How Long Should a Web “Season” Be?

Network TV series generally have 22 episodes per season, cable series hover around the 13-episode mark, but there is no standard “season” for web series. The prevailing theory seems to be make as many as you can — but is that always better? Given the attention span of online audiences, asking them to stick with you over countless episodes seems risky.

Roommates had 47 episodes. Quarterlife is up to 27. Prom Queen had 80 episodes, and its sequel Summer Heat had 15. LonelyGirl15 seasons one and two have a whopping combined 427 episodes.

I understand that with each episode being between 90 seconds and three minutes, it could take a ton of episodes to complete a story arc. But are people sticking around for the whole series? A rough look at the data suggests a mixed answer to that question.

To see if there are any attendance trends along the lifespan of a web series, I looked at the four web series listed above. Then, I pulled and assembled the play counts from each episode for each series into a chart. Caveat: I only pulled them from their “homebase” for traffic and did not include all of the syndication partners. For instance, I only used MySpace numbers for Roommates, quarterlife, Prom Queen and Prom Queen: Summer Heat. For LonelyGirl15 seasons one and two, data was pulled exclusively from YouTube. The rationale for this is that each source was a main (or exclusive) distribution outlet for each series, and once people knew where to watch one episode, they knew where to come back if they wanted more.

Between frolicking by the pool and test driving Ford automobiles, Roommates, pulled in 6.2 million plays over 45 episodes. A look at this graph shows that after the big spike in traffic from episode 13 (titillatingly titled “The Exhibitionist” with a thumbnail of a girl stripping in the shower), traffic settled down in the tens of thousands (the spike from the episode title “Wet and Wild Car Wash” notwithstanding).


This might be where I eat a little crow, but it looks like Marshall Herskowitz’s ode to whiny twenty-nothings is actually finding an audience. Though it too fluctuates, it seems to consistently pull in around 100,000 viewers per episode on MySpace.


Though Prom Queen enjoyed some initial bumps in traffic, its traffic really died around the episode 20 mark, with traffic only breaking 100,000 mark sporadically across the rest of its run.


Just like in the movies, maybe the sequel wasn’t such a great idea. Summer Heat went cold immediately, barely breaking 50,000 plays out of the gate and spending most of its time drawing fewer than 20,000 viewers. No wonder Eisner lost money.


Ugh. The LG Army is going to crucify me in the comments section for this one. LG data is harder to track. The show was on Revver for a while, it has a presence on MySpace, and not all videos are submitted to the “lonelygirl15” channel (in keeping with the shows intricate plot, different characters upload videos through different accounts). Since YouTube’s Lonelygirl channel had the lion’s share of plays, that seemed like a reasonable measure of the show’s popularity.

LG15‘s first season certainly held its audience. Though the spikes in traffic didn’t cross the million mark as often after episode 31, the show continued to rack up consistent six-figure plays. That’s great considering its episode count was over a hundred at that point.


Maybe it was the loss of Bree-nnocence, but the second season of LG15 definitely dropped off a cliff, with plays stuck down in the tens of thousands, and far fewer episodes crossing the 100,000 threshold. Perhaps they should have just let Lonelygirl go and come up with another show? (Yes, I know about KateModern).


Oh, don’t let that giant spike in traffic fool you. That 4-million plus performer is the most-watched LG15 video of all time, and I’m not sure what it says about our society or Internet culture (other than “duh”), but the name of that episode is “Girl Tied Up” and featured a cute girl in restraints. Here’s a revised chart without that episode.


So what does all this tell us? Well, individual episodes within a web series season spike wildly from episode to episode, which makes sense in an environment where something else entertaining is just a click away. Plus, since they are all on the web, we can access them whenever we like, at least in theory. The fact that episodes posted months ago still don’t generate plays suggests that people aren’t going back.

To get some perspective, I spoke with TV by the Numbers guru Robert Seidman. “Barring special events, the [TV] numbers don’t vacillate that much. The numbers don’t change week to week, but year to year there are changes.” Again, this makes sense for TV where the longer time format allows viewers to create an emotional attachment to characters and return from week to week to watch them.

As we all know, the online video industry is still finding its legs. We don’t have a quantifiable standard for measuring hits, and we’re just starting to experiment with episodic storytelling. But perhaps, we should make our experiments a little shorter.