White House press secretary Sean Spicer became the subject of Twitter mockery and journalistic hand-wringing Saturday when he asserted that President Donald Trump drew the largest crowd ever — period — for an inauguration. Multiple news outlets shot that claim down by comparing aerial photographs and public-transit statistics for President Barack Obama’s first inauguration in 2009 with those of Trump’s.
Spicer’s conflation of Nielsen-measured ratings and digital stats touched on an ongoing battle at the intersection of television, digital media, and advertising over how best to report viewership — and the confusion that clouds that argument.
According to numbers released Saturday by Nielsen, Trump’s inauguration averaged 30.6 million total viewers. That same day, CNN Digital released a statement saying that it had measured 16.9 million digital views of the inauguration on its platforms. In Nielsen-measured television viewership, Trump’s inauguration fell short of Obama’s in 2009 (37.7 million viewers), and Ronald Reagan’s in 1981 (37.4 million), and several pre-Reagan inaugurations. On Monday, Spicer asserted that by adding the Nielsen number and the CNN number, Trump comes out ahead of Obama and Reagan.
“I’m pretty sure that Reagan didn’t have YouTube, Facebook, or the internet,” Spicer said. He continued, “Let’s just take the Nielsen ratings, which are 31 million and add CNN, 16.9 million. That’s a little higher.”
Not that much higher. Nielsen and CNN reported numbers drawn from dramatically different methodologies. Nielsen measured average viewers per minute across 12 networks from 10 a.m. to 6 p.m. on inauguration day. CNN’s 16.9 million reflects the number of digital streams started between 6 a.m. to midnight. It does not account for how long a viewer watched or even whether single viewers clicked start multiple times. In unique starts over the same period — number of viewers who clicked start on a video — CNN measured 7.1 million.
But number of unique starts is not comparable to number of average viewers, either. According to numbers provided to Variety by CNN, from 11 a.m. to 2 p.m. the network’s digital platforms averaged 1.54 million concurrent streams. That number is more comparable to Nielsen’s, although it measures a three-hour period instead of the full eight hours that Nielsen reported on. Average concurrent streams over a longer period would likely be lower.
Spicer’s mistake in likening Nielsen ratings to digital views is one that media executives and observers often make themselves — much to the consternation of researchers. As advertisers continue to shift money to digital video, television executives complain that digital numbers are improperly reported and often unreliable. Because of the difference between the two industries’ standard methodologies, digital viewership numbers have an inflated value when compared to television numbers.
Sports, like an event such as the inauguration, is largely viewed live — and thus can illustrate well how such inflation works. Sunday’s NFC championship game averaged 46 million viewers on Fox. In addition, Fox reported 4.1 million digital views for the game. But when measured by average viewers per minute, the digital audience was only 307,000.
Trump, more so than most other political figures, has displayed a fascination with TV ratings rooted in his role as former host of “The Apprentice” on NBC. He even took to Twitter earlier this month to criticize “The New Celebrity Apprentice” host Arnold Schwarzenegger for the lackluster ratings drawn by the series premiere of the franchise’s new iteration (which Trump executive produces). “Wow, the ratings are in and Arnold Schwarzenegger got ‘swamped’ (or destroyed) by comparison to the ratings machine, DJT,” the then-president-elect tweeted. “So much for being a movie star — and that was season 1 compared to season 14. Now compare him to my season 1.”
Joking with the White House Press corps Monday, Spicer noted, “As you know we are all about big viewerships and large audiences here.”