As YouTube Show So Shall YouTube Reap

I can't wait for pilot season next year. Which network will have the guts to post any of its pilots on YouTube and then use viewer feedback to determine which shows to pick up for the Fall season? We'll find out whether ''Nobody's Watching'' had any effect at all.

I reckon that no networks or production companies want to be so bold, but I can hope. Can't I? If I'm in charge of the CW, I might try it to fill my last sitcom timeslot this way. That netlet has absolutely nothing to lose, and it can only gain viewers by giving the masses a sense of empowerment. Forget about next season, I would let the netizens choose my mid-season replacement sitcom.

Here's what I would do. Post three pilots and advertise widely that you're going to let the viewers select one of these three. Tell the production companies their shows definitely won't be picked up if they don't participate. Hell, I might even stack the deck the first year with one really good pilot and two crappy ones just to test the process.

The infrastructure is already there, I just need to have a few analysts comb through the data after a two- or three-week trial period. I don't know how much it costs to assemble multiple pilot screenings, but I bet that I could save money by paying YouTube to host the event and YouTube would generate a much-needed revenue stream for themselves.

Think about it. I can track behavioral data as well as comments and ratings with the existing features at YouTube. Here are some metrics to consider:

• # of views for Part #1: This metric measures the number people who found the premise interesting enough, the star(s) appealling enough or a friend's recommendation compelling enough to make an effort to view the video.
  
• # of time shared: This number should correlate with the potential word-of-mouth advertising. Are viewers keeping their guilty pleasure a secret or do they want the world to be watching with them?
• average viewing time: Of course, this metric simply measures viewer retention. Are viewers going to switch channels within minutes? During the first commercial break? Or watch all the way to the end? (If average viewing time isn't available, then use # of views for Part #3 divided by # of views for Part #1.)
  
• ratings distribution: Average rating isn't good enough. I'd much rather have a show that has 80% 5.0 ratings and 20% 0.0 ratings than a show with 100% 4.0 ratings, even though both shows have an average rating of 4.0.
  
• comment keywords: I'm sure each network already looks for keywords — both negative and positive — when it reads viewer feedback forms from live-audience pilot tests. Now it's even easier to find these keywords since all feedback is in electronic form in the comments section.
  
• # of times favorited: This metric should be correlated with the projected size of the show's core audience.
  
• average age of members that chose show as favorite: A show can only be favorited by YouTube members, and prospective members must provide their age when they sign up. This metric should confirm whether or not the show targets the appropriate core audience. (This metric might not be available, depending on YouTube's privacy policy and anything the viewer might agree to before watching the video.)
  
• # of results for title searches: Use Google/Yahoo!/MSN or Google blogsearch/Technorati or other search engine(s) to measure just how much people talk about the show. It's one thing to markup the video as one of your favorites, but it's another thing to tell the world about it on your blog.

Les. If you're reading, please let the CW be bold sometime soon.