Depending on who's doing the counting, NBC's Feb. 4 broadcast of Super Bowl LII was either the lowest-rated NFL title tilt in nearly a decade or the most-watched program in the history of TV. And while there's a good deal of ambiguity surrounding the deliveries of TV's last great reach vehicle, buyers are not at all equivocal in their belief that the era of soaring Super Bowl inventory rates has come to an end.
According to Nielsen live-plus-same-day data, NBC's coverage of the Eagles-Patriots joyride averaged 103.4 million linear TV viewers, the smallest turnout for a Super Bowl since 2009. Toss in the 2.6 million fans who streamed the game on various NBC and NFL platforms, and Super Bowl LII delivered 106 million total viewers—a more respectable number, even if it still adds up to a nine-year low.
On Thursday, however, NBC revealed that another 12.2 million viewers had taken in the game at a bar, restaurant or someone else's home, bringing the overall TV audience to a record 115.6 million viewers, topping the high-water mark (114.4 million) the Peacock itself set three years ago with Super Bowl XLIX. Add the streaming figure, and now advertisers would appear to have reached an average audience of some 118.2 million viewers.
So which of the three totals goes into the books as the official Nielsen tally? On the one hand, NBC should get credited for the additional eyeballs that took in the action from out-of-home venues or via non-traditional platforms. After all, there's no empirical evidence that indicates that those who watched the game while planted atop a barstool paid or at a friend's Super Bowl party were any less engaged with the commercials than the vast majority who tuned in from the comfort of their own family rooms. On the other hand, there's no way to compare NBC's out-of-home numbers to those from previous years, as Nielsen only began measuring that subset of the TV audience in 2016 and last year's Super Bowl host, Fox, didn't start subscribing to the out-of-home reports until four months after the Big Game.
Perhaps more to the point, when calculating the cost of generating 1,000 ad impressions, most buyers don't factor in the out-of-home deliveries. Many media buying agencies are pushing back on accepting guarantees for such deliveries, noting that the cost of reaching the bar/restaurant/gym/hotel crowd has always been baked into the negotiated CPM. (The networks don't furnish ratings guarantees for the Super Bowl, but buyers are still very aware of what they get for their money.)
According to Lyle Schwartz, GroupM's president of investment, the CPM for Super Bowl LII worked out to $109.48, a figure derived from the broadcast's live household rating and an average unit cost of $5 million per 30-second spot. For the sake of comparison, Super Bowl XLVIII, which aired Feb. 2, 2014, on Fox, fetched an $81 CPM. In other words, in just four years, the cost of reaching 1,000 Super Bowl viewers has skyrocketed 35 percent, while the average household rating and number of viewers who tuned in dropped 8 percent.
Schwartz says the dizzying increase in the cost of reaching the Super Bowl audience has begun to have a chilling effect on how the game is priced. After Fox first crashed the 110 million-viewer mark back in 2011, the average unit cost for an in-game spot began growing at a near-metastatic rate, jumping from just under $3 million that year to $4 million in 2014. Another three years saw the unit cost leap an additional $1 million—whereupon all that giddy forward progress ground to halt.
"I think you're seeing a certain amount of push-back from advertisers because while the unit price has gone up substantially, the CPM went up even more," Schwartz says, noting that the $1 million price hike between 2014 and 2017 amounted to a 25 percent increase, while the cost of reaching 1,000 viewers in that same period jumped 30 percent. "Regardless of the overall size of that Super Bowl audience, you're still getting less for more."
Buyers say NBC's $5 million Super Sunday price tag was more or less in line with what Fox charged in 2017, an assertion supported by Kantar Media's preliminary ad sales estimates. Kantar eyeballed NBC's in-game take at $414 million, slightly shy of the $419 million Fox scooped up during Super Bowl LI. When pre- and post-game spend are accounted for, NBC's overall haul should easily clear $500 million.
Schwartz believes that until there's a vast improvement in how viewership is quantified, pricing is likely to continue to be impacted by the perceived slackening of interest in the Super Bowl. "The way we measure it is not accounting for the multiplicity of the ways in which people are watching it," he says. "I think for the most part a lot of the 'losses' on the traditional TV side would be made up for by this alternate viewing."
Anyone who knows Schwartz's history understands that he's not exactly waiting around for the industry to solve the measurement crisis. Back in 2006, when he served as managing partner and head of broadcast research at GroupM, Schwartz was instrumental in implementing the C3 ratings currency which, to this day, remains the coin of the realm for buyers and sellers of TV. In much the same spirit, he is now looking to develop a more agnostic approach to gauging exactly how many viewers are watching the programming... and the ads, naturally.
"Much like we did 10 years ago, we're working to better understand a device-independent audience," Schwartz says. "We have been working with Nielsen, to some extent, and ComScore and our network partners in order to move to the next stage of commercial ratings, the measurement of audiences across all platforms, within a seven-day window. … We have to figure out how to measure and monetize all this inventory that's not currently being accounted for, and we need to move fast."
