During the 2008 Presidential election, the Obama campaign pushed the envelope in two significant ways: they set out to change the face of the electoral map by mobilizing new and young voters; and they took the guesswork out of their resource allocation strategy to achieve that goal.
Any real change to the political system needed a change in the electorate. Rather than fighting over the same aging, well-off, white constituents, the Obama campaign went after the young and unregistered voices–a heretofore untapped resource estimated at 55 million potential voters as of 2004 (Hayes 2008).
According to a report published by the Pew Research Center on January 9, 2007, 18- to 25-year-olds, a group the Center calls ‘Generation Next,’ are the most tolerant age group in regards to divisive social issues like immigration, drug use, race, and homosexuality (Pew 2007). Yet, four out of ten eligible Generation Next voters were unregistered at the start of the Obama campaign.
The report also shows that Generation Next’s use of mobile and online technology to stay connected is double that of those aged 26-40 (Pew 2007). Hence, integrating mobile and online technologies to the core of the campaign was a deliberate choice aimed at following the voters to where they were already gathered (Carr 2008).
The Obama campaign had both the resources and the technical leadership to envision the campaign process as a living organism and utilize technology and social media to organize and mobilize their audience, gather feedback, and adapt to new challenges in real time (Delany 2008).
Rather than engage in the futile exercise of attempting to control an outcome from the top-down, the Obama campaign structured its efforts to model the complexities and nuances of a political campaign in a networked society the same way scientists model an ecosystem, a weather system, an immune system, or a social system–using the concept of a Complex Adaptive System.
A Complex Adaptive System (CAS) is a term that comes from Complexity Theory. It is a way to understand natural and social phenomena that are intricate and constantly adapting in response to changes in the environment.
A CAS is a dynamic network comprised of parallel agents that are components of the system; air and water molecules in a weather system, or supporters in a social system. Agents in a CAS have the ability to interact with and affect each other, send and receive feedback, self-organize, and produce non-linear dynamic behavior. The resulting system exhibits collective properties that are different from, and not reducible to, the properties of its agents; a storm has more power than any individual water or air molecule on its own.
The most interesting properties of a CAS are: simple rules, diversity, communication, self-organization, iteration, edge-of-chaos, emergence, and co-evolution. Each of these terms has a theoretical significance which we will not consider in details here, but for the purposes of our discussion, it is both an apt analogy and a practical technique that was employed by the campaign. An understanding of the significance of this theory was present at all levels of the campaign, from the policies expressed, down to computer simulations of each target state that were updated weekly with incoming field data from volunteers.
A CAS starts with a few clear rules. Winning was the simple and guiding principle for everything done at all levels of the campaign. From speech writing to hiring technology leaders like Chris Hughes and Joe Rospars.
A CAS requires diversity to thrive. At the center of the campaign stood a man at the intersection of many social, economic, political, religious, and racial groups. Obama as a candidate was able to connect to the experiences of a large and diverse group of people and draw them into the system.
The diversity was not only present in the types of individuals attracted to the campaign, but also in their level of involvement. Although at the end of the day the election was won by the boots on the ground that had the skill, the time, and the connections to register and mobilize new voters, the Obama campaign offered something to do for each level of willingness. The tiered levels of engagement allowed even the most casual of supporters to feel welcome and to be part of the campaign (Delany 2008). Through my.barackobama.com (My.BO), the central hub for the campaign, a supporter could register, buy a t-shirt, contribute $15, and call it a day. Alternatively, she could use the same resource to organize a house party or a larger get out the vote event.
The campaign did not shy away from diversifying their communication methods either. In addition to My.BO, the Obama campaign used a variety of existing methods to communicate with their supporters including TV, radio, text messaging, email, Flickr, Twitter, YouTube, and other social media. They also experimented with an iPhone application and a Firefox plugin.
A living system talks to itself and the campaign’s demonstration of this rule was near exemplary. Special attention was given both to the message and the delivery methods. Joe Rospars, Obama’s New Media Director, attributes the success of the Obama outreach to good content and well-crafted messages that created many touch points with many and diverse groups, thus generating cross-discipline influences on public discourse and cementing the relationship between the supporters and the campaign.
Obama maintained an active presence in the blogosphere throughout his campaign and had fifteen profiles on a variety of mainstream social networks, including Facebook and MySpace, and on some high-profile Asian, Hispanic, and African American sites like AsianAve.com, MiGente.com, and BlackPlanet.com (Edelman 2009). But while social media is important, it is not the whole story. Obama depended on the bridges between online and offline connections, and to do that he provided the source material and central hub at My.BO to communicate his message not just to his supporters, but rather through them; allowing them to replicate it and spread it.
Communication from the bottom-up was just as essential. Through a combination of feedback from volunteers in the field and sophisticated analysis tools built into these channels, information was flowing back into the campaign’s central database allowing for better data mining and an unprecedented level of micro-targeting.
At first glance, it is easy to get blown away with the sheer scale of the massive operation led by the Obama campaign. Special attention is always given to the number of emails and cell phones numbers collected, but at the end of the day it is about what they did with this massive quantity of data that put Obama in the White House.
The campaign followed a targeted approach for mobilizing people. Volunteers analyzed voter data to craft an appropriate message to send. The language and the layout of an email differed between a supporter and undecided voter. The follow-up channel itself, whether it be a Facebook reminder or a text message, was chosen depending on which one was the most suitable for the voter and the most likely one for her to act upon.
Door-to-door volunteers were able to slice and dice the data down to which doors to knock, and report back on which voters they had reached, enabling the database to remain current (Delany 2008). Voter information was stored and updated in real time in the campaign’s database. Boots on the ground tracked users and pulled their information on a very granular level. All of this allowed volunteers to hyper-target the hardest to earn votes.
Voter registration drives require a massive amount of time and energy. Generally, they are the most painful process of any and every election and campaign. The process happens in stages. First, you need to identify and find those unregistered voters. Next you need to follow-up with them in any way possible, by phone, via mail, and even in person. The cost associated with running this effort solely though paid campaign staff is prohibitive. The Obama campaign invested in finding and enabling volunteers to do that work for them.
Volunteers used My.BO to organize a thousand phone-banking events in the last week of the race and another 150,000 events throughout the campaign. By the end of the campaign, supporters had self-organized into more than 35,000 groups based on location and shared interests. (Stirland, 2008).
By providing training and access to the tools traditionally only available to paid campaign staff, the campaign was able to exponentially augment its workforce with volunteers. The volunteers became organizers and managed other volunteers. Independent voter registration operations sprang out all over the country without a single paid staff overseeing the operation, however they had the tools in place via My.BO to report back the data they collected to a central campaign database (Hayes 2008).
The best place for a CAS system to be is right at the edge of chaos where the largest number of possibilities exist and the highest level of creativity is possible. Agents of a CAS cannot be controlled; but they can be nudged or stimulated to respond. While the campaign only showed about 2,000 official YouTube videos, it generated 442,000 user-generated videos on YouTube (Hartman 2008). The campaign provided good content for sharing source materials for the public to use in order to create their own messages. The spread of viral videos was not and could not be controlled by the campaign. If they had feared malicious use of the source material by the opposition, they would have missed out on some viral exposure via videos like Yes We Can Obama Song by Will.I.Am. which, according to the Viral Video Chart, had 26,344,683 views across 8 different video positing and garnered 8,014 blog posts and 136,936 comments.
A CAS system cannot be planned. It depends on patterns emerging from the behavior of its agents to inform the behavior of the system. This was very true in this campaign as evident by Joe Rospars’ blog entry from May 2, 2007; “We’re flying by the seat of our pants, and establishing new ways of doing things every day. We’re going to try new things, and sometimes it’s going to work, and sometimes it’s not going to work. That’s the cost and that’s the risk of experimenting.”
A CAS system learns from experience through iteration; small changes that result from the initial conditions going through the feedback-emergence process a few times. According to David Plouffe, Obama’s campaign manager, the campaign used data gathered by their volunteers to build detailed and segmented models to predict response in each state. The gathered data from volunteers was used to obtain data as granular as chunks of voters on a city block in a given media market (Delany 2008).
The models were used to make constant changes to resource allocations and methods of communication. From changing the layout of an email to determining where a campaign representative, including the candidate, would visit next (Delany 2008). At the end of the campaign, about 1 billion emails were sent in 7000 variations to ensure the message resonnated with the recipient (Vargas, 2008).
In addition to learning from its volunteers’ experience, the campaign kept pace with changes in the political, social, and economic scenes both online and off. They used strong branding, tags, and search optimization to make their content easily findable online. The campaign fought smear campaigns with videos under the same tags so the same search terms that people used to find the negative videos would also lead them to positive ones. They also aggressively bought search ads and TV spots to make sure people would find friendly information first (Edelman 2009).
Perhaps the most magical aspect of a CAS system in full swing is the moment known as the threshold effect, which is a point when the system change is exponential, the moment when a weather system turns into a storm, when the market swings wildly from bear to bull, or the moment when the spread of a political movement turns into Obamania. The success of the next generation of U.S. politicians belongs not to those who can convey their message to the public, but rather those who can galvanize the public to spread their message and grant it moral authority.
Cross posted on http://www.com546.kennethrufo.com/
Bibliography and Sources
- Carr, David (2008, November 11). How Obama Tapped Into Social Networks’ Power. New York Times, Retrieved January 25, 2009, from http://www.nytimes.com/2008/11/10/business/media/10carr.html
- Daily Kos Staff. (2008, November 2). Obama Uses Houdini. Daily Kos, Retrieved January 26, 2009 from: http://www.dailykos.com/story/2008/11/2/182319/265/737/650137
- Delany, Colin (2008, December 24). David Plouffe: The Obama Campaign Used Grassroots Data and Computer Modeling to Allocate Resources in Real Time. TechPredsident, Retrieved January 25, 2009, from http://www.techpresident.com/blog/entry/3348/david_plouffe_the_obama_campaign_used_grassroots_data_and_computer_modeling_to_allocate_resources_in_real_time
- Delany, Colin (2008, December 3). Inside the Obama Numbers: Tiers of Engagement. TechPredsident, Retrieved January 25, 2009, from http://www.techpresident.com/blog/entry/33328/inside_the_obama_numbers_tiers_of_engagement
- Delany, Colin (2008, February 3). Is the Obama Campaign a Model for Online Politics?. TechPredsident, Retrieved January 25, 2009, from http://www.techpresident.com/blog/entry/21119 is_the_obama_campaign_a_model_for_online_politics
- Delany, Colin (2008, December 11). Joe Rospars and A Billion Minutes on YouTube: Content was Key (and Overlooked) Part of Obama’s Online Juggernaut. ePolitics, Retrieved January 25, 2009, from
- Edelman Digital Public Affairs. (2009, January). The Social Pulpit: Barack Obama’s Social Media Toolkit. Retrieved January 25, 2009 from Edelman Web site: http://www.edelman.com/image/insights/content/Social%20Pulpit%20-%20Barack%20Obamas%20Social%20Media%20Toolkit%201.09.pdf
- Hartman, Jalali (2008). Obamanomics: A Study in Social Velocity. Retrieved February 2, 2009, from Yovia Web site: http://www.yovia.com/Obamanomics.pdf
- Hayes, Christopher. (2008). Wanted: New voters. Nation. 287 (6), 24-33.
- Ifill, Gwen (Interviewer), & Plouffe, David, & Axelrod, David, & Davis,Rick, & McInturff, Bill (Interviewees). Inside Campaign 2008. Retrieved January 21, 2009, from C-Span Archives Web site: http://www.c-spanarchives.org/library/index.php?main_page=product_video_info&products_id=282851-1
- New Voters Project. Fact Sheet on Youth Vote and Text Messaging. Retrieved January 21, 2009 from New Voters Project Web site: http://www.newvotersproject.org/text-messaging
- Pew Research Center For The People & The Press. (2007, January 9). How Young People View Their Lives, Futures and Politics: A Portrait of ‘Generation Next’. Retrieved February 1, 2009, from Pew Research Center For The People & The Press Web site: http://people-press.org/reports/pdf/300.pdf
- Pew Research Center For The People & The Press. (2008, October 23). More Than a Quarter of Voters Read Political Blogs: Liberal Dems Top Conservatives Reps in Donations, Activism. Retrieved February 1, 2009, from Pew Research Center For The People & The Press Web site: http://people-press.org/report/?pageid=1411
- Rospars, Joe. (2007, May 2). Our MySpace Experiment. Retrieved January 24, 2009 from MyBarakObama Web site: http://my.barackobama.com/page/community/post_group/ObamaHQ/CvSl
- Rubel, Steve. (2009, January 17). Inside Obama’s Social Media Toolkit. Micro Persuasion, Retrieved January 27, 2009 from http://www.micropersuasion.com/2009/01/obama-social-media-tools.html
- Stirland S.L. (2008, November 11). Propelled by Internet, Barack Obama Wins Presidency. Wired, Retrieved January 28, 2009 from http://blog.wired.com/27bstroke6/2008/11/propelled-by-in.html
- Vargas, J. A. (2008, November 20). Obama Raised Half a Billion Online. Washington Post, Retrieved January 29, 2009 from http://voices.washingtonpost.com/the-trail/2008/11/20/obama_raised_half_a_billion_on.html
- Vargas, J. A. (2008, August 19). Obama’s Wide Web. Washington Post, Retrieved January 29, 2009 from http://www.washingtonpost.com/wp-dyn/content/article/2008/08/19/AR2008081903186_pf.html
- Winograd, M., & Hais, M. D. (2008). Millennial makeover: MySpace, YouTube, and the future of American politics. New Brunswick, N.J.: Rutgers University Press.