#Electronic democracy

Scotland is engaged in the most significant exercise of the franchise in living memory. Just a few days after this magazine is published we’ll be voting in the referendum on independence. Then, with barely enough time to digest the result, we’ll find ourselves in the middle of the 2015 General Election, followed in quick succession by the 2016 Scottish Parliament election. 

Setting aside the sense of dread that all evokes in those that aren’t political anoraks, this pace and number of elections poses challenges to the key players: how do the political parties manage all this? The answer is found in the increasingly sophisticated technologies that Scottish parties deploy to maximise their vote.

Until relatively recently, a typical campaign room would have been recognisable by the sound of a (usually ancient) Gestetner machine shaking the foundations as it produced leaflets. An enthusiastic volunteer would normally have a bundle of said leaflets thrust in their hand along with a canvass sheet. Their activity would be governed by the all-important canvass cards; produced by nothing more than a paper copy of the electoral register, big scissors, glue and a stack of index cards. Returning canvassers would report back results, to be marked up with highlighter pens on the cards, which would usually be covering all available tables. Although these manual methodologies were proven and helpful within the constituency, they had obvious limitations for command and control of the national campaign.

Modern campaigns are still rooted in these techniques, but today canvass runs have been replaced by iPads. Similarly, the visual check for party colours on an organiser’s table has been replaced by complex data analysis that highlights supporters, strength of feeling and likelihood to vote. 

This technology revolution started right here, in Scotland, with the SNP’s campaign in the 2007 Scottish Parliament election. Voter targeting was achieved, to great effect, by the use of software that was related to the 2008 Obama campaign. The revolutionary aspect of that campaign was the new ability to identify core and swing voters – and then focus resources on securing their votes. This was achieved by creating a dataset that combined canvass returns with publicly available data – then applying business intelligence to identify target voters.

For instance, canvass returns tell campaigns how voters are aligned and their declared strength of commitment, the ‘marked-up’ electoral register reveals how likely they are to cast a vote, and demographic classification (a service provided by the major consumer information agencies) helps the campaigns understand what issues the voters are likely to be most concerned with. Combining these datasets means the campaigns can target communications for greatest effect – if the party has a policy that benefits, for instance, professional couples, then these technologies can predict who they are and ensure that the policies are highlighted in targeted campaign communications.

Roll forward to 2014, and election technology has leapt forward again, this time, harnessing the world of social media (Twitter and Facebook) to connect with voters and track sentiment. This provides the political parties with two crucial benefits: namely, the volume and relevance of information that can be sent out to voters. Delivering targeted letters in person is costly and time consuming, but connecting with voters on social media is instant and can address specific concerns. However, arguably, the greatest power of social media is that it enables voters to distribute content themselves, either from central campaigns or user-generated. The recent independence campaign has seen social media used effectively by campaigners on both sides to build interest groups and allow voters to connect directly to each other. 

Such technology is not without drawbacks, though. The online independence referendum campaign has seen distasteful behaviour at the fringes, and expensively produced campaign materials have been ‘creatively’ amended into parodies that are less than helpful to the side that originally produced them. Social media may be useful for disseminating a message, but once content is published it becomes completely outwith the control of the central campaign co-ordinators – it becomes public property.

What may also surprise those outside politics is that the next generation of voter-ID systems are able to track social media behaviour, meaning that parties can see how supporters and target voters react to events – in real-time. Campaign managers can now see how sentiment shifts during an election period and target communications to convert swing voters or ensure core voters get out and vote. 

The technologies can go further – we are now seeing American campaigns able to link constituency casework to voter profiling. Such applications are unlikely to be seen in Scotland in the near future; data protection law in Scotland, along with the privileged nature of casework, means that Scottish politicians would avoid sharing constituents’ data.

So, as we conclude this referendum campaign and look forward to the General Election, the revelation is that Scottish politicians have a lot to teach the public sector and industry on using technology to connect with stakeholders. Public bodies and consumer-focused businesses would do well to mirror the way that Scottish parties are now using data to understand their stakeholders, focus resources and tailor communications with them. Yet perhaps the greatest learning point from the political parties is that no amount of data, no amount of deep insight, no amount of complex analysis will ever replace the power of a canvasser standing on a doorstep and personally asking for your vote. Personal interaction is still the most powerful weapon in the armoury.

Chris Jones is a client manager with a major UK consultancy and sometime academic lawyer, specialising in procurement and IT law.