Practicing Failure Helped Obama Win
May 23, 2013
By George Stenitzer, VP-marketing and corporate communications, Tellabs
Harper Reed, chief technology officer of the Obama 2012 campaign, told the story of the technology that helped win the election. Reed spoke to about 300 people at Fuel, a BMA Chicago event held on May 14.
The key lessons that Reed learned on the Obama campaign: Hire the best team. Use all the “things” that work. Focus on Big Answers, not Big Data. Microlisten. Practice failure constantly.
Hire the best team.
“Most problems as a CTO are about people; they’re not about technology,” Reed said. “We hire engineers to do engineering. Yes we code.
“Engineers are weird, right? I suggest an exchange program. Sit near each other to understand priorities and how things work. It’s about being transparent on the engineering side,” Reed said.
Reed hired 40 engineers from top companies to build a technology platform—10 times the number of engineers who worked on the 2008 Obama campaign. The team built a platform called Narwhal, which gave them freedom to focus on products. “We built a very large infrastructure. We required the cloud to be successful.”
With Narwhal built, the team created tools for telecommunications, a dashboard, mobile applications on Android and iOS, social media and Big Data.
‘Use all the things.’
Reed applied a “use all the things” framework, including every social media platform—except Pinterest. “Pinterest turned out to be a red state of its own,” Reed noted.
Facebook generated a huge number of fans and a huge opportunity to share photos. By some calculations, it reached the entire U.S. population.
The team also used Twitter, Tumblr and Reddit. “No channel was too small. No channel was too weird—except Pinterest, because it didn’t work.”
Reed cautioned marketers about hype surrounding Big Data. “Big Data has been solved for years. The key is the questions. Big Answers should be the focus, not how and where to store data.”
To engage, ask. And listen.
Reed emphasized the application of technology tools, not the tools themselves. “Nothing changed with email,” he said. “We just made it easier.”
“Ask people to do something,” he emphasized. “We asked for $56. We texted for contributions. With social and personalization, we asked people to ask a friend to vote. We raised boatloads of money with better content and distribution.”
The campaign emphasized microlistening, making contact with voters by knocking on doors, then building a profile to inform the digital content. “We sent direct messages preemptively to influencers, which led to conversations. Listening worked really well.”
Fail, over and over again.
The Obama technology team spent the entire month of October practicing failure. They intentionally broke each piece of technology, one at a time, to find out what would happen and how to recover most quickly. They broke databases, social platforms and other technologies separately and in combinations.
“It was an emotionally draining exercise,” Reed said. But practicing failure produced a thorough playbook on how to fix each problem quickly.
On Election Day, the Narwhal platform indeed failed. But because the team had practiced failure, “we went to the playbook, page 35, and fixed it before the users ever even realized there was a problem,” Reed said.
The Romney campaign platform Orca also failed on Election Day. Since the campaign had not practiced failure, it was unable to recover from “a small bug.”
“We started from zero. We raised $1 billion in 18 months. We focused on execution,” Reed said.
"Practicing Failure Helped Obama Win." George Stenitzer, VP of Marketing and Corporate Communications at Tellabs. BMA Buzz. 5/23/13.
You must be logged in to submit a comment.