• by Joe Mandese  @mp_joemandese, December 8, 2006

Less than a year after it won an important TV marketplace victory - convincing the networks to guarantee their advertising deals based only on real-time TV viewing - Madison Avenue is poised to lose its so-called "live" TV ratings. Exactly how and when "live" ratings will die isn't yet known, but they were officially put on the death watch Thursday morning during a special client meeting hosted by Nielsen Media Research to determine how time-shifted viewing should be reincorporated as part of the TV industry's advertising currency.

During that meeting a consensus appeared to emerge between both buyers and sellers that a combination of live and time-shifted viewing that occurs within one or a few days of an original telecast, might be the best way to go. The consensus emerged after Nielsen revealed new research showing that the share of viewing occurring via playback on digital video recorders is growing faster than many people might have thought.

"The data you just showed is stunning," said NBC research chief Alan Wurtzel, declaring, "The future of half of all viewing is time-shifted. If not now, certainly in a year or two, 'live' is going to be a completely irrelevant measure."

Wurtzel was referring to Nielsen data indicating that more than 40% of all prime-time viewing by adults 18-49 is done in playback mode in DVR households. While DVR households currently represent only about 11% of Nielsen's sample, the impact is significant enough that playback now represents 5% of all viewing done in the average U.S. household. And it is growing fast.

"It seems to me that if we are going to actually move to some kind of commercial-based measurement, 'live' is going to go away - live-plus seven is going to go away," concurred Steve Sternberg, the head of audience analysis at Interpublic's Magna Global unit, referring to two of the three data streams Nielsen has begun distributing to the industry since it began reporting DVR households in January. The third stream is live-plus same day of DVR playback, which could emerge as the new standard for advertising deals, or potentially a new stream including "live" plus two or three days of playback.

The reason, executives said during the meeting, was that Nielsen's analysis shows that the predominant amount of playback occurs within one or three days of the original telecast.

The meeting was intended to help Nielsen decide which data stream it should use for reporting its new average commercial minute ratings, when it begins releasing them early next year on an "evaluation" basis, but the decision on which stream is most relevant to the TV advertising marketplace ultimately is expected to shift the basis of advertising deals from "live" only to include some portion of playback.

While the industry has not yet agreed to switch to commercial ratings, that shift would resolve one of the major concerns advertisers have with DVR playback: fast-forwarding through commercials. Because Nielsen uses audio codes to detect audiences, fast-forwarded minutes would be excluded from those data streams, because they emit no audio signal.

Much of the debate during the meeting centered on how, and when the commercial ratings data should be released, and ultimately incorporated as trading currency for TV advertising buys. Several executives said they wanted to wait until the data was vetted and proven to be accurate. Others said the marketplace was shifting too rapidly and that the move should be made now and in time for next summer's upfront advertising marketplace.

"Everyone saw the data up there. I personally do not think we can have an upfront negotiation that does not take into account the impact of DVRs," said CBS research chief David Poltrack, adding, "CBS is definitely using this data as part of its upfront strategy. We're never going to be in a position to model the next 10% of DVR penetration based on the current 10%. It has to be part of the upfront."

Questions surrounding the accuracy of the new ratings data are not inconsequential, said Mediaedge:cia's Lyle Schwartz, suggesting that bad data might be worse than no data, because it would force agencies and networks to spend additional time verifying their accuracy and correcting mistakes.

Earlier in the week, during a presentation the UBS media conference in New York, CBS' Poltrack also pointed the financial impact of the new ratings data, noting that following the shift to "live" only ratings currency, the major broadcast networks have not been getting "paid" for about 7% of their audience - the portion of their audience that is currently played back on DVRs.

During the Nielsen meeting, Poltrack made another observation suggesting a profound battle looms in the future over another important goal of Madison Avenue's: The shift from Nielsen's current "average commercial minute" ratings - ratings based on an average of all commercial minute appearing in a TV program - to individual commercial minute ratings.

Citing a new, but not widely distributed report from Nielsen analyzing the "standard error" of average minute ratings vs. individual minute ratings, Poltrack said the individual minute ratings would be too unstable to use as the basis of advertising deals that are guaranteed by the networks to advertisers.

Poltrack said the standard error for individual commercial minute ratings was three times higher (17.6%) than for average commercial minute ratings (5.1%), according to the new Nielsen report.

"In a marketplace where we pay off on the downside and don't get anything on the upside, that standard error costs us money," Poltrack noted, referring to the fact that advertisers do not currently pay the networks for bonus audience delivery.

"If you go to individual minutes, you should also go to [paying] off on the upside and [paying] off on the downside," he said.