"AdExchanger Politics" is a recurring feature that tracks developments in politics and digital advertising.
Today’s column is written by Chris Nolan, founder at Spot-On.
Regulations such as the EU’s General Data Protection Regulation and California’s Consumer Privacy Act have focused a spotlight on consumer data privacy. But there’s another conversation that needs to take place about the use of voter data and its deployment for political advertising.
As we head into the 2018 midterm elections, ad buying and selling platforms need a cleaner distinction between what constitutes political data and the customer data used in brand advertising. Political and advocacy ads shouldn’t be sold or trafficked in the same way as consumer goods and services.
Why? Well, it’s the data.
Consumer data is collected via location tracking, past purchase records, browsing behavior and other means. Online consumers now notice ads that are tailored to them. But they’re also increasingly uncomfortable with the exchange of information. New privacy laws allow consumers to opt out or at least understand what is being used for what purpose.
That isn’t the case with political data.
For the most part, voter data is a public record. What political ad buyers call a voter file doesn’t contain any information about how you voted – ballots are secret – but it does include party registration, voting history, name, age and address. Some state voter rolls include mobile phone numbers and email addresses.
Political data firms parse this information, layering in attributes like household income information from the census tract or licensing data, such as hunting permits or a medical license information, to create even more targeted voter profiles.
For now, the only way for consumers to opt out of sharing this information is to never register to vote, since targeting by political party demographics doesn’t appear to be off-limits under new privacy laws.
But targeting isn’t always as effective as political ad buyers believe. In keeping with policies established by the Digital Advertising Alliance (DAA), some information used for targeting is anonymized so political ads aren’t being sent to specific voter groups but to a larger group that shares the voters’ characteristics. This can give political ad targeting a hit-or-miss quality.
Facebook and other walled gardens, by contrast, have well-developed voter targeting, but to use it effectively advertisers must play by their rules – which go above and beyond what’s required by law.
And it’s worth asking: Is the mess that’s been created on Facebook about to play out across the open web?
As part of its attempt to avoid the missteps of the 2016 election, Facebook created its own political ad disclosure and disclaimers. The social network also fiddled with the priority it gives news and political ads in users’ feeds, resulting in much less engagement.
It hasn’t been a happy transition. Facebook blocked unrelated ads because the company name or city was identical to famous political surnames. And in California, days before the primary, candidates and campaigns saw their ads pulled down – for not complying with Facebook’s rules, even though they followed state law.
The DAA and IAB are also contemplating a political ad labeling scheme, which they presented to the Federal Election Commission in a recent public hearing.
But if Facebook’s troubled introduction of its system is a preview of what’s to come, political ad buyers and the platforms that serve them could be in for some rude surprises as a system designed to protect the use of ad targeting results in a series of missteps and further confusion.