After a whirlwind week in Cannes, and three bleary-eyed connecting flights, I was in the back of a NYC cab when I saw it. Running at the bottom of the backseat screen were the words: “Trump Starts Using Online Ad Software to Target and Track Users.” After months curiously waiting for a headline on the presidential candidate’s digital advertising campaign, I found one. Imagine my excitement!
My first order of business was to figure out if Trump’s campaign had bought ad inventory through the Index Exchange platform. Sure enough, from April to June, Trump ads have passed through the exchange. This meant it was prime-time for another election post. Today, we’re comparing how the Clinton and Trump campaigns have spent from January to date.
The graph above shows Clinton and Trump campaign spend from January 2016 to date. As you can see, Trump was absent from the exchange for the first few months of the year. In April, he showed up, suggesting the digital campaign set up their tents. Through the month of April, the Trump campaign spent about a quarter as much as Clinton’s. Though a fraction, it was major influx of spend, considering Clinton’s upward trajectory starting in March.
Looking into how the campaigns accessed programmatic inventory was interesting also. Trump relied on one DSP (I’m afraid I can’t tell you which one), while Clinton bought through six. Trump allocated all of his spend in the open market, while Hillary put 3% of her total spend into private marketplaces.
Trump’s impression footprint was relatively smaller than his spend, compared to Clinton’s. His campaign bought a seventh of Clinton’s total April impressions. As we’ve seen before, the value of Clinton’s impressions are lower – she gobbles up buckets of impressions on the very low end of the Index Exchange CPM spectrum. Trump’s higher spend on fewer impressions shows the impressions his campaign purchased, on average, were at a higher market price. This suggests they are more valuable impressions: impressions are more valuable when they are in prominent places on the page (i.e., viewable), for more valuable ad inventory (i.e., bigger), or appended to some sort of demographic/psychographic/behavioral data. It seems Trump may be targeting users with more sophistication than Clinton in the programmatic market.
In the table below, you’ll see Trump and Clinton’s average CPMs from January 2016 to date. Trump was bottom feeding in April – trying to get his message anywhere he could. The CPMs jump in May and June, which suggest after presuming the nomination he started targeting and buyer inventory with higher value.
Though it may seem small, Trump’s spend in the programmatic exchange after months of zilch is significant. It suggests the campaign has a digital marketing team in place that understands the value and importance of the programmatic frontier. Trump became the presumptive nominee on May 4, 2016. Perhaps the campaign wanted to use programmatic to push himself over the delegate threshold. Can’t you picture him girding his team to wrap up the primaries, musing “there has to be something else we can do”, and finding programmatic? Surely, it will be exciting to see what happens when Election 2016 proper blasts off come July.