Filter bubbles and fake news - welcome to the 2017 election

David Reed, director of research and editor-in-chief, DataIQ

Political campaigns are created around a core narrative. That’s why “strong and stable” on one side is pitching against “hard Brexit” on the other, with little in the way of nuance or detail. The paradox of this year’s general election campaign is that these simplified positions are being played out in the highly-targeted arena of social media. Sharp tools with blunt purposes.

That is also giving rise to a secondary narrative about how, exactly, political parties are reaching out to supporters and potential voters. Expect to hear a lot about psychographic targeting of ads and the transfer of campaign tools used in the recent US presidential election into the UK. 

But are these stories examples of marketers’ own filter bubbles or even, whisper it, fake news? Clearly, there is money to be made by companies who demonstrate their skills in the (usually unprofitable) arena of political marketing because it attracts commercial clients who want the same effect for their brand. That’s why a whole cohort of American social media practitioners can be found on the conference circuit talking about what they did for Presidents Obama or Trump. Whether the techniques are truly transferrable is open to doubt, given that political allegiences tend to run deeper than brand preferences.

You are what you like

That said, psychographics is being used in the reverse direction - using brand likes, content engagement, comments and so forth to model out political affiliation. Unlike the US, where voters register as supporters of one party or another, politics is more secretive in the UK. Undecideds can still swing the outcome, as they did in both the last general election and the EU referendum.

So political marketers are working hard to understand who to hit with messages that will either sustain an existing vote for a party or win it from the uncommitted. The science behind this is well established - delegates at DataIQ Future back in 2013 would have seen Michal Kosinski from the Cambridge University Psychometrics Centre show how to predict with 85% confidence whether voters were liberal or conservative, based on Facebook data from 60,000 volunteers. (The Centre eschews political clients,)

Parties will be drawing on publicly-available social media data, as well as their own engagement histories, to create target audiences in these channels. But remember, for all the scientific effort, audiences are still bought by the thousand, not one-by-one. In marginal seats, this dilution could render the targeting effort meaningless.

Class still counts

Unlike the US, British voters bring to the polling booth a powerful awareness of their class and that of the candidate or party. Big shifts in support only tend to happen when one party manages to appeal across historical boundaries, as Thatcher did by winning over the aspirational working classes in the 1980s and Blair did by reassuring the middle classes in the 1990s.

The big unknown this time around relates to the disaffected who voted Leave in the referendum and who still feel disenfranchised from the two main parties. With UKIP struggling to gain traction at a national level, these voters have no natural home. They may choose to stay at home as a result, meaning that whatever majority may be won could be seen to be undermined by a low turnout. 

Kosinski’s method does allow for targeting by IQ, as well as by the Big 5 OCEAN personality traits (openness, conscientiousness, extraversion, agreeableness, neuroticism). But this is only valuable if a party has a message that resonates with this targetable group. That doesn’t seem to be the case this year.

Less paranoid, more savvy

What swept Trump to victory last year was a wave of paranoia about global trends and supposed conspiracies that left millions feeling that the system was working against them. Cynical manipulation of this emotion, at live events and through online content, got the candidate over the line.

In the UK, consumers are generally more savvy about the messages they receive and are less likely to believe that “the 1%” is secretly manipulating everything. (Think about the outcome of the last two national votes and it’s obvious those in control weren’t.) That doesn’t mean fake news won’t work in the UK or that Brits don’t have their own filter bubbles. But the best defence against these techniques may simply be the national tendency to take the piss out of everything. Against mockery, even the most confident struggle to stand.



Please note that blogs are the sole view of the author and that they are not neccesarily the view of IQ ddg Ltd and should not be interpreted as advice. Please read our full disclaimer

Director of research and editor-in-chief, DataIQ
An expert commentator on all things data, David has been editor of DataIQ since its inception in 2011.

You have....

to be GDPR compliant.

Register with us for all the news

Sign-up to hear about the latest DataIQ news, content and events.