Last year, Nielsen announced that it would be adding “broadband-only homes” to its television ratings sample and viewing universe. This category consisted of people that not only didn’t subscribe to cable television, but didn’t even have an antenna to watch broadcast television, and thus couldn’t watch any programming on any platform that Nielsen normally measures, so their inclusion in the sample must have seemed superfluous and useless. As a result, ratings, and the estimated universe of people that could watch cable channels, fell. On the other hand, Nielsen also announced that starting this year, it would begin including online viewing of content in its TV ratings… so long as the ad load on those programs was exactly the same as when it aired live.
This unusual outcome is the result of the tension between Nielsen’s actual role in the television industry and the role it inadvertently fills as a result of it. Television networks pay Nielsen to tell them how many people are watching the ads accompanying their programming, because the ads are what are paying for the programming and the people who buy ad time want to know if they’re getting their money’s worth and where they should spend it if they want to. For most of Nielsen’s history, that meant measuring how many people were watching the programs, and as such Nielsen became the barometer for how popular America’s TV shows were.
As time-shifting became more popular, however, and as Nielsen’s measurement practices became more refined, these two purposes became increasingly at odds with one another. Today the currency in the TV industry is “C3”, or how many people are watching each minute of commercial time either live or within the first three days of DVR playback; some media buyers this year have adopted “C7” as their currency, which is exactly what you think it is. In other words, if you fast-forward past the commercials, your viewing counts for jack all to the networks even if you’re in a Nielsen household. Neither of these are widely reported, but it doesn’t matter because most people do, in fact, fast-forward past the commercials, and waiting the amount of time it takes for the C3 or C7 ratings to come out isn’t always practical (especially if you have a ratings flop on your hands), so the live-plus-same-day ratings that are widely reported are good enough for most purposes. (Nielsen’s definition of “live” is so restrictive that there are enough same-day viewers watching enough commercials to be useful.)
Nielsen’s move to counting broadband-only homes is a direct response to criticism from outside the TV industry that Nielsen dramatically undercounts the true popularity of many shows, especially in the most valuable demographics, by not counting viewership on alternative platforms besides live TV and time-shifted DVR – an attitude that expects and assumes Nielsen to be primarily concerned with its role as barometer of shows’ popularity. But in order for measurement of online viewing to be in any way relevant to the networks that are only concerned about who’s watching the commercials they sold for those shows, Nielsen has to impose the bizarre “same ad load” requirement, which no network or online platform would put in place without the incentive of being counted in Nielsen ratings, preferring dynamic ad insertion techniques that can adjust based on a viewer’s location and Web browsing habits. I try to stay away from authenticated TV Everywhere services, but I did have occasion to use my Dad’s account to use WatchESPN recently, and I found that even there, even while watching the live feed of an ESPN channel that is supposed to be no different from watching it on television, the ads were not the same as on TV, meaning no one using WatchESPN could be counted in Nielsen ratings. Heck, there were one or two commercial breaks where no ads were inserted into the feed, and I still wasn’t getting the ads that were being shown on television, just a placeholder slide.
It is certainly true that the model of television on which Nielsen is based is becoming outdated, but the reality is that Nielsen shouldn’t have had to create such contortions to count online viewing towards its TV ratings, because no matter how many viewers aren’t being counted, as far as the networks are concerned, Nielsen is working exactly as it should. The problem is not that Nielsen is falling short on the goal it doesn’t really have to serve as barometer of the popularity of television shows; the problem is that that role is still relevant even though Nielsen should not really be concerned with filling it. The problem is that the success or failure of television shows is staked to a system that, structurally and by design, can only capture a fraction of its popularity. And this is not a problem with Nielsen, but with the networks.
The vast majority of big-budget, big-studio shows are still widely assumed to need a place on a linear television network’s schedule, to be underwritten by the network and distributed by them to the network’s audience. The network, however, only cares about the show – or at least, should only care about the show – insofar as the show can attract people to the advertisements they can intersperse throughout the show. If not enough people are watching it live to serve as a captive audience for the commercials, the network can and will cancel the show. If a show is on network television, its existence is dependent on the commercials the network airs, or else the network can cut bait and abandon the show, potentially driving it out of existence no matter how popular the show may be on platforms that don’t expose their audience to the same commercials.
Shows should not be dependent on this system, on networks that will stake the show’s existence to a particular set of commercials inserted into the network’s feed. The presence of a show on a linear television network, and thus a show’s ability to attract audiences to a linear network’s commercials, should not be a precondition for a show’s existence; rather, a show should have a presence on a linear network only if that network has reason to believe that they can sell commercials off it and attract the show’s audience to those commercials by giving them a reason to watch it “live”. We’re a long ways away from the day when a show’s presence on linear television is a recognition of its value to the network rather than a precondition of its existence – we’ll know that day has arrived when a show that originated on the Internet moves to linear TV rather than the other way around – but we’re at least seeing halting steps towards throwing off the tyranny of the linear networks and of Nielsen, through the original shows on Netflix and Amazon and through Yahoo’s recent move to give Community one last season. The arrangement between CBS and Amazon for Under the Dome also frees that show’s fate from being dependent on the Nielsen ratings, though as it happens the show has done quite well for CBS, especially for a summer show.
Just as I don’t think linear television is necessarily completely obsolete in the age of the Internet (and it may in fact be of paramount importance, if lessened compared to pre-Internet days), so I don’t think Nielsen needs to worry about its core business going under; even with the prospect of broadcast linear television colonizing mobile devices, given the appeal of that prospect to the consumer and the basic nature of the technology there will always be a place for Nielsen’s measurement methods so long as the transmission of advertising isn’t dependent on a two-way connection over the Internet. I only hope that, so long as linear television remains the primary mode of video consumption, Nielsen does not overly hobble the prospective future where it is not, and that by the time that future arrives both networks and ad buyers (and to some degree the public) will be fully aware of Nielsen’s limitations.