and Facebook) that help to get people involved. This potential is significant when considering the principle-agent problem parties usually face when using members (who often lie at the ideological extremes) to communicate with voters. As Enos and Hersh’s work (2015) has shown, parties’ reliance on members that are unrepresentative of the general public can prove counterproductive in attempts to campaign. By drawing on the energies of citizens who may not feel sufficiently enthused to join a party, but who may nevertheless share party values, satellite campaigns can provide parties with a wider set of advocates, who may be better placed to articulate their appeal. Digital therefore helps to enable the transition back and forth between being a party-sympathizer to carrying out the role of a party-activist, further blurring the lines between models of party membership and affiliation (Chadwick, 2007, 2017; Scarrow, 2015; Guaja, 2015). This can occur in the confines of one party or across party boundaries, as organisations like Campaign Together directed citizens to campaign for a range of different parties in order to minimise the Conservative Party’s electoral success. Intermediary bodies can therefore enhance party campaigns by providing new activists and resource.

An additional benefit of satellite campaigns is the potential for innovation. As organisations less restricted by legal requirements and responsibilities, these bodies have the space to innovate and trial new tools that parties may be wary of promoting. In this way, Momentum developed the ‘My Nearest Marginal’ tool which allowed campaigners to identify marginal seats and offer lifts or car shares with others from their area who wanted to travel to campaign. This tool helped to target the campaigning activities of hundreds of volunteers into the areas where campaigning was deemed to have the most significant effect. The capacity of a central party to develop and regulate such software is far more complex due to legal duties (especially when encouraging car sharing), hence innovation might emerge more easily when originating from beyond parties (Williamson, Miller and Fallon, 2010).

Despite these advantages, satellite campaigns also raise multiple questions regarding party control, specifically in terms of how parties should link to and work alongside these campaigns. Although some intermediary organisations have permanent infrastructure, others emerge purely around elections. Whilst, as Marengo (2013) argued, electoral campaigns are an opportunity to reach out to and empower non-party members, there is a challenge in capitalising on such links throughout the electoral cycle. It is not yet clear how embedded these organisations are in the campaign landscape, and so, even for strongly partisan organisations, parties cannot necessarily rely on—or even predict—their support. Furthermore, it is unclear whether the support offered by satellite campaigns is always welcomed. In the high-profile seat of Sheffield Hallam, where Labour defeated the former Liberal Democrat leader Nick Clegg, the contribution of satellite campaigns was unclear. On the one hand the winning Labour candidate Jared O’Mara was quoted as saying, ‘The contribution of Momentum members in South Yorkshire and beyond was exemplary … It was a blessing to have them on board campaigning to get me elected.’ (The Week, 2017). However, an interview piece in The Guardian provides a different take:

Momentum has, incidentally, tried to claim Hallam for one of its victories. But O’Mara isn’t having this. ‘No, no. I reject that entirely. I was grateful for their help, but it was a victory for every shade of red in the party. There are some really good eggs in there, but there are also a few people that … well, I maybe want to put a bit of distance between them and myself.’ (Cooke, 2017)

Parties will therefore need to consider whether and how they relate to intermediary organisations and satellite campaigns, and whether there may be institutional advantages to creating links that help to sustain and harness this enthusiasm. This is particularly important because organisations such as Campaign Together and Momentum maintain their own activist lists and communication channels that parties do not control. This lack of direct access to a reserve army of additional volunteers renders parties reliant either on building productive links with intermediary bodies, or developing their own systems by which to capture contact information and attempt to involve such individuals in party activities. Yet such activities may undermine what is attractive about satellite campaigns: that they are flexible and orbiting, not integrated into official party campaigns. They appeal to activists who consider themselves as ‘doers’ and not ‘joiners’ (Scarrow, 2015). If these organisations become more embedded in the campaign landscape in the long term, negotiating this boundary will be key.

3. Conclusion

The 2017 digital election campaigns may well be remembered for the normalisation of Facebook advertising and for the developing significance of satellite campaigns, but we should remember that digital remains one of many tools used by political parties. Like the printing press, typewriter or computer before it, digital technology enables parties to carry out existing functions more efficiently and within more expansive parameters, but its capacity to transform current practice is not deterministic. Rather, political activists, and especially those within parties, must consciously decide to engage with digital tools to promote a different kind of practice if lasting change is to occur (Lilleker and Jackson, 2010, p. 92).

Clearly, parties’ use of Facebook advertising relocates activities that have previously been conducted using internal party databases on to a digital social media platform. Traditional electoral campaigning has not been revolutionised in this sense— it has just gained a new dimension alongside face-to-face canvassing, leafleting, and phone banking. But the normalisation of this type of campaigning does raise important questions about public acceptance of such tools, as well as the complexities of regulating the digital sphere and the subsequent power of finance to buy electoral advantage.

Digital media in the general election of 2017 has also facilitated the growing visibility of satellite campaigns. However, the success of these campaigns and the intermediary organisations that drive them rests on contingent factors that make it difficult to assess at this stage both their influence and permanence as a feature of electioneering. Much of this infrastructure appears left-leaning or more specifically driven by support for the current Labour leader, and hence may be far more unfamiliar to the Conservative Party or UKIP. Given the unpredictability of the 2017 results, and our understanding of the difference that grassroots campaigns can make (Fisher, Cutts and Fieldhouse, 2011), this suggests that more traditionally right-wing parties could benefit from promoting and encouraging such bodies to emerge. However, if there is no organic support for such developments, parties could be accused of ‘astroturfing’ such campaigns, which is unlikely to get them the support and resources they need to target marginal seats in an effective way.

Cumulatively, these insights reveal that there is considerable ambiguity about the implications of these trends, specifically in terms of what we expect from campaigns. Whether driven by ethical concerns over the conduct and regulation of parties’ Facebook advertising, or reflecting ambiguity over the boundaries and scope of parties’ ‘official’ campaigns, our understanding of what constitutes electoral campaigning is evolving in line with developments in the digital realm. For parties it appears that there are considerable benefits to be gained from experimenting with new technology, learning from others, and perhaps most controversially, being willing to relinquish some control over their election campaigns. And yet the longer-term implications of these trends are by no means clear. The public’s tolerance of new practices, and their willingness to embrace different organisational forms and ideas, requires further investigation to examine what is wanted and expected of parties’ campaigns today.

Footnotes

References