Every major election cycle in the U.S. brings with it a volcanic eruption of media. With the presidential primaries popping off all around us, you don’t need to be reminded of the sheer urgency with which campaigns have been pushing out their messages. But just so you know, political advertising is going to get even crazier very soon, as heated campaigns on the state and local level jump into the fray.
All of this media attention, and the volume of campaign messaging coming through ad channels, makes the political programmatic advertising space a really interesting one to watch right now. To get a fuller sense of the nuances of political programmatic, and both the complexities and the benefits of playing in that space, I reached out to Eli Kaplan. As a Founding Partner of digital ad agency Rising Tide Interactive, Kaplan works closely with political campaigns, nonprofits and advocacy groups. He’s also the Chief Strategist for programmatic buying shop DSPolitical, which he described as a “sister company” to Rising Tide and a “joint venture” with data firm Catalyst. At OPS on June 7 in New York City, Kaplan will be talking about ad tech’s very compelling role in the 2016 election, but in the interim, AdMonsters wanted to take a minute to ask what Rising Tide and other political programmatic players have been working on and why it matters.
BRIAN LaRUE: How much of a challenge it is target and schedule political programmatic buys in certain geographical areas, when demand can change so quickly?
ELI KAPLAN: If you’re trying to do a direct buy in a relatively small geographical area—an inefficient media market, a state House race, or something like that—an individual publisher isn’t going to have enough scale for a direct buy to even be worth the time. The programmatic space was designed to fix that problem. But as you noted, the flip side of that is it can be difficult to estimate the number of impressions you’re going to be able to serve in a specific geography and timeframe.
We work on enough DSPs to have the ability to turn levers and get additional scale when we need. I’d prefer to deliver in full than to leave a little extra money on the table. The nightmare scenario in politics is that you lose a close election by a very small number of votes, and have to go back to the client and say, “We have all this money that we actually weren’t able to spend, and if we did, maybe you would have won.”
BL: To that end, what are some of the other major challenges you face—especially challenges brands might not face in programmatic?
EK: The ability to have performance metrics that are meaningful in real time in a persuasion campaign is very difficult, compared to what you’re able to do with a constant feedback loop of conversion data. When we’re doing fundraising or acquisition campaigns, we have that feedback loop and can optimize very quickly. What we’ve seen, based on a variety of studies, is that the click-through rate has very little correlation to the effectiveness of a persuasion campaign. We get polling results back every so often, but obviously those polls reflect what’s going on in TV, in earned media, and in mail. It’s often very difficult to parse out the digital element.
BL: To get back to available inventory, what might publishers do to make inventory available and price it to their best advantage?
EK: So many people don’t vote, and so many are completely set in their ways, with no chance of persuading them. Enabling buyers to book first-party data against impressions is very important. There’s such a premium placed on data and audience targeting in the political space. You can have really powerful, high-impact inventory, and if you’re not giving buyers a mechanism to line that inventory up with the people who matter, you are losing out on political dollars.
BL: I’d seen a little about how scarcity of inventory can be a challenge in political programmatic, because of, for example, the environments buyers want to or don’t want to appear in. To what degree is scarcity really a problem?
EK: It really depends on the state. You have a bunch of states this cycle—I can’t think of a better example than New Hampshire—where you have a competitive presidential election, a competitive governor’s election, a competitive US Senate election, and a state that’s not particularly efficient for television. Inventory’s going to be very, very tight. In states that aren’t competitive in the presidential contest, I think it’ll be very easy to get high-impact pre-roll inventory in a programmatic media environment.
BL: In terms of the inventory you’re seeing, how are things different in this election cycle than in 2014 or 2012?
EK: In 2012, there were not a lot of tools that made it easy for a relatively unsophisticated buyer to purchase programmatic media, especially video. But, if you knew what you were doing, 2012 was actually a pretty good year—there were a lot of places where you could be ahead of the game.
In 2014, there was a lot of video—and there was also a lot of fraud. We place a premium on first-party data, for a lot of reasons. Avoiding fraud is one of them. If you have a list of likely persuadable voters in a particular election, and you’re matching them to cookies, it is very difficult for a bot to pretend to be Eli Kaplan, who lives at a particular address. It’s very easy for a bot to pretend to be male between the ages of 30 and 40. The industry’s gotten a lot better at cracking down on fraud since 2014, and I think that’s good for everyone in the long run.
Obviously because there was no presidential election in 2014, there was a lot less spending. If you’re in one of these states that’s competitive on the presidential level, it’s going to be tougher this time to find inventory.
BL: What kind of new opportunities and challenges have come up in video, and how are you navigating that?
EK: This is so new to clients that there’s a lot of education that needs to get done. Video is similar to the TV format political campaigns are used to. But on platforms such as Facebook, you don’t actually have to sit through the video and listen to the sound in order to get to your content. It requires difficult conversations with clients about why it’s a much better deal to buy inventory that costs more. So many publishers are putting video inventory out there that is not what you would traditionally think of as a high-impact pre-roll video commercial.
BL: I’m curious what publishers can do to work with demand partners in the political programmatic space to make things more efficient and to find more value.
EK: Well, I’d think if you’re a publisher, you probably want to sell direct as often as possible because you get a higher margin. But truthfully, at least in politics, you have to make your inventory available to book first-party data against it at scale. In practice, individual publishers probably aren’t going to have enough inventory booked against first-party data segments to be worth salespeople’s time. To get the scale of likely voting ticket-splitters in a particular geography that I need to talk to , I would need to aggregate hundreds of publishers.
BL: What’s the role of private marketplaces in this election cycle? It seems like they might offer some solutions to some of the pain points you’ve mentioned. Plus private marketplaces are generally more mature than they were even in 2014.
EK: You can get better inventory. You can book first-party data against it. If you’re in one of these states where there’s a lot of demand for video inventory and you’re just looking for everything on the open exchange, there’s not a lot of great stuff out there. And, if you’re going publisher direct, you lose a lot of the targeting you need for your buy to be effective. We’re a big believer in these private marketplaces and we think they’re going to play an enormous role in the 2016 cycle.