Kevin Reilly (an executive who has worked for NBC and now works for Fox) recently made a few headlines at a conference by arguing that the fall schedule system is antiquated, and that the networks need to find a more efficient system than the current one. Of course, network executives have been saying that for years, and nothing really changes. The system the U.S. networks use is very inefficient, no question about it. To have a bunch of shows ready for September, they have to make a bunch of pilots all at once, which means they’re constantly competing for actors, writers and directors; they have to decide rapidly between a lot of pilots, and frequently pick the wrong ones almost on a whim. (There have been fewer flops this season than last, so there aren’t as many “what were they thinking?” moments. But there are a fair number of shows that were picked up for what now seem like obscure reasons.) So every few years, executives announce that they’re going to shake up this system and move to something that makes sense: get away from the fall/midseason splits, phase out the pilot system.
It doesn’t happen. It may happen someday, but it hasn’t happened yet, because there are certain advantages to the current system that other systems can’t replicate. Bringing out most of the new shows at the same time does create a glut, and makes it harder for a new show to stand out. On the other hand, most people know that the new seasons and new shows are coming in or around September. Letting people know that a new season is starting, even with the increasing number of places where networks can put their ads, is still a very inexact science. Most people don’t read articles to find out when all the new shows are coming, and even when online viewing is a bigger chunk of TV viewing, there will still be the same problem of letting people know there’s a new episode. (Not everyone is on Twitter, not everyone is on Facebook, and the U.S. networks have not yet collected all our personal e-mail addresses to send us alerts. Yet.) Switch to a cable-style system where shows can air more or less any time, and you immediately accept one of the challenges of cable, that a lot of people who might like the show aren’t aware that it’s on. Or you get the Canadian system, where shows sneak on the air and tiptoe off without anyone noticing.
So networks still need the fall roll-out, and if they need to have a lot of shows ready for September, they also need to have a lot of pilots ready earlier in the year. And they need to make up their mind about the fall schedule when they know which shows will and won’t be back in the fall. Which kind of limits the time frame in which they can make the pilots and choose which ones they’re picking up.
That doesn’t mean there are no alternatives to the current system. There are some things networks used to do that people have called for them to do again. Like actually letting people see the pilots they didn’t pick up, something that (as with Seinfeld) can identify a pilot with more potential than the executives realized, and make them change their minds. Or instead of bidding frantically on every new pitch, sometimes go back to old pitches or rejected pilots and make them again. (There’s also the frequent suggestion that networks should make more 6 and 13-episode seasons, but as I’ve said in the past, I personally think broadcast shows often have too few episodes a season, not too many.) But the actual concept of the fall season as the foundation of the broadcast TV world – I don’t know if that can change for now.
The other thing network executives have been arguing as long as anyone’s been listening to them is that Nielsen is an antiquated system, and we need to switch to a system that measures exactly who is watching what. Well, it’s true; Nielsen is antiquated, but antiquated polling systems often work about the same as the technologically advanced ones (In politics, polling didn’t really catch up with the cell phone era, but was still pretty accurate for many elections. Although there was talk that the “landline-only bias” would skew the 2010 election polls so much as to be useless, they turned out to be about as accurate as usual. Of course it helps there that the people least likely to have cell phones are among the most likely to vote.) Nielsen deserves all the criticism it gets, but when it comes from executives, it sometimes sounds like an excuse – as if a more accurate system would prove to advertisers that the flops are really hits.