Netflix Data Reveals Exactly When TV Shows Hook Viewers — And It’s Not the Pilot

Orange Is The New Black TCA

Netflix crunched cold, hard viewing data for more than two dozen TV shows and says it has determined which specific episode grabbed most subscribers to the point where they watched the entire first season.

However, none of the shows Netflix looked at, which included originals and licensed series, hooked viewers with the pilot. In fact, two shows — “Arrow” and “How I Met Your Mother” — didn’t hit the tipping point until episode 8. In the traditional TV biz, conventional wisdom holds that a show’s pilot is the most critical linchpin to igniting viewer interest, given the nature of how new television programs debut.

But don’t get the wrong idea: Netflix has no plans to use Big Data to rejigger the way TV shows get made, in order to put the strongest emotional hooks earlier in a season (which would result in more viewing by subscribers). Instead, the company sees the metrics as validation of its binge-release strategy of delivering all episodes of a season at once.

“This won’t have any direct effect on the creative process for our showrunners/creators,” a Netflix rep explained.

Netflix pinpointed the episode for each show’s season 1 for which 70% of viewers who watched it went on to complete the entire run.

For example, in “Breaking Bad” season one, the “hook” was episode 2: the one in which Jesse Pinkman dissolves a drug rival in a bathtub — and the disintegrated remains crash down through the ceiling. For prison dramedy “Orange Is the New Black” (pictured above), which Netflix execs have said is the service’s most-watched original series, it’s episode 3. That’s when Crazy Eyes (Uzo Aduba, who just won a Primetime Emmy for the role) drops both poems and fluids in the course of her imagined romance with Piper (Taylor Schilling).

See More: Binge-Viewing Is Becoming a Less Shameful Activity: Survey

“Given the precious nature of primetime slots on traditional TV, a series pilot is arguably the most important point in the life of the show,” Netflix chief content officer Ted Sarandos said. “However, in our research of more than 20 shows across 16 markets, we found that no one was ever hooked on the pilot. This gives us confidence that giving our members all episodes at once is more aligned with how fans are made.”

Netflix found slight geographic differences in the “hook” study. For example, Germans showed early fandom for “Arrow” whereas French viewers were hooked earlier on “How I Met Your Mother” than the worldwide average.

For the study, Netflix analyzed data from accounts of subs who started watching season one of the selected series between January and July 2015 in Brazil, Belgium, Canada, Denmark, Finland, France, Germany, Ireland, Mexico, Netherlands, Norway, Sweden, the U.K. and the U.S., and between April and July 2015 for Australia and New Zealand. The company noted that the hooked episode had no correlation to overall viewership numbers or viewer attrition for a particular series.

Here’s the full list of shows Netflix analyzed, with the number of the episode when 70% of viewers were hooked (all series are the first seasons):

  • “Arrow”* – Episode 8
  • “Bates Motel” – Episode 2
  • “Better Call Saul”* – Episode 4
  • “Bloodline” – Episode 4
  • “BoJack Horseman” – Episode 5
  • “Breaking Bad”* – Episode 2
  • “Daredevil” – Episode 5
  • “Dexter”* – Episode 3
  • “Gossip Girl” – Episode 3
  • “Grace & Frankie” – Episode 4
  • “How I Met Your Mother” – Episode 8
  • “House of Cards”* – Episode 3
  • “Mad Men”* – Episode 6
  • “Marco Polo” – Episode 3
  • “Orange Is the New Black” – Episode 3
  • “Once Upon A Time” – Episode 6
  • “Pretty Little Liars” – Episode 4
  • “Scandal” – Episode 2
  • “Sense8″* – Episode 3
  • “Sons of Anarchy”* – Episode 2
  • “Suits”* – Episode 2
  • “The Blacklist”* – Episode 6
  • “The Killing” – Episode 2
  • “The Walking Dead”* – Episode 2
  • “Unbreakable Kimmy Schmidt” – Episode 4

* Denotes series that are not currently available in all Netflix territories.

Filed Under:

Want to read more articles like this one? SUBSCRIBE TO VARIETY TODAY.
Post A Comment 23

Leave a Reply


Comments are moderated. They may be edited for clarity and reprinting in whole or in part in Variety publications.

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

  1. Leonie says:

    Where can I find the original Netflix study (source)?

  2. vobricen says:

    How to get away with murder: episode 4

  3. jess says:

    Seinfeld: Episode 24!

  4. marieugorek says:

    This should honestly not be a surprise to anyone. Pilot episodes have to introduce SO much basic material about characters, world, etc. that fitting in an emotional hook as well is pretty much going to fail. Add to that the fact that emotional hooks are more effective when viewers have actually had some TIME to develop empathy (or antipathy) for characters and it should be obvious that most shows will do best if the big stuff shows up in episodes3-8.

    This is also part of why some of the best series I have seen (Firefly and Miracles, e.g.) were cancelled before a full season was broadcast: if you have a detailed world and complex characters AND you screw with viewing times and/or mess up your marketing so the people who want to watch it don’t get to see at least 2-4 of the first 5 episodes, network execs are going to make bad calls. I firmly believe that those shows would have stayed on the air if introduced on a Netflix platform, where they would be suggested for those who like that sort of thing and would NOT be bumped around at the whim of less important stuff like sports events.

    • A show where the concept itself is the strong hook might be an exception. iZOMBiE is an example (Netflix just got season one so they didn’t report data on it); I expect that viewers will either be hooked immediately or not at all.

  5. macemoneta says:

    Once I’ve determined that a show’s subject matter is of interest (from the trailer or description), I usually give the show three episodes. If I’m not hooked by then, I move on. Even for some of the shows listed as taking much longer, I’ve been hooked earlier – so the hook isn’t the same for everyone.

    If a show takes more than three episodes, the story arc or presentation isn’t properly developed. That’s just a bad sign in any case.

  6. Hookme says:

    This is cla”ssic (and wrong) Hollywood thinking:

    “Netflix has no plans to use Big Data to rejigger the way TV shows get made, in order to put the strongest emotional hooks earlier in a season (which would result in more viewing by subscribers).”

    Emotional hooks don’t work in a vacuum and they aren’t something you can just move around. They work when you’ve gotten to know the characters and their world enough to understand and care about them. The statement about makes about as much sense as a restaurant saying “diners rave about our desserts so we’re going to start serving them after the appetizers.”

  7. Bill says:

    This is based on what demographic? Out of all the shows listed the only one I saw all the episodes so far is Better Call Saul. I have seen some of the Walking Dead and Sons of Anarchy.

  8. canadagraphs says:

    I’ve always had a strategy to my show watching. I give a show 3 episodes… even if the pilot is junk, to prove it self. lots can change between a pilot and eps 2 and beyond incl cast, writers and show runners. So I like to give it to ep 3 to show me what it has.

    So far, this method has never steered me wrong (for a 1st season at least). A couple times I’ve been on the fence after 3, so i’ll give it 3 more. By ep 6 it’s either going to have fully grabbed me, or probably never will.

  9. mike lawless says:

    I see a lot of eps 2 and 3, which, on Netflix is a little like a premiere. I remember my first binge OITNB , I want 5 straight,

  10. J-dog says:

    There is no goofy algorithm for this “study”. The first episodes of a new series are always going to have the most viewers. People don’t start in the middle of a new series or skip episodes on Netflix. All they are identifying in this study is at what point does most of the erosion stop (we stopped the bleeding from this point on) — it isn’t rocket science.

  11. J-dog says:

    This seems backwards to me. Aren’t they really just measuring the episode when erosion finally stopped happening (not the magical episode when people started watching).

    • They may be measuring additional things as well, like when people start binge-watching (viewing multiple episodes continuously or nearly so) rather than one at a time separated by days.

    • PG says:

      Certainly looks like an ‘end of attrition’ rather than ‘worth finishing’ point – but honestly… isn’t that splitting hairs? When attrition ends, those left (by definition) finish. So the raw data is essentially the same.

      However, the specific conclusions become suspect. It ceases to be the case that Episode X is necessarily a good one, and may instead suggest that X – 1 is underpar. Usually the idea/advertising/subsequent word-of-mouth is the most critical thing: if no-one hears about or is interested in even watching the pilot, the show is sunk. *Then* the pilot “simply” has to just not drive everyone away, and, ideally – but secondarily – be a good follow-through on the promised idea/hype.

      It’s surely been a fairly common-knowledge semi-fact that it’s the second and third episodes (as borne out here) that are key: the secobd sets more clearly the budget and thrust of the show after the flashier/more expensive pilot, and the third episodr is where everyone starts to find their feet after the pressure of the pilot/first and upheaval of the second.

      It’s also an unmentioned point that “pilots” and “first episodes” are not always identical. The pilot sells the show and draws the initial audience, but (historically) may be retooled – or reshot, in parts or whole – and only then become the first episode of the first season. So the pilot isn’t always the key to anything..

    • I’m assuming that they looked at all of the members who watched a series’ first season, and took out the ones who watched every episode. Then they looked at the average or median episode the people who didn’t finish the season, and said the “hook” was the episode after that?

    • nerdrage says:

      I think they decided to undermine Amazon’s announcement today of their new pilots by casting shade on the very idea that pilots are important. So they looked at how many people persist in a series after the pilot, and found that it was around 70%. They arbitrarily decided 70% was the magic number (why not 65% or 88% or any other random number) to “prove” what they wanted to find.

      As far as we know, pilots hang onto 69.999% of viewers and then it falls a tick to 70% in the second or fourth or whatever episode. Without seeing a time sequence for all the episodes and what percentage of the final episode audience (representing 100%) is present at each episode, we know both jack and shit about what is going on, other than Netflix’s PR team is working overtime.

    • JoeMcG says:

      I’m sure it’s not that simple and that their logarithm measures a variety of metrics to arrive at this number. And I think this logic tracks… a pilot is only a proof of concept. The key is whether subsequent episodes can sustain that concept, and in the case of many hit series, end up improving on it. A number of hit series didn’t hit their stride until later episodes, and even in later seasons. What’s backwards is how quickly a network will kill a series without giving it a chance to blossom.

  12. 85wzen says:

    Well if you’ve ever been ‘locked up’ it may be less amusing to watch a prisoner crouch down and urinate on the ground. But yeah, after that I just didn’t feel right about the Orange show…

  13. Nelly says:

    House and Friday night lights had amazing pilots.

  14. Geoff says:

    not hooked by the pilot? clearly they didn’t test mr robot

More Digital News from Variety