The context

Barack Obama won not one, but two elections: first, the contest to be the Democratic Party's candidate and then the Presidential election against John McCain (and others). From the Iowa caucus on 3 January 2008 through to the final public election on 4 November, there was a ten-month stretch — and in fact Obama was campaigning for even longer. There was no official start to the contest to be Democrat candidate, but all the front-runners were already campaigning regularly through 2007.

This long campaign period through 2007–2008 allowed large data sets to be accumulated, numerous tactics and techniques to be tested and refined, and provided a regular supply of deadlines and pivotal moments that could be used as marketing hooks to raise money, recruit volunteers and spur action.

Summary

The closeness of the contest between Obama and Hillary Clinton for the Democratic nomination made these marketing opportunities all the greater, because while in previous years many of the primaries and caucuses became dead events as the nomination was effectively settled before they were held, in 2008 the contest continued on and on.

Although, for the public election against McCain, Obama was the front-runner and favourite for a long period of time, the experience of previous front-runners with large leads imploding and losing (most notably, Michael Dukakis in 1988) meant that the excitement, uncertainty and tension that fuels marketing opportunities continued. Not only was there a need for marketing (uncertain result means a need to mobilize help and to persuade the public) but there was also a compelling story to power it (we might lose).

Political marketing in the US

US politics and society are not the same as the UK's, a point often overlooked when people seek to learn lessons from American elections for the UK.

One of the key differences is the role of national media. Political TV advertisements are legal in the US, while they are banned in the UK. The importance of TV advertisements is strengthened by the size of the country, which makes the reach of TV particularly appealing. Moreover, in the US, there is relatively little in the way of national newspapers, with USA Today being a relatively minor player in media consumption and papers such as the New York Times having only limited reach outside their core markets and audiences. By contrast, the UK has a large national media, plus it has the BBC which accords significant coverage to politics.

As a result, national contests in the UK can be played out in the national media — with campaigns aiming to win and influence national media coverage — in a way that US campaigns can only hope to match by running TV advertisements.

Partly because of this importance of expensive TV advertising, the ability to raise money to fund campaigns is not only a more important influence on US than UK elections but it has also come to be seen as a measure of judging the health of a campaign in itself. Regulations require fund-raising totals to be regularly published, with the result that good or weak figures boost or hinder election campaigns just as a good or weak debate performance or run of opinion polls can.

One reason for raising money is therefore simply to be seen to be raising money — as that is a yardstick by which campaigns are judged.

In addition, although the complexities of US electoral finance regulation allow people to make many donations in many different ways, there are low limits on how much can be given to individual campaigns for their own direct use ($2,400 in hard money to a candidate for their primary contest and the same again for the general election). This places a strong incentive on candidates to seek a much wider fund-raising base than in the UK, where there is no upper limit on donations, and donations of more than £1 m each are seen.

The marketing objectives

Obama's campaign had two complementary strands of marketing objectives: mobilizing activists and winning votes.

Mobilizing activists involved building up communication channels (eg email lists, friends on Facebook and records in volunteer databases), getting people to give money and getting people to give time.

Winning votes then, in turn, involved not just benefitting from the above but also getting people to register to vote, getting people to make use of postal ballots and getting people to vote.

Both of these strands existed before the creation of social media, the internet or even computers. However, the development of technology has opened up new intermediate objectives (eg getting people to sign up to an email list) and provides new methods (eg getting people to help by making phone calls using a computerized phone bank system).

The Obama campaign was unusual in the very heavy emphasis on recruiting volunteers to campaign at a local level. One British visitor who helped out explained that, ‘Every supporter was asked to volunteer. Every volunteer was asked to recruit more volunteers. Even when there were more volunteers than we knew what to do with, we recruited more’. This was combined with a heavy investment in training, from the 5-minute chat for someone who walked through the door of a local office to do some clerical work through to intensive weekend boot camps for the more dedicated volunteers.

Even the smallest of improvements or efficiencies can, in a finely balanced situation, make a large impact on the final result in an election. Gary Hart's run for the 1984 Democratic nomination may well have turned out differently had online fund-raising been available to him so that the momentum from victories turned into extra money in time. Instead, for him, the slowness of direct mail responses meant that the cheques came in too late to be spent when it mattered.

Two-way communication

However, where technology has had its most strategic impact in 2007–2008 is in the emphasis on two-way communication.

What distinguishes social networking, in particular, from previous communication mediums is the emphasis on two-way communication; not just sending out messages but listening, responding and interacting.

While those traits have always been possible — you can respond to a letter, you can ring someone back and so on — services such as Facebook not only make this easier but have also developed a strong culture in which that is the expected normal way of behaving.

The Obama campaign expanded on an insight from Howard Dean's bid for the Democratic nomination in 2004, in which his campaign team realized that by placing such interactions in the public sphere they could tap into the voluntary efforts of their supporters. Dean's web presence was centred on a blog, in which people could post comments. This meant that not only could people post questions or make negative comments, but also that his supporters could see these and respond themselves rather than the campaign having to rely on its own team to respond to everything (or, when overworked and out of time, fail to respond at all — as has been the fate of many emails sent to American political campaigns over the years).

More generally, this idea of citizen marketing — getting ordinary members of the public to take up your messages and distribute them — complemented the current marketing Holy Grails of personal recommendation and word of mouth. By providing people with convenient and high-quality resources to do this marketing, the Obama campaign was able to retain a strong control of its brand at the same time. Rather than worrying about how people might use the campaign's official logo, and thereby trying to restrict its use, the campaign made it easily and widely available — so that there would be no need for anyone to use anything else.

Fishing where the fishes are

The Obama campaign made widespread use of social networks, including not only the most popular ones at the time such as Facebook and MySpace but also niche ones such as AsianAve, Black Planet and Eons.

The logic was to fish where the fishes are, that is, to take the campaign to where people are located — a logic that applies as much in the virtual world as in the physical one. This included getting over 3.2 million fans on Facebook and over 1,500 groups being created on the photo sharing site Flickr, a reflection in part also of the photogenic characteristics of the candidate.

These sites allowed people to generate their own content, which in turn spread the Obama message more widely. The classic low-key example of this was people who volunteered for the campaign could take a photo of themselves at the office and share it easily with friends through the likes of Facebook, Flickr and other services.

However, this myriad of different presences all tried to funnel people back to the core Obama campaign's website or email list as quickly and efficiently as possible.

Part of the core website presence was Obama's own social network, with a name that crossed his initials with the format of MySpace's name: MyBO. On MyBO, people could create their own profiles (over two million did), write their own blogs (around 400,000 posts were composed), organize events (around 200,000), network with other supporters, raise funds (approximately $30 m) and volunteer to help with phone calls and door-to-door work.

Email lists

Also central to the Obama campaign was its email list, variously claimed to be between 10 and 13 million strong. Some caution should be attached to those numbers as the email list was open for anyone around the world to join — and Obama had strong international appeal when candidate — and the figures also may not allow for bounce rates.

Although not disclosed, it is very likely that the bounce rates on the list were high, as one of the campaign's suppliers, Blue State Digital, have gone on the record to say that the most popular source of email addresses for the Obama campaign was offline, for example passing round sheets of paper at rallies.

It was, nonetheless, a very large email address list — yet the money it raised came in small quantities. The trick was to make them add up to large totals. The Obama campaign did not have an unusually large number of small donors, but, by getting them to make repeated small donations, it raised large sums from small donations. The average donor made 2.5 donations to the Obama campaign.

Emails were carefully segmented, with 7,000 different messages sent out among the 1.2 billion emails sent to the Obama list.

The campaign raised around $500 m online, of which two-thirds came from people clicking on a link in an email sent to them.

Blue State Digital have said that, for a typical 9-month US campaign, they now expect to raise around $5.00–$7.50 per opted in email address, which is a fairly low figure. What turns it into a large pile of money is having a large volume of addresses on the list.

SMS

While other election campaigns since the turn of the century have made heavy use of email, the Obama campaign was the first to break out into using text messaging (SMS) on a large scale. SMS campaigning and marketing has been held back in the US compared to the UK by the different pricing models in the two countries. In essence, in the UK, you pay to send; while in the US, you pay to receive, making people much more reluctant to let others know their number for texting purposes.

The central texting effort of the campaign was, however, rather botched. People were encouraged to sign up for text messages by the promise of being the first to hear of Obama's pick for his Vice President running mate. However, in the event, the traditional media broke the story first, by spotting the increase in security around Joe Biden's house just ahead of the announcement. As a result, the media ran the story first and the text message was then rushed out at a time when many recipients were asleep.

This hiccup aside, texting played a particularly useful role in reminding people of events — to turn up for a rally, to vote and so on — making use of the fact that 80 per cent of text messages are read within 2 minutes of receipt. In total, the Obama campaign ending up with around three million mobile numbers opted in to receive information.

Migrating from technology to personal interaction

In 2000, Howard Dean's bid for the Democratic nomination achieved stunning success at online fund-raising and recruiting some highly motivated supporters. Where it failed was in converting that to more traditional doorstep and phone campaign activity, which won votes.

The use of SMS was but one of the ways in which the Obama campaign strove — successfully — to avoid falling into the same trap. As with Dean, money raised online was used heavily on TV advertisements to reach a wider public, but, in addition, the Obama campaign was much more successful at using technology to funnel volunteers towards making phone calls (particularly via their virtual phonebank software that let anyone become a call centre operator from their own home), calling on doors and taking part in other local, offline activity.

In that sense, the Obama campaign was not just the technological successor to the 2004 Dean campaign but also the organizational successor to the 2004 Bush campaign with its highly organized focus on targeting the right voters with grassroots campaigning.

Using other's technology

MyBO and other Obama campaign technology were based on systems developed for the Howard Dean campaign. By building on what others had already done, the Obama campaign not only had technology that worked from day 1 and through the campaign but also could improve its sophistication rather than having to recode the wheel.

Some of this ability to avoid redoing the work of others was based on the growing popularity of open-source software: available for free and for anyone to adapt and amend.

Fragmented and integrated data

One of the major innovations that the Obama campaign did make on technology was integration. Traditionally, American campaigns have used a myriad of different systems and databases with, at best, one database for voters and managing their records and a separate database for managing donors and helpers, even though many people fall into both categories. In practice, the databases have often been even more fragmented, such as with a separate set of records from online donations from those processed by the direct mail team.

Owing to the range and depth of different database needs for different parts of the campaign, the Obama team did not attempt to have just the one database but, instead, put significant effort into data-matching across the systems. In effect, there was one system that handed out unique identification numbers for each person and, when any of the other systems added a new person to their own records, they checked against these records to append the right identification number (if the person was already in another database) or to get a new identification number (if the person was new to the campaign's records).

This type of data matching is never perfect, particularly when different mediums lend themselves to getting different pieces of data from people, leaving doubt then as to whether two records are really for the same person. However, it did in principle — and largely in reality — allow an integrated approach to data with knowledge about people's activities across different areas being able to be utilized for tailoring the right marketing for them.

Allies and data

In addition to the Obama campaign itself, there was a range of other organizations campaigning directly or incidentally for an Obama Presidency, such as other Democrat candidates up for elections, various advocacy groups and Democrat organizations including the Democratic National Committee itself.

Full integration across all these operations was not permitted by law. However, there was much greater integration at the data level than in the past due to the work of a firm called Catalist. While their approach of providing data services to a range of clients was in itself nothing new, they added to this a strong emphasis on different clients sharing at least part of their data with each other.

Catalist had more than 90 clients in the 2008 election cycle, whose activities enriched a database of 260 million people. Of these, 49 million were contacted at least once by a Catalist client, with a total of 127 million contacts made. Over 60 million pieces of response data (eg an answer to ‘how are you going to vote?’) were collected from over 35 million different people.

Other marketing changes

The rise of social networks and the heavy use of email were the most dramatic changes in marketing technologies for this Presidential election cycle compared with previous ones.

Less dramatic, and in many ways more marked in other recent American elections, are three other changes.

Firstly, the decline of automated calling — while this is still a substantial business in the US, it is far from being the coming thing. A combination of increased phone screening by the public and pressure for more regulation means it is a political marketing technique that is slowly fading away.

Secondly, increasing variation in personalized direct mail as printing techniques make it financially feasible to vary more and more of the contents of a piece of direct mail beyond the old-fashioned approach of only varying the name and address. Variations now often include swapping in and out graphics and photographs — in colour — depending on who the recipient is.

Thirdly, a use of search engine optimization and online advertising to tap into the surges in web traffic, which occur off the back of ‘real world’ events, such as in people searching out information on a topic that has featured in the news.

The Obama campaign, in particular, used this tactic to respond to personal attacks and smears on Obama. A dedicated website, Fight The Smears, and other approaches were used both to arm supporters with the arguments to rebut such attacks and also were there to be easily found by people searching out information on the smears. This meant the campaign could fight smears even when it felt it appropriate to be quiet in the traditional media for fear of giving the smears more publicity.

Message, not technology

Although these technological achievements were impressive, their success primarily came from the success of the candidate in winning popular support. Enthusiastic people can overcome faulty technology; apathetic people are not motivated by slick technology.

It was a point made by one of the pioneering Democrat bloggers, Jerome Armstrong, just after Obama's victory when he cautioned Republicans against looking at technology as the answer to winning in future: ‘Until the [Republicans] get ideas to sell that people believe are real, it doesn’t matter how good they use new media and the internet, they will be in the minority’.

The Democratic Party's subsequent electoral travails in 2009 and 2010 demonstrate how the technology is not enough on its own and, in a different political context with a more motivated opposition, the huge databases and clever systems look rather more fragile.

Indeed, comparing the software packages and lines of code used by Obama with those used by Hillary Clinton and John McCain shows an edge for the Obama campaign, but not massively so. What really gave the Obama campaign its edge was the commitment from the top to use technology and take part in the two-way communications culture. Married to an early start with data collection, a commitment to using data well and to innovate where necessary, this attitude produced a campaign that excelled not just in oratory but also in marketing.