People meters start today - a new era in Baltimore TV
Today is the day that they start measuring TV audiences differently in Baltimore. As one general manager of a Baltimore TV station put it in a memo to his staff: "Television starts anew in Baltimore today! Let’s move out of the past – and position ourselves to be the leaders of the future!"
What's new is the use of Local People Meters, an electronic device, to measure viewing patterns of Baltimore area residents.The new Nielsen system replaces a way of counting viewers that relied in part on diaries that some researchers considered unreliable and out of date given new digital technology.
The general manager's memo went on to succinctly described the advantages of the new technology this way: "The local people meters include a larger sample and daily extensive demographic data that was not available under the traditional set meter ratings. The new LPMs allow multiple individuals in the same household to record their television viewing simultaneously. The sample includes for the first time cell phone only homes...."
But the meters have also brought shakeups in the ratings and controversy in some other cities, like Miami where they introduced this spring.
I have written about the switch to people meters for the Sun and this blog. Here is some of what I said in a Sun story last month after explaining that Nielsen ratings are the primary source of data used by station to sell advertising time, and therefore central to the business of local TV:
Previously, Nielsen relied on a decades-old method of handwritten diaries. The new People Meters offer instant, real-time demographic data that rules out the variable of faulty recall by viewers in filling out their diaries.
But as good and technologically improved as Local People Meters have seemed to be on paper, their arrival in other cities has often caused major disruptions in viewing patterns, complaints from stations and even lawsuits.
What tends to happen in most cities is that the market-leading station or stations see their audience ratings drop. Sometimes, they even fall from the top of the heap altogether.
Local People Meters were introduced in Miami in October, and last month, WSVN, the Fox affiliate, filed suit against Nielsen, claiming that flawed data from the People Meters will cost the station $12 million in revenue this year.
According to the lawsuit, the station says that ratings for its 10 p.m. newscasts fell 50 percent after the People Meters were introduced, while ratings for American Idol, the hit prime-time show from Fox, fell 40 percent.
"Nielsen's imposition of the Local People Meters in the Miami-Fort Lauderdale market has been an unmitigated disaster for its broadcast television customers there," Sunbeam Television Corp., the owner of WSVN, said in its lawsuit. "From the outset, Nielsen's LPMs produced defective, wildly inaccurate ratings data, which literally overnight, created havoc in the market."
Speaking of what happened in other markets, Don Lowry, vice president of government and public affairs for Nielsen, said in a Baltimore Sun interview, "There have been shifts. I wouldn't call them dramatic. But there have been shifts that as a television operator you would definitely pay attention to."
Lowry said that company representatives have installed the full 600-home sample of Local People Meters in Baltimore and have been offering the stations here test data for several months that are consistent with the ratings derived from the older methodology.
But while Lowry said he didn't expect any major disruptions, sources familiar with the Baltimore test data say that it has shown considerable change from what came before. So far, the People Meters in Baltimore have generally served up lower ratings for WBAL and WJZ - and even offered data that would have reversed the standings in one key news race during the May sweeps.
Station managers say they are prohibited by contract with Nielsen from discussing the test data publicly.
I am not naming the general manager who wrote the memo quoted at the start of this post. Nor am naming his station or reporting any of his specific suggestions to his staff. I think it unfair to give competitors such information.
But, in a general sense, viewers should look for more animated, higher energy local newscasts starting today with a tighter feel and quicker pace. People Meters instantly measure tune-out in a way that diaries could never record.
More than ever, stations need to grab fast and hold onto viewers if they want to score well with the new technology.
As for the new ratings data, general managers at Baltimore's network owned and affiliated stations were unanimous in cautioning against making too much of what the People Meters tell us in July. This week, for example, is a holiday week with strange viewing patterns that further accentuate the traditional abnormalities of local TV viewing during the summer.
We will have to wait until September to get a real fix on the new technology, but we'll be on it long before that at Z on TV.
Comments
In this day and age of technology, why can't this data be gathered directly from the cable boxes? This would eliminate the human factor of bad record keeping. It may also expose the commercials that cause people to immediately change channels. I should think that would be useful feedback.
Posted by: Neo | July 2, 2009 11:22 AM
I can think of a couple reasons why using cable boxes, as suggested by Neo, might be problematic.
First, by definition, it excludes anyone that doesn't have cable... like those that receive over-the-air broadcasts, or those with Fios or satellite dishes.
Secondly, it opens the door to monkey business on the part of the cable company. They could possibly alter results for those shows/networks in ways that would be most advantageous to their company.
That said, I think it's a wise step to bring an automated system into play. Having been a Nielsen diary holder before, I can see that there is a great deal of room for inaccuracy in manually filling out diaries. Filling out the diaries after the fact, as you mentioned, leaves room for memory gaps. They ask for entries at 15 minute intervals. Any channel-flipping during commercials is likely left un-recorded.
Secondly, there is always the possiblity of intentionally misleading entries, either to make the diarist look better... like avoiding listing the embarrassing shows that some of us would rather not admit to watching. Or in the same vein, one could prop up a particular show or network that the viewer is particularly fond of to a greater extent than that which would be depicted by his actual viewing.
Or worse, one could be offered compensation to watch a particular show or network by someone with an interest in the results.
It comes down to the fact that a Nielsen diary holder can put anything they please in the diary, whether it reflects actual viewership or not, and there is no oversight or deterrent. With the black boxes, assuming they function correctly, one would have to go to greater lengths to fudge the results. But at least the TV would actually be on. And as Neo mentioned, it could allow for a little finer analysis of viewing habits, like which commercials or segments of shows get turned off.
Thanks for the story, Z...
Hi Tony, Yes, and thanks for this analysis. I am sure it will help readers better understand the technology and issues involved. Z
Posted by: Tony N | July 2, 2009 2:24 PM
Actually, there are industry initiatives underway to better use data from cable, satellite and TiVO set-top boxes. The biggest problem is that while they do provide great data on what is being watched and when commercials might be skipped, they do not collect the important demographic data (i.e. who is watching) that the new Nielsen local people meters do.
Z, you might not be aware that Arbitron is starting to measure Baltimore radio audiences this month with their new electronic portable people meter (PPM), replacing diaries. It promises to change the radio landscape as much as the new Nielsen meters will do in TV.
Jay Guyther
Hi Jay, Thanks much. This is great information -- especially on Arbitron. Z
Posted by: Jay Guyther | July 2, 2009 4:36 PM
Jay and Tony sure must have done their research. Their blogs were great. I have used those cumbersome diaries here in Ohio. Any electronic way of measuring the results would have to be more accurate in my opinion. I wanted to say to all of you over there in Baltimore and beyond have a great 4th of July from an Ohioan! I really am glad I found this blog. Wow what about the bombshell that Sarah Palin dropped today? She set off some fireworks of her own.
Posted by: Sherry T. | July 3, 2009 6:21 PM
Many folks I know do not watch TV this way. They watch on demand - particularly on the computers. These boxes will do nothing for that.
Posted by: Lauretta Nagel(ConstellationBooks) | July 3, 2009 8:51 PM
Actually, the LPM device and the Nielsen sample can measure what's being watched and when for those individuals who are watching via DVRs or OnDemand. Viewing that takes place after the initial live airing is generally called timeshifting.
The LPM device can determine viewership that is not a live broadcast...Nielsen refers to this as Live +3 (viewed within three days of digital recording) and Live +7 (viewed within seven days of digital recording).
Not to get too technical, but the LPM device is connected to the TV's audio system. Every network and station embeds audio signals differently with their own signature source coding. Think of it as each station's digital finger print. Based on the embedded audio coding, during playblack the LPM device can determine which show is being aired.
Also, the LPM device will note if the TV is in DVD or gaming (XBOX, Playstation, etc.) mode. Not what you're watching or playing per se, but the mode in which your particular television activity is currently engaged.
Posted by: Bill J. | July 21, 2009 1:07 PM