From The Futon Critic:
Cancelled: An Inside Look at Ratings and Why Shows Get Axed
December 17, 1999
The one question I get asked the most is "Why did they cancel show (blank)?"
It's probably the most difficult question to answer in a way that would set the askee (is that a word?) at ease. After all, I know what they're going through. I've seen numerous shows I love die miserable deaths (if you must ask, "Cupid" and "EZ Streets" still ring the hardest) and know all too well the frustration that comes with it.
In all seriousness the chief reason shows get canceled are low ratings. Now I'm sure those two words are bringing out a groan the size of Maryland, after all what is a "rating" anyway? "I have never met a Nielsen family nor am I convinced they even exist," someone wrote to me once. Truth be told they do, in fact I know very well someone who has been one. So scratch that theory.
Actually I'm sure people's frustration comes from the fact that Nielsen Media's (the company that gathers the ratings information) methods and data aren't wide public knowledge.
That fact alone is key. You and I shouldn't have access to the list of Nielsen families. It could corrupt the data by having networks or producers sending reminders or free stuff to them in exchange for viewership. But I'm getting a little ahead of myself.
Nielsen ratings are a measurement of the number of people watching a particular channel at a particular time. This information is divided into total viewers, rating and share. Total viewers is essentially the number of people watching the show. The rating is equivalent to 1 point per 1,008,000 sets watching the show (notice that this isn't viewers, just television sets, meaning there can be more than one viewer on each set yet the rating doesn't change). And the share is the total percentage of sets watching the show. This information is used to determine who are the audiences and what they are watching. From this advertising rates are determined (that's how they make TV free you know) and guarantee levels are set. Basically NBC and the other networks say that for "X" amount of money, show "Y" will get at least rating "Z." For example, NBC might say for $100,000 a 30-second-spot during "Friends" will have a 25.8 rating. If they meet that they break even; if they go above, they get a bonus; and if they don't meet it at all they pay back the difference to the advertiser. Starting to see how important Nielsen data is (and more to the point why networks can't afford to keep low rated shows on the air)?
Now this all might seem a bit complicated (or convoluted) so let's take it step by step.
Let's start with the method Nielsen Media uses to get the ratings data. It's called statistical sampling and the idea behind it is this - since it's inefficient (or more to the point impossible) to poll everyone with a television the next best thing is to "sample" a given population, that is poll a small cross-section of a metropolitan area in order to find out the entire population's makeup (what they're watching). Hence the Nielsen "family" is born. Each household has its television sets outfitted with a box that monitors their viewing habits and each member of the family is given a diary to fill out what they actually watched.
Now I'm sure this theory looks full of holes for the moment but bear with me. It might seem like that by only looking at a small percentage of an entire population you'd think you would miss people (more importantly fans of a particular show might feel exluded, like their "vote" doesn't count to give an analogy). The truth is though that it can be proven mathematically that if you sample a given population's viewing a set amount of times you can within 99% accuracy find what the whole population's viewing. Trust me on this folks I took a class on it in college (and did well in it too
).
Now what about the 1% or so margin or error. That's certainly a factor to be looked at but the truth is 1% of television audience isn't going to make that much of a difference. If a show really sucks in the ratings it sucks and 1% won't really save it. If that's the case (it's a close call) there are other factors to look at which we'll get to in a minute.
So exactly what information are these Nielsen people acquiring then? Well, essentially what's on the TV sets of the people in the Nielsen families. The most common method used is a box that sits on top of your television that records every remote click you make (I know, how Orwellian). In essence, it records exactly what you are viewing. Now there's two important factors to consider about this method that doesn't make it 100% solid.
Let's say you're a "Snoops" fan (all three of you) and you tape the show and watch it on Monday instead of watching it live Sunday night. That viewership is not included in the Nielsen data. Why? Well, since Nielsen data is primarily used to set advertising rates (the more viewers you reach the more money you get) it's important that the viewers be there for the commercials (which people fast forward through when they tape). If anything this is a key flaw in the ratings process. I rarely watch television live yet I spend a solid 10-15 hours a week watching TV on tape. Should my viewership count then? Yes and no. While acknowledging taped viewership would dramatically change the ratings makeup it would not change the number of people watching the commercials. This then is why taped viewership should not be considered (in theory anyway).
My thoughts on the whole thing is that such information (that is, taped viewership) should be accumulated albeit separate from the "live" viewership. I say this since if say ABC finds out 30% of its total viewership for "Snoops" tapes the show that means it might be best to move the show to another night. Acquiring taped viewership data lets networks know how many potential "live" viewers they may have. Chances are a series with a high total viewership rate but low live viewership rate on Fridays and Saturdays would do better elsewhere in the week in live viewership.
The second is that just because a TV set is on doesn't mean people are necessarily watching it (or more to the point watching the commercials). You could be eating dinner, having sex, asleep and left the TV on, etc. So just because a TV is "on" doesn't guarantee active viewership. This however isn't considered that much of a factor since the area of most concern (that is channel flipping during commercials) is offset by the Nielsen box recording if the commercials are being shown. Advertisers are generally willing to risk the fact people watching the actual show watch the commercials or at a weighted percentage (hence the member diary's usefulness).
Now with all these if, ands and buts you're probably wondering what's the point of measuring all the data at all if there's so many factors that could spoil it or misinterpret it. The matter of fact is that it's the best we have and if you consider the mathematical stability of statistical sampling (say that three times fast) you're going to get a legitimate measuring stick as to the viewership number and demographic.
Whew! Now that we have all this information down and hopefully digested, let's look at how this information is applied. Since it's obvious that the networks don't take the top shows and cancel the bottom ones across the board, let's look at how at some other factors.
1) Network viewership.
Simply put NBC has more viewers than UPN. A good part of the fact is that NBC is carried on 99.9% of the nation's airwaves while UPN has only 70% or so. Also, NBC has been around since the beginning of television and people know the network and trust it. My 6-year-old cousin is older than UPN. This means that UPN likely won't have a show in the top 20 any time soon. Because of this UPN will renew shows that are near the bottom of the Nielsen ladder. 5 million viewers and a 3.0 rating wouldn't fly on NBC but on UPN it's a success. So don't think that just because NBC's "The Mike O'Malley Show" ranked higher than just about every show on WB and UPN that it meets NBC's expectations.
2) Night by night viewership.
Again, this is pretty simple - more people watch television Sunday through Thursday than on Fridays and Saturdays. This means that shows on Friday and Saturday don't get as high of a rating as shows on Wednesdays and Thursdays. People go out on the weekend and don't watch as much TV, hence a more weighted view is placed on shows that air during those nights. (Why do you think NBC loves "Providence" so much? - It's on a crappy night and still does killer ratings.
)
3) Demographic viewership.
People always wonder why shows on the WB are considered hits and yet they don't place above 80 in the Nielsen rankings. The reason is demographically speaking, WB gets more 18-30 year old viewers than any other network on a regular basis. That demographic is considered the ideal to advertisers (you know, get them hooked while they're young), hence they are willing to pay just as much to shows that do well overall as they do for shows that do well in the younger demographics but overall don't do as well. In fact, I'd go so far to say that advertisers prefer the WB's shows because they know they are guaranteed to always have a younger viewership. There aren't too many 50-year-olds watching "Buffy the Vampire Slayer."
4) Lead-in viewership.
Another wonder of the TV audience is why shows like "Union Square" and "The Single Guy" got canceled even though they were in the Nielsen top 10 (outside of the shows being vapid holes in creativity). The answer is lead-in viewership. If a show is highly rated but loses 25% of the audience from the half-hour before it, it negatively effects the shows it "leads in to." The idea is that if people turn off a network because they don't like the show, there's less of a chance they'll come back for shows later on. This means that the networks want to make sure you're watching their network instead of channel flipping (where gasp! you could find something else to watch). Essentially, just because "Jesse" is ranked 8 this week doesn't mean it's a success. If it's hemorrhaging 25% of "Friends" audience it won't be around for long.
5) Syndication.
This is where the networks can be accused of being kind of shady. With each new season a series comes closer to qualifying for syndication, that is the possibility of selling repeats of the show to local stations outside of prime time. These are those "Seinfeld" repeats that air on your local FOX or WB affiliate at 5 or 6 p.m or wherever they want. And you know what, the production companies make millions off of selling shows, more than they make when they actually produce it. Jerry Seinfeld got $175 million when he sold "Seinfeld"'s syndication rights. This is big money folks. Basically then if a network has a stake in a show (many networks produce part of their show lineup), then they are more inclined to keep it on the air in hopes of getting it to the syndication level. What's the syndication level? Either 4 full seasons or 100 episodes, whichever comes first. A show isn't allowed to air on another network (there are exceptions we'll get to in a second) until this level is reached. Shows that are canceled before this level can still be sold into syndication but are less likely to get a big paycheck. The exception to this is that a production company may ask for an early syndication package as part of the deal of airing it on a network. This is why "Once and Again" airs on Lifetime and "Law & Order: Special Victims Unit" airs on USA. The producers made a deal to have it air on the other networks in order for ABC and NBC respectively to keep the show instead of going elsewhere.
So basically if "Suddenly Susan" was in it's third full season and doing so-so in the ratings there's a better chance it will come back knowing that NBC gets a nice payoff at the end of another season. (You mean that could happen? Oh wait, it did.
) People generally get upset about this since it shows favoritism (why would they put an NBC produced show on Saturday night when they could put it on Thursdays?) and arguably it does. The trouble is that it's just like any other business property - do you want your own stuff to succeed or other people's?
6) Cost.
Money makes the world go round so it should be no surprise television works in a similar way. It's no secret that science-fiction series are often the most expensive to produce, hence they are not a largely represented genre. On the flip side, newsmagazines ("20/20," "Dateline NBC") and reality series ("Candid Camera," "Whose Want to Be a Millionaire?") are among the cheapest to produce hence their often overwhelming presence on many networks. Why would a network want to run a marginally successful drama series when they could get similar if not better ratings with a newsmagazine in its place. While such philosophies may stifle one's need for creative programming, it is however a logical business manuever.
7) Politics.
Just like any other business television is susceptible to all kinds of internal and external politics. Networks are likely to be more kind to producers that have brought them hits in the past. Many people scratched their heads as shows like "Millennium" and "seaQuest" were renewed despite abysmal ratings. The answer is likely it had something to do with Chris Carter giving FOX a hit with "The X-Files" and "seaQuest" coming from Steven Spielberg. This is the hardest of all areas to speculate on since it's generally the first people look to instead of considering a show's renewal or cancellation from an objective standpoint.
So what have we learned class?
We learned about how ratings work and what they are used for.
As much as we'd all like to believe there's a conspiracy against each of us to cancel our favorite shows, the truth is 9 times out of 10 there's a legit reason behind it. The ratings system might not be perfect and even the implementation of it is fallible (that is until those brain implants start rolling out
), but it's the best we've got.
So keep steadfast my television friends, just like everything else in life, enjoy it while you have them. And if the networks try to pull some crazy shit on you, I'll be the first to scream it from the mountain tops.
