Nielsen Has A 'LOT' More Problems, Goes To New Lengths To Fix One

In the latest in a series of debacles stemming from Nielsen's effort to begin including ratings for time-shifted viewing this year, the TV researcher has informed clients that all of the data it has released on an increasingly important metric - length of tune, or "LOT" - is wrong. The data, which calculates the average amount of time viewers are tuned to a network, program or daypart, is used by some advertisers, agencies and networks as a surrogate measure of "engagement," an increasingly important but often elusive effort to understand how deeply consumers are involved with the media and advertising content they are exposed to.

But in a series of notices sent to clients last week, Nielsen said all of the length of tune data it has issued since Dec. 9, 2005 is "incorrect." That's a problem for a number of cable networks, which have already made upfront sales pitches based on that data, and for advertisers, agencies and other networks who've incorporated the data into their 2006-07 upfront strategies.

advertisement

advertisement

"We understand that length of tune data is important for upfront presentations," Nielsen acknowledged in its first client communiqué disclosing the SNAFU last Wednesday, adding that the glitch was "being treated as a top priority."

In another communiqué sent on Friday, Nielsen identified the source of the problem, an error that occurred when it made changes to the way it manages its data in preparation for its shift to reporting time-shifted viewing, including ratings for digital video recorder playback, and said it was working quickly to correct the situation.

Nielsen spokesperson Anne Elliott late Friday told MediaDailyNews that Nielsen's systems would be updated to re-run the length of tune data for household ratings effective today, and that by May 1, the database would be corrected for all demographic data as well.

She added that the glitch, which uniformly lowered length of tune estimates during the period, should not have impacted the relative relationship among networks and programs.

"Basically, the stories don't change. If you were the stickiest network [before], you're still the stickiest network," she said, borrowing the term "sticky," which is normally used to refer to the amount of time online users spend on a website, and one which increasingly is being used by the TV industry to understand television engagement.

While the data glitch appears to be uniform, it throws another wrench into the works as the TV industry begins pitching advertisers and agencies for their 2006-07 advertising budgets, including some networks, such as Court TV, known to be pushing engagement.

Media buyers could not be reached at presstime for comment, but several network executives said they were pleased that Nielsen went public with the information and was taking steps to correct it quickly, and before 2006-07 upfront negotiations begin in earnest. The head of research at one major cable network said he did not expect any "collateral damage" as a result of the SNAFU, but added, "If Nielsen got this wrong, it makes you wonder what else might be wrong."

Earlier this year, Nielsen's shift to time-shifted ratings uncovered other strange anomalies, including situations where ratings that added time-shifted viewing were actually lower than those with out them. That problem was attributed to the way Nielsen "weights," or factors, various components of its sample.

Next story loading loading..