I keep seeing stories about how the premiere of The Beast on Thursday drew 3.6 million viewers. And while it's true, it took two episodes to do it. The actual premiere at 10pm averaged 2.4 million and a re-broadcast of the premiere immediately following the premiere averaged 1.2 million. I've seen this method of counting used for other cable shows from Damages to HBO's True Blood and Showtime's Dexter.
I don't really have a problem with the accounting for shows like The Beast and Damages, as it seems to be the case that the advertising is sold across the multiple airings. But it is misleading in terms of actual viewership. It also begs the question of opportunity cost. Repeats of CSI: Miami frequently average more than 1.2 million on A&E, though unfortunately I don't have the 18-49 break outs for those airings (or what it costs A&E to air them, for that matter) to make any definitive conclusions based on opportunity cost.
My fear is that the practice will be universally embraced by the PR mavens and make its way into the mainstream where everything gets added together. We'll see headlines like Scrubs draws 11.26 million. Interestingly, a rebroadcast of Tuesday's 9pm Scrubs drew more viewers when it aired on Thursday at 8pm - it's not surprising since it didn't have to go up against American Idol, the second time around. The 9pm edition on Tuesday averaged 4.628 million and then 6.634 million as a repeat on Thursday at 8pm.
My problem is where does it stop? Will A&E start putting out releases like CSI: Miami seen by 36.63 million? That's how many the 30 episodes of the show added up to for the week of January 5-11. And if you think that's beastly, you haven't seen anything. NCIS totaled up to 51.27 million over 17 episodes for that week, and that too is...nothing beastly.
The dominator, by any metric, is the yellow and porous one, weighing in at one hundred and eighty-two million, two hundred fifty-one thousand for the same week over 102 airings (63 on Nickelodeon, 39 on Nick Toons). That's right 182.251 million. The Beast indeed.
While I understand your point, comparing the spin on The Beast to trying to spin CSI: Miami or Spongebob the same way is not an equal comparison. When they do combine episodes it’s of the same episode. Those 30 episodes of CSI: Miami or 102 of Spongebob were not all the same episode. (There probably were some duplicates, but not enough for it to be 30 million or so.)
But I agree, it’s rather silly. I mean, they won’t be able to sell a 30 second spot for a show getting 3.6 mil when it’s really only getting half.
Julia, as I said “where does it stop?” everything above that was comparisons of the same episodes. I wouldn’t be at all surprised if duplicates of SpongeBob did over 30 million, but I’m sure you’re right about CSI: Miami.
Did NBC and MSNBC did the same with the election coverage last year right? I saw a PR saying they were #1 but with the combined numbers.
My mom got her Fox e-mail newsletter recently and it boasted that
“‘American Idol’ outperforms the competition with a stellar season premiere (over 71 million viewers) this week!”
and
“27 million people tuned in to see the clock start ticking on a new day on ’24′ as President Allison Taylor (Cherry Jones) takes the oath of office.”
That is not just public relations spin, those are flat-out lies.
Cool, I think it was Telemundo, too.
Demonhog, it’s probably just the total audience measurement (those who watched at least 6 minutes), and not flat-out lies. We report “average audience” numbers typically (number of minutes viewed divided by duration) and rarely see total audience numbers.
Cool, the dreaded (yet common!),”The Networks of NBC“
Re: Robert – I assumed that they combined the (total audiences of the) two nights.
What it looks like the networks might “pay under the table” sort of speak to boost their horrible shows.Making it seem more people are watching than what the REAL truth is so people tune in to watch their shows.They know tv is half dying people are sick of the trash they are seeing and tunning out!
Interesting article Robert. I knew “the yellow and porous one” was prolific, both with airings and viewers, but those figures are insane!! Might we have the Law & Order of the next decade here? (On a side note, there must be a bunch of repeated episodes in that batch, as there are only 113.5 episodes produced according to wikipedia; individual segments, regardless of length, currently total 218) Ridiculous!
I don’t necessarily see a big issue with the way these numbers are being generated. At home I know Scrubs was about to be an issue since the main airing will now be against both Fringe and The Mentalist which both get DVR’d (acceptable verb yet?). But Comcast has it on demand (in HD, sweet!), so either the VOD or those reairings would’ve had us as first-time viewers. And if the networks sell ads for these shows based on multiple airings and the numbers that come with them, what’s the big deal? It doesn’t seem all that different from including Live+7 numbers in the final data for each show. Plus rebroadcasts are likely to get a chunk of viewers watching live, so the ads would definitely get some more eyes, maybe making the viewership at least as relevant to advertisers as the DVR numbers (where many are skipping ads anyway, I forgot the percentages on that though).
Another example I recall was back in the ’90s. We were live-watching (and VCR-archiving) Star Trek: Voyager (bash away if you must, but I was only in grade school, and it was fun to watch at that age!), but there was at least one time when a power outage kept us from doing either for a new episode. Fortunately, UPN/WPWR rebroadcast every episode on Saturday nights, so we were able to watch without waiting months for the rerun. I would say those eyes were relevant to the total viewership of that episode (if we had been a Nielsen house).
I understand your concerns about extending this kind of math to all airings of a show, regardless of the number of episodes. However, if I were an advertiser, I might want to know that my spots might be seen by 180 million “viewers”, even if it’s really only a few million distributed through the 100+ airings. That’s still a lot of exposure. But it shouldn’t be the sole figure being put out about syndication.
Frankly, I’m more troubled by the way the whole Total Audience Measurement thing works currently. Especially for non-news/sports programming, I’d like to see Nielsen determine how many people actually watch the entire episode, not just show “average viewers” through the duration. A lot of hour-long shows have acts that run beyond 6 minutes, so those numbers aren’t going to be wholly accurate to advertisers since some viewers may drop in and out for one act and never see an ad. And I’d like to think that creators and producers would have some interest in knowing how much of their “average viewership” is actually devoted to watching the show. If they can figure out how many people are watching for at least 6 minutes, they should be able to calculate how many are staying for the whole hour/half-hour/etc.
For those who are crazy enough to still be reading this (yeah, I mean you Robert!), I guess we can agree that the whole system does still has some kinks to work out. It’s functional, but I think we can do better. Hopefully Nielsen will continue to improve things as we go along. And hopefully PR people will resist the kind of practices you mention. Even though we all know they won’t!
Double-counting is an excuse for poor ratings.
It’s clear that The Beast underperformed A&E’s expectations and for that reason, they double-counted.
Jason, I don’t mind the total audience measure to the degree that it’s stated that was what’s measured. I like the average audience measure, because it counts total engagement for a show (total minutes of viewing). But the advertisers — at least the ones who pay for them — see how many people watch the commercials. Nielsen does measure that but that data is almost never released into the wild.
The cynic in me thinks Shane probably has it right. though the 10pm premiere of The Beast will be A&E’s #1 program for the week, it’s surely its most expensive too.
I just remembered that there is actually a Nielsen rule that allows two broadcasts to count as one as long as all the ads are the same. If you’ll recall, this was an issue in the beginning of last season when the Heroes premiere’s numbers were officially reported at a very inflated number because of this rule. Any chance this is a case of that? (I don’t agree with the rule either way, but it would let the spin people off the hook for this one.)
Julia, after all the bruhaha last year over NBC’s attempt to do that with Heroes, in the end Nielsen decided to abandon the practice altogether. I’m pretty sure the ratings points for shows like The Beast and Damages are actually sold across the multiple airings, so in terms of ad sales I’m not sure it matters. I was just whining about spin
Gotcha Robert. I wasn’t aware that Nielsen had that data, good to know it’s out there. And agreed on your response to Julia above, spin sucks. But then that’s what sites like this one are good for, cutting through the spin and b.s.!
Well, glad that the rule was abolished. It sucked.
How do they track whether the 1.2 million viewers of the second showing are not part of the 2.4 million people who watched the first showing? It’s possible that only 2.4 million ever saw the show.
Lawrence, in the case of the “average audience” measurements cited in the post above (for all of the shows) they don’t, average audience measure is not stripped of duplication across multiple airings since it’s only measuring viewing of a particular airing.
I chose to record two programs live at the same time as the first showing of Beast but recorded Beast later only because the other shows were NOT available at the later time slot.
A better idea is the broadcast networks need to rebroadcast the PRIME TIME shows a second time and count all the views of both showings as first time viewers even if DVRd and watched later. Too often viewers have to chose between as many as three GOOD shows that compete at the same time causing GOOD programs to be cancelled even when a viewer base exists to make the program viable if available at a different time. A second showing of PRIME TIME over night would be valuable and more revenue producing than infomercial in the middle of the night.
Thats nothing. Jericho fans have invented ratings theft. Jericho airs 7-8pm so, they claim that the “8pm” ratings are tied to Jericho, not the 8-8:30pm average for the first half hour of the Sunday night movie that most know it is.