Facing the DVR’s Future

Mar 15, 2004  •  Post A Comment

In the TV advertising business, even seemingly innocuous changes in the methods used by Nielsen to gather and report its TV audience ratings can have a profound effect on the way advertising is planned, bought and sold.
With Nielsen now ushering in an unprecedented array of changes-some representing major shifts in the way it measures and even defines TV viewing-media planners are bracing for a fundamental shift in thinking about TV media planning.
Most acknowledge that many of the changes are necessary, even overdue, but that doesn’t alter the fact that they will have a substantial impact on how agencies calculate TV audience impressions.
Among the changes under way, or scheduled to take place in the next year:
* The rollout of local People Meters.
* The gradual integration of local People Meters into an expanded national People Meter sample.
* Reweighting of Nielsen’s national ratings sample designed to make it more representative of the actual population.
* The introduction of Nielsen’s new A/P metering system designed to measure digital TV transmission.
* The inclusion of viewing in households with digital video recorders (DVRs).
Each of these would be considered a huge development in its own right. Combined, they represent a virtual zero-basing of TV audience measurement that has many worried how these changes will impact the stability of the data and, perhaps more pragmatically, its comparability to historical TV ratings patterns. Without that, they say it may be difficult, if not impossible, to trend TV audience patterns, which is essential for planning TV advertising buys.
Of the changes, the plan to incorporate time-shifted viewing via DVRs may be the most profound, changing almost everything about how people think of TV audience ratings.
“On July 12, 2005, everything will change,” said Nick Schiavone, former head of research at NBC and now a consultant for Fox’s cable networks and DVR marketer TiVo, referring to the date the first time-shifted Nielsen ratings report for national TV will be released. (The change will actually impact local TV markets three months earlier under Nielsen’s plans, though it still is unclear how Nielsen will deal with DVRs in the multitude of diary markets.)
Among other things, Mr. Schiavone said, the industry will have to come to grips with how it deals with two sets of data. One, based on the initial “live” viewing in households unaffected by DVR usage, will constitute an initial release of ratings in much the same pattern they are issued by Nielsen today: overnights and weekly reports. The second set of data will be based on any viewing that was impacted by DVRs, which will be delayed for seven days, after which the data will be integrated into new, final national ratings for the previous week.
During its national client meetings last week, Nielsen said all DVR-impacted viewing-regardless of how long it was impacted or even when it was played back-would be excluded from its “live” ratings reports, a decision that sparked some frustration and outrage from some clients who believe that any “near-live” viewing-viewing that occurs within the same day that a program was scheduled-should be included in the live viewing reports.
That would be a significant number, said Kate Lynch, senior VP and global research director at Starcom MediaVest Group, which has conducted its own research with TiVo to get at the magnitude of it. She said most “TiVo-ed” programming is viewed in “near-live” mode. According to a study conducted by Starcom and TiVo in April 2003, 75 percent of all recorded viewing occurs within 48 hours of the time the original program aired.
Even more significantly, she said the near-live viewing is more likely to occur with the most popular network TV shows, the ones that are most likely to have the highest ratings under Nielsen’s current TV ratings system and the ones that tend to command the highest advertising rates in the marketplace.
“It’s the `ERs’ and the `Friends’ that will be most affected,” Ms. Lynch said, referring to Nielsen’s plan to delay reporting on the time-shifted viewing of these shows by a week.
The findings suggest profound implications for how agencies plan television advertising buys by daypart and genre. Not surprisingly, the data indicates much lower “commercial defection rates” for live programming, such as audience participation shows, awards and pageants and news programming, than for dramas and sitcoms.
As a result, the immediate impact of Nielsen’s DVR measurement plan would be to substantially lower the initial ratings of some of TV’s highest-rated shows. And since a significant portion of the viewing that is thrown into Nielsen’s new time-shifted basket will also suffer from reduced commercial exposure, demand from advertisers and agencies for these shows could also suffer. According to the Starcom study, an average of 53 percent of TV commercials are skipped in DVR households, but the ad zapping is much higher-77 percent-for the portion of programming they view on a recorded basis.
While the data released by Starcom did not explicitly address the impact on television dayparts, it could have a profound effect there too, since certain programming genres such as daytime soaps are known to be heavily recorded and viewed back at later times. While these programs and dayparts likely benefited under Nielsen’s rules for measuring VCR recording, which credit “viewing” when a show is taped, the shift to playback under Nielsen’s new DVR rules could begin the hurt these same shows.
But the whole notion of dayparts, and even TV scheduling, is likely to erode as Nielsen begins implementing its DVR scheme, since viewers essentially will be creating their own schedules and viewing programs independent of traditional dayparts.
These changes are more than just semantics. They will challenge essential precepts of the marketplace. Audience “shares,” for example, will lose virtually all meaning as a substantial portion of the DVR audience will be shifted out of real-time viewing. A share point is the percentage of households or viewers that are tuned to a given channel when programs are scheduled to run. Audience shares “will become a relic of the past,” predicts a top network research executive.
The move will also obviate the meaning of overnights, he said. Because an increasing number of ratings will be delayed for a week, the validity of overnight ratings will erode over time.
Nielsen estimates that DVRs are in 3.5 percent to 4.0 percent of TV homes and cited forecasts projecting they could be in 10 percent or more by July 2005.