How political media buyers misuse programmatic

For a lot of political operatives “programmatic” means one-to-one targeting and that’s about it. In the context of political advertising, there’s been a heavy reliance on such microtargeting — the use of extensive online data to tailor persuasive messages to voters. However, it’s a strategy whose utility has been lately challenged, not least in an MIT-led study published in 2023.

It’s worth saying that microtargeting in politics works in a different way than it does for brands who advertise clothes, cars, or chocolate. It’s more challenging to generate the data needed to train the political targeting models and distribute the resulting ads, according to David Rand, an MIT professor and co-author of the study.

Nevertheless, we in the political industry tend to just use the same tool over and over again. It's like trying to build a house with just a hammer. Why do we do this? Fear of losing deals and credibility by standing out from the herd. 

There is a focus on microtargeting because many of us are taught that it’s the smart way to achieve efficient victory. 

Nothing is more prized in this extremely competitive, impossible to evaluate business than demonstrating that you belong in the category of “smart” operatives. So once the conventional wisdom around what is smart forms, it becomes nearly impossible to dislodge. 

Most people think the smart way to use programmatic (and, in fact, all digital) is to only deliver to a one-to-one audience matched to the voter file. But as the available inventory on connected television has grown thanks to major streaming platforms such as Hulu and Peacock, so too the need to rethink this conventional wisdom.

We can do this by adopting a more layered approach that combines individual, contextual and geographic targeting across screens and ad units in a way that is less wasteful. In short, this new approach is likely to yield an incremental audience of voters that otherwise may have been missed with a more tightly focused lens. Top DSPs offer a broad suite of targeting and inventory selection tools — and provide campaigns both immediate scale and endless optionality in delivering advertising. And this means new opportunities for political operatives.

We as an industry have followed the belief that microtargeting is the key to not only being a smart operative, but also winning contracts and receiving the hollow adulation of friends and rivals who all conform to the same strategy. Though, in fairness to all of us, this is also driven by the historic need to fight for budget over wider audience and less nimble TV firms. 

After all, Karl Rove rose to prominence while targeting segments of wedge voters with direct mail, but the broader atmospherics — which included him pushing for gay marriage bans in eight states — had a much larger impact on his political trajectory. And, we were told, the success of both Obama presidential campaigns was built on the then cutting-edge use of social and digital media to finely target voters and deliver them the exact message they needed to hear to embrace Hope and Change. And yet, by 2008, the Republican brand was in the toilet, and the Obama campaign doubled down by allocating the vast majority of its budget to broadcast TV ads attacking the Republicans.

So what if all that microtargeting talk was more about all of us proving our bona fides as digital smarties in a credential-less industry than it was about matching media plans to the objective of winning the campaign? 

After all, the always difficult to prove match rates have been falling as privacy wins on the internet. And as the definition of “digital” keeps expanding, more and more of the top-quality ad networks do not allow custom audience targeting for some political buyers. And yet, we continue to insist that one-to-one is the way. 

Following digital buyer dogma despite the evidence was sort of not that big of a deal when digital played a supporting role. But now that digital is playing a leading role in many political campaigns (thanks to the fact that it’s slowly consuming linear television) it’s becoming a major problem. 

Microtargeting precepts that worked fine in display and preroll are proving catastrophic in streaming. The prioritization of targeting over inventory has led political buyers to intentionally buy lower quality CTV inventory rather than higher quality streamers because they want their delivery to match their over-promising on a spreadsheet. In a connected TV household like my own, where there’s typically more than one person watching, that strategy alone could spell a missed opportunity.

Any political buyer who has prioritized one-to-one targeting in an open exchange buy, assuming they look at the domain-level CTV reporting, knows this is true. But because there is so much reputational sunk cost based on promising that you can deliver TV ad units with display ad targeting, few political ad buyers are willing to be honest about it. 

Instead of trying to justify some always faulty promises, we should be embracing the power of being able to deliver quality television-equivalent ad units into much tighter geos than a cable zone — and being able to do so based on the actual real-time media consumption choices of voters rather than our prescribed view of what they watch. We can think of this approach as a way of blending a legacy media approach with the targeting virtues of digital. That’s what I mean by layering.

As “digital” moves from tightly targeted advertising to more or less everything, we need to adjust the way we think. If digital media planners in politics continue to think narrowly about what we should do in our medium, we’ll continue to play a supplementary role to linear buyers. We’re artificially limiting ourselves to serve an outdated shibboleth while the industry is trying to hand us the keys to the kingdom. Maybe we aren’t so smart after all. 

Andy Barr is Managing Director of Uplift Campaigns, an agency that provides creative, ad buying, data and software to Democratic campaigns, firms and organizations throughout the United States. 

This op-ed represents the views and opinions of the author and not of The Current, a division of The Trade Desk, or The Trade Desk. The appearance of the op-ed on The Current does not constitute an endorsement by The Current or The Trade Desk.