Tomorrow the UK goes to the polls to decide which parties will form the government for the next five years. The BBC is not alone in calling this the “first social media election”. Which may seem odd as many gave that tag to the 2010 general election as well. When you look back at 2010 though, the impact of social media – both the efforts of the parties and the comment from the electorate – seems to have been minimal on the outcome. Trying to use quantitative data from social media to predict the result of the election was problematic (I know, I tried). One of the problems in using this sort of analysis for predicting UK elections are the difficulties in breaking down the results by parliamentary boundaries, but these models can breakdown even when you do have good geographic data to work with.
What is acknowledged by most as the real “first social media election” was the 2008 US presidential race. Barack Obama won his first term in the White House thanks – in part – to mobilizing the youth vote through social media in general, and through Facebook in particular. Analysis highlighted then by the authors of Throwing Sheep in the Boardroom: How Online Social Networking Will Change Your Life included a map of declared support by American Twitter users. This found every state overwhelmingly Democrat, apart from South Dakota – which was only ‘mildly’ Democrat. Clearly, Obama didn’t win every state, so there must be other factors that need to be taken into consideration.
Some seven years later the social media strategies and tactics of the UK political parties can still learn a lot from that campaign and Obama’s re-election bid. Though the differences in campaign funding, rather than social-savviness, may be the driver for that. Nahendra Modi, whose BJP party swept to power in India last year, is also hailed as a paragon of social media politics. Though their politics are different, President Modi has a lot more in common with President Obama than the current crop of UK parliamentary hopefuls. Notably, big budgets and the use of social media to engage the young, urban electorate. Ed Miliband’s dalliance with Russell Brand aside, there doesn’t seem to be much evidence of any UK party adopting this strategy.
Barack Obama may have over a year and a half left in office, but many candidates to replace him have already started their campaigns – and they have started them on social media. Hillary Clinton announced her decision to run on YouTube. As soon as the campaigns started, so did the attempts to judge the candidates on their social media engagement. This is premature for a number of reasons, not least the fact that the platforms being used might be ones that have never been used in a campaign before. Jeb Bush seems to be taking to Instagram in a big way and the rise of live streaming apps will certainly have an impact, if only to make sure no gaffe slips below the radar.
As we get closer to polling day in the UK, the amount of analysis of the parties’ social media performance in the mainstream press is ratcheting up. Whilst these analyses are more sophisticated than those which were possible five years ago, I’m not sure how helpful we (or the parties themselves) will find them in predicting support for any particular party. With this sort of analysis (in fact this applies to analysis of brand marketing as well as political campaigning) you need to understand three things in order to be able to make actionable conclusions.
The first is sentiment (whether people are speaking positively or negatively about the party) and to be fair, many of the analyses do include this metric. The second is intent (whether what people are saying on social media indicative of intention to vote one way or the other), this can be determined using machine-learned classifiers. The third – and this brings us back to the difference between Obama/Modi and the UK political parties – is demographics. The audience who are active on social media in the UK is different to the audience that votes. Understanding your audience is the key to a successful political (or marketing) campaign and being able to see how the different parties are performing in the segments of society that are actually likely (and eligible) to vote would give a clearer picture of the real state of affairs.