David Plouffe served as the campaign manager for President Barack Obama’s 2008 presidential campaign and was appointed a senior advisor to the President in 2011. He is now a strategic advisor for the ride service, Uber. David Plouffe visited the Goldman School during “Stop the Clock” week, which provides students and faculty with a chance to interact with leading decision-makers who have worked in public policy and government. The following is an excerpt from a conversation between him and students Taylor Myers, Andrew Wilson, Cassandra Bayer and Ian Perry of PolicyMatters Journal.

The full interview will be available on PolicyMatters Journal.

At the Goldman School, we use a strategic framework called the Eightfold Path to think about policy issues. When you are thinking in the realm of political strategy, do you use a similar framework?

With elections, it always starts with how many votes you need. There’s not a lot of debate about what the issues are—the candidate already knows that going in. But then you think through, “what’s your base?” “Are there swing voters? “What concerns do they have about you and your opponent?” You have to find ways to turn out your supporters. It’s more about inspiration than it is about technology. Voters are either motivated or they’re not. So that’s got to be your guiding light. The way you measure is by always making progress towards a vote goal. Even the money you raise has to be married to a campaign to acquire a vote goal.

In terms of public policy, it depends. Some public policy debates can seem like they’re all about making the right data-driven decision. But it’s often not that easy. It’s also about building public support and the permission structure for people to be supportive of something. So there’s the data, and there are the policy recommendations, but who will be affected by this? What’s the best human face to put on this? That’s really important. I think that blending that—the story telling with the policy—is difficult, but when it works well, that tends to be how you are successful.

Is there is similar strategy for electorate vote goals, or working in the House or the Senate?

Yes, you have to count votes. And in a close vote in the House or Senate, you’ve got to build a campaign for each swing voter—what do they care about substantively? What’s their political situation? Who would be an effective advocate to talk to them? That could be another colleague, that could be a policy expert. We even did this in the Iowa Caucuses in 2008. We had people who were concerned about energy or health care, and we had policy experts who were volunteering on the campaign and we would say, “would you like to sit down and talk with this policy expert?” and they would!

So how does that translate to the private sector? Are you seeing that you’re using the same kinds of strategies in your work with Uber?

Well, it’s a little different, but if we have a vote in the city council or legislature, yes. You are trying to understand where people initially are and what would be the strongest arguments to them. So in our case, it could be about flexible work or reducing congestion—you have to understand your audience.

If I’m a 55-year-old person who is looking for extra money and my only image of Uber is of young people, I might not think that’s for me. It’s the same thing in politics. There were a lot of people in 2008 who thought, “Obama...hmm...young...race….” So one of the ways people got comfortable was to see people like them in their community being supportive; this gave people the permission structures to come out and support him.

It’s important to note that it just can’t be the data or even the top-line message. You have to get some sense of who the audience is and what’s the best way for them to experience your argument. Many times, there’s a visual. If you can’t express your message in an infographic or short video, you’re probably not going to get it through.

So talking about consuming information, it’s hard to remember a presidential election where we have followed delegate counts and nomination details so closely. What are your thoughts on the delegate situation, and perhaps the possibility of a brokered convention?

The 2008 Obama–Clinton race was the first time that a lot of the news media covered the delegate situation. Before 2008, media coverage was just who won the state. So I would get on the phone with news directors and anchors and say, “You guys have to start covering the delegates.”

I remember our Super Tuesday—it was big: 22 states. Generally, Hillary Clinton won California, New York and Massachusetts, so a lot of the coverage was that Hillary Clinton had a better day because she won big states. Our challenge in the 24 hours after was to actually say, “we won more delegates.” It’s a byzantine process, but people certainly seem to be interested by it and they are certainly learning a lot about how we elect a president.

Will there be a contested convention? I think it’s 50–50 that Trump will not have enough delegate pre-convention, which means it will go to Cleveland. If he doesn’t secure enough votes in the first round or two, you will have something that political junkies have dreamed about for a very long time. You’ll have multiple ballots like we used to use before we had primaries and elections to decide the nominees. At the end of the day, if Trump is denied the nomination, and if Cruz is not picked, the message being sent will be that the voters had their say, but the party bosses are coming in. I don’t think this will happen. That’s a very destabilizing message.

This interview originally appeared in the Spring 2016 edition of Policy Notes.