Tuesday, December 9, 2014

Obama For America 2012 and Google Analytics: Changing the World $3 at a Time

Though not a traditional e-commerce retailer, the Obama-Biden 2012 campaign is the most successful uses of Google Analytics (GA) in the history of politics. Though the campaign did offer a delightful online store with well-designed and clever merchandise (I still own a “cup of Joe” mug with Vice President Biden’s photo on it), the real product for sale was the Obama for America (OFA) brand.  In the course of the 2012 election, the Obama campaign raised $733M in donations from 4.4M individual donors, the majority of whom donated $200 or less to the campaign—a stark contrast to the Romney campaign that raised $479M from 1.1M donors, the majority of whom donated $2000 or more (Washington Post, 2012). Through a sophisticated use of both data analytics and digital analytics – the campaign had two separate analytics branches in the organization focused on different uses of data – voters were targeted with specifically crafted email messages asking for small amounts of money, often “just $3,” to not only fundraise but to earn the allegiance of their visitors (Engage Research, 2013). For a relatively small personal donation, the campaign’s consumers could feel fully engaged with the Obama brand. This engagement would translate into brand loyalty and consequent votes on election day. Using both traditional product marketing strategies and unconventional applications of e-commerce techniques, the OFA team created a dynamic and responsive marketing campaign for a winning political platform.

To cover the entire breakdown of the Obama 2012 analytics techniques is far beyond the scope of this blog post, however, the OFA campaign’s use of GA to enhance their digital analytics, and particularly their real-time responsiveness ans optimization is a masterful ecommerce strategy that could be replicated by any fundraising or highly visible enterprise to continuously engage customers.

Hey.
If you in any way aligned yourself with the Democratic National Committee, or the Obama campaign, you were added to the OFA email list. Jon Stewart may have poked fun on The Daily Show about the voluminous quantity of email sent on a daily basis from the campaign but these emails were not simply designed to wear down a contributor’s resistance. They were also the result of a complex A/B testing process and GA event tracking to determine which subject lines prompted opening the emails and clicking through to the website. According to Amelia Showalter, OFA Director of Digital Analytics, “We did extensive testing not just on the subject lines and the amount of money we would ask people for but on the messages themselves and even the formatting.”  The campaign would test as many as 18 variations before selecting a version to send to its massive email list.

The most popular subject line? “Hey.” The casual and familiar tone encouraged recipients to open the email. Additionally, staffers reported that their instincts were practically useless when designing email blasts. “We were so bad at predicting what would win that it only reinforced the need to constantly keep testing. Every time something really ugly won, it would shock me: giant-size fonts for links, plain-text links vs. pretty ‘Donate’ buttons. Eventually we got to thinking, ‘How could we make things even less attractive?’ That’s how we arrived at the ugly yellow highlighting on the sections we wanted to draw people’s eye to,” according to Showalter. A/B testing proved to be the reliable measure of potential success. The extrapolated monetary difference between versions could add up to millions of dollars from a single email blast (Green, 2012).

Secondary Calls to Action and Ease of Re-Donating
After an initial signup, the digital analytics team used GA Advanced Segmentation and Flow Visualization features to determine the best way to reengage their audience and craft messaging for these segments. The digital analytics team used these tools to test the efficacy of sharing on social media as well as measuring what messages resonated with particular audiences. The result was continuous optimization in response to the content that proved most resonant with existing subscribers (Google Analytics, 2013).

Additionally, the digital analytics team continually tested the efficacy of their donation pages. By observing donor behavior and bounce rates, the campaign was able to reformat the donation pages, conduct A/B tests and raise their conversion rates by 49%. Additionally, by merely increasing the load speed of the donation pages by 60%, the team saw a 14% increase in donations. By simply monitoring user behavior on pages and noting where users dropped off in the funnel results, the team enabled the campaign to maximize supporter donations (Engage Research, 2013).

Binders Full of Women
The OFA campaign anticipated that a significant majority (64%) of viewers utilize the internet to fact check statements made during presidential and vice-presidential debates. Though the digital analytics team anticipated many of the potential searches, not all of them could be foreseen. Many quotable moments are created during the course of a broadcast, so the digital analytics team generated real-time reports in GA to a) identify new organic search terms, and b) create and place relevant ads in the search results for these terms. During the second presidential debate Governor Mitt Romney uttered this unfortunate statement:
“And I—and I went to my staff, and I said, ‘How come all the people for these jobs are—are all men.’ They said: ‘Well, these are the people that have the qualifications.’ And I said: ‘Well, gosh, can't we—can't we find some—some women that are also qualified?’ And—and so we—we took a concerted effort to go out and find women who had backgrounds that could be qualified to become members of our cabinet. I went to a number of women's groups and said: ‘Can you help us find folks,’ and they brought us whole binders full of women,”

In response, the OFA campaign was able to create immediate targeted keyword advertising focusing on the President’s achievements in equal rights legislation running alongside the search results for “binders full of women” (Google Analytics, 2013).

Get Out the Vote
One of the most innovative uses of GA during the campaign was the Get Out the Vote effort as election day drew near. OFA used GA to allow the entire team to view the results of advertising and website data to empower rapid decision-making in the final days of the campaign. The team utilized the Google Analytics API and scripts to automate their reporting of marketing efforts in various channels, with a specific emphasis on geo-targeting accuracy. A very useful features of the OFA site in both the desktop and mobile versions was the polling place lookup feature. By customizing the GA reports, the digital analytics team could determine whether voters were being directed accurately, and specifically whether their location was in a high priority area for the campaign. If the lookup feature wasn’t performing as expected or desired, it could be adjusted on the fly for more precise results (Google Analytics, 2013). Additionally, the campaign spent millions on Get Out the Vote election ads on mobile devices, targeting down to a neighborhood level young, female, and Hispanic voters to ensure that the OFA campaign was maximizing voter turnout (Heine, 2012).

In Summary
Everyone knows how the story ends, but not the analytics that contributed to making that ending a reality. By engaging voters to be part of the solution through responsiveness and real-time optimization driven by advanced GA features, the OFA digital analytics team was able to better understand their consumer/voter and provide the appropriate opportunities and incentives to engage with the campaign brand.


References

Engage Research. (2013). Inside the Cave: An In-Depth Look at the Digital, Technology, and Analytics Operations of Obama for America. Alexandria, VA: Engage Research.
Google Analytics. (2013). Obama for America uses Google Analytics to democratize rapid, data-driven decision making. Mountain View, CA: Google Analytics.
Green, J. (2012, Nov 29). The Science Behind Those Obama Campaign Emails. Retrieved Dec 7, 2014, from Bloomberg Businessweek: http://www.businessweek.com/articles/2012-11-29/the-science-behind-those-obama-campaign-e-mails
Heine, C. (2012, Dec 18). Here's One Advertiser Who Swears Mobile Ads Work: Obama. Retrieved Dec 7, 2014, from AdWeek: http://www.adweek.com/news/technology/heres-one-advertiser-who-swears-mobile-ads-work-obama-146044
Washington Post. (2012, Dec 7). 2012 Presidential Campaign Finance Explorer. Retrieved Dec 8, 2014, from Washington Post: Campaign 2012: http://www.washingtonpost.com/wp-srv/special/politics/campaign-finance/ 

No comments:

Post a Comment

Web Statistics