When Barack Obama was elected to the United States presidency in 2008 the marketing story was the power of social media, and how his campaign successfully leveraged tools such as Facebook to drive unprecedented voter interaction (view for free the Obama for President case study). Four years later, Obama is celebrating another victory and this time the talk is all about big data.
Big data has become an ubiquitous phrase in marketing circles as brands and agencies hasten to get to grips with the torrent of information unleashed by the digital age. Underlying the general excitement at the sheer quantity of consumer data available is the nagging but persistent question of what exactly to do with the data.
Politics provides one compelling answer as insiders trumpet the role of big data in Obama's re-election. Time magazine devoted a feature to the role of data crunchers in helping swing the close contest in Obama's favour, declaring that "the role of the campaign pros in Washington who make decisions on hunches and experience is rapidly dwindling, being replaced by the work of quants and computer coders who can crack massive data sets for insight".
As far back as February, The Economist reported that the era of big data has transformed traditional notions of political advantage, noting that: "One of the less obvious [advantages] may be the leisure to recruit a strong team of boffins. Team Obama has long been scouring the nation for scientists. It has sought out computing experts, mathematicians, programmers and statisticians."
The Obama campaign recruited Rayid Ghani, former senior researcher at Accenture Technology Labs, to head its data team as it made deep dives into a vast pool of information. American electoral rolls were linked with commercially available consumer data, The Economist reported, allowing the Obama campaign to cross-reference names and addresses to uncover "everything from magazine subscriptions and home ownership, to hunting licences and credit scores."
The aim was to develop an unprecedented level of targeting to drive voter activation and fundraising. Though success in the former field was not confirmed until the final returns on 6 November, big data gave the Obama camp confidence behind the scenes. Campaign managers, speaking on the condition of anonymity, told Time reporter Michael Scherer that:
The polling and voter-contact data were processed and reprocessed nightly to account for every imaginable scenario. "We ran the election 66,000 times every night," said a senior official, describing the computer simulations the campaign ran to figure out Obama's odds of winning each swing state. "And every morning we got the spit-out - here are your chances of winning these states. And that is how we allocated resources."
According to Time, the campaign was able to buy TV 14% more efficiently and ensure the ads targeted the most persuadable voters. It also drove decision making about what celebrities to engage as fundraisers, and what mechanisms encouraged repeat donations.
The Obama argument (to coin a phrase) for big data could not be clearer: it delivered a hard-fought election and over $1bn in campaign donations. But what - if any - are the lessons for brands?
It may appear unique, but the American election can be examined like any other marketing challenge. The Obama campaign needed, to varying degrees, what all brands need: to create awareness, foster engagement, increase preference and ultimately spur action.
Whether you are a politician or a retailer, the rules for successfully exploiting big data are the same:
- Clearly define your objective
- Recruit the best available talent
- Set measurable key performance indicators at every stage
- Have a consistent brand message
- Iterate and modify in real time
Further reading on Warc:
Obama for President (free access), ARF Ogilvy Awards, Gold, 2009
Innovative online research - The US presidential campaign of Barack Obama case study, Ron Riley, ESOMAR Online Research, 2009
AMA Mplanet 2009: Marketing Obama all the way to the White House, Geoffrey Precourt, 2009