Netflix Ratings Revealed: New Data Sheds Light on Original Series’ Audience Levels

Charlie Cox Daredevil

Research firm measures viewers of 'Daredevil,' 'House of Cards,' more

One of the media business’s best-kept secrets isn’t entirely confidential after all: Netflix audience ratings.

Select content companies have been turning to San Diego-based Luth Research, which has assembled a sizable panel of Netflix subscribers in the U.S. as a means of determining the most popular programs on the streaming service.

An analysis shared exclusively with Variety revealed the numbers for many of Netflix’s most recently launched original series — including which new shows already seem to be challenging “House of Cards” in popularity.

Daredevil,” the first of multiple superhero dramas coming to Netflix as part of a deal with Marvel, premiered April 10, and is seeing strong sampling, with an estimated 10.7% of subscribers watching at least one episode in its first 11 days on the streaming service.

By way of comparison, the third season of “House of Cards,” which premiered Feb. 27, attracted 6.5% of subs over its first 30 days of availability. New comedy “Unbreakable Kimmy Schmidt” also bettered “Cards” in its first month (7.3%), while new drama “Bloodline” appears to be a slow starter (2.4%).

netflix ratings

Still, any way you look at it, “House of Cards” is clearly a big attraction for Netflix. When all three seasons of the show are taken into account, the program has been the most popular series on all of Netflix in March (6.4%). Its third season was also binge-viewed more than any of the other aforementioned originals, with nearly half of subs having watched at least three episodes in a single day in the first 30 days after release.

The data was drawn from a sample of 2,500 Netflix subscribers watching via computers, tablets or smartphones. There’s one caveat: Luth Research does not yet track Netflix viewing on TVs, whether Internet-connected sets or those linked to streaming-media players or gaming consoles. Because Netflix hasn’t offered much recent insight into its audience composition across devices, it’s not easy to conclude whether TV viewers watch different programming than those watching via other platforms.

But the data provides a rare glimpse into viewing patterns on Netflix, which has frustrated many in its refusal to divulge audience data. That’s proved problematic for producers creating original programming for Netflix and for the studios that pocket billions of dollars in licensing fees from the company.

Netflix declined to comment regarding the Luth findings. While the streaming giant’s execs have credited the sophistication of their extensive user data with helping inform everything from their content-acquisition strategies to audience-recommendation-engine algorithms, the company has also long held firm on its disinclination to publicize any measurement findings. Because Netflix does not carry any advertising, the company professes not to care when a subscriber watches content.

Luth Research captured the viewing behavior with its ZQ Intelligence tool, which the company bills as an “industry-first” solution capable of extracting encrypted data within Netflix’s apps. Luth can drill down to viewing patterns on an episode by episode basis including audience demographics cross-indexed with user behaviors outside of Netflix.

ZQ Intelligence hasn’t been the only third-party research source to shed light on Netflix viewing patterns. Companies tracking broadband usage like Procera Networks have been able to calculate the consumption of specific programs, but Netflix has made adjustments to thwart such efforts.

With its focus on device-based Netflix consumption, Luth Research could prove a good complement for industry research giant Nielsen, which has pledged to offer measurement of subscription VOD services over connected TVs. However, that data isn’t expected to extend to Netflix original programming.

As with Nielsen data, the Luth numbers can provide an approximation of the audience size for a given series. For instance, if 10.7% of the 40.9 million domestic subs Netflix has watched at least one episode of “Daredevil,” that would mean nearly 4.4 million tuned in over the first 11 days. The 2.3% who tuned in first day of release? 940,000 viewers.

That’s far from an exact comparison to Nielsen ratings for particular episodes, but they can provide a ballpark sense of just how many are watching top shows on Netflix on a given night or over a given period.

Luth expects to add Amazon Prime–which also doesn’t disclose its audience data–to ZQ Intelligence later this year, according to Becky Wu, senior executive VP at the company. Luth began offering ZQ Intelligence in March and has data going back to the third quarter of 2013.

Filed Under:

Want to read more articles like this one? SUBSCRIBE TO VARIETY TODAY.
Post A Comment 27

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. Ramekin says:

    Daredevil was excellent and gritty…I agree with other commenters that the only flaw with the series is the horrible costume at the end! It destroys the tone of the show, making it campy and the effect is jarring. It just doesn’t look good.

  2. Dimitri says:

    Daredevil awesome series… But shoot the costume designer of the suit in final episode…

  3. Eddie says:

    The company may not care, but producers and content creators DO care. Many are building extensive licensing and merchandising programs and trying to create long-term brand equity in their productions. Netflix’s “FU” attitude toward this continues to be shocking and will continue as long as they keep making the money they’re making. But as they try to woo more and more content producers, the natural question to ask is: how many people will watch my show? That’s important stuff to know. There’s not even an indication how many of its 44 million subscribers actually watch on any given day. WEEKS can go by in my house without turning on Netflix.

  4. Donna Marie says:

    There is also Hemlock Grove :) Huge fan base …………..

  5. hassan mustafa says:


  6. TobiasM says:

    Who would be interested in this data? It is such a small sample and so flawed.

  7. Julia (multifandom-madnesss) says:

    Reblogged this on Multifandom-Madnesss.

  8. Dr D Noble says:

    I dumped cable/satellite 5 years ago and get my content via a Roku box and subscribe to Netflix, HULU and PlayOn. I will admit my preference is for anything other than US Network television as it is of such low written quality and is cast to fit whichever size 0-2 is handy, not necessarily talented. Netflix Original, HBO & AMC are giving us alternative programming along with smaller niche channels that so far are casting actors and allowing writers to be creative and actually write characters. AMC’s Halt & Catch Fire is a series that is a slow starter and would have probably gotten the axe on Network but it’s been allowed to grow for season 2. Although I lost interest in Daredevil after 3 episodes, I really liked Unbreakable, and House of Cards is brilliant but but will be hard pressed to make 2 more seasons. I don’t think their current rating system is particularly accurate excluding those with Roku, AppleTV, Chromecast and SmartTV systems. But it may give glimpse if these newer platforms are able to give a new series a 6-8 episode test run and then decide to offer to renew if viewers bite. Unfortunately clichéd dreck like OITNB will still get churned out as there is a market for it, but hopefully we will still continue to see mostly creative programming with real talent and leave the networks to keep producing mindless reality shows, laugh-tracked sitcoms and wave after wave of cop/lawyer/hospital soap-dramas. If not, well thankfully these days I can stream BBC1-4 all day if I need to. Content still trumps all there

  9. bbock says:

    I am curious how Netflix will decide how long a show will be in production. I like that they aren’t beholden to people seeing the show the day of release or within 7 days like the nets. That’s a stupid metric in this day and age to determine how much of a show to produce. But what do you measure to tell how many seasons of House of Cards or Daredevil, or Kimmy Schmidt are enough. Or do you let the show producers tell you when they are out of story. (Some producers are good at quitting at the right time. Others… yeesh.)

  10. bbock says:

    Is anyone troubled that they are “extracting encrypted data within Netflix’s apps”. That would seem like hacking if one program on your phone is spying on another app. I think that alone is worthy of a story to find out what the heck that means.

    • Doug says:

      Yeah, they’re stealing information because they found a way to do it… And the media is here to glorify their theft.

  11. Neil ruane says:

    These ratings are only for computers, tablets and phone views. That has to scew the numbers dramatically. Especially for older viewers, or for shows that look better on a flat screen. Like Marco Polo. I’m 28 and watch Netflix almost exclusively on TV. Reporting this as a valid ratings study is a bit ridiculous.

  12. I can’t wait for the next Daredevil season. Netflix is doing a great job by providing its users with high quality original shows. I’m curious to see how the industry will be 5-10 years from now, my dream would be paying double (or even triple) the subscription value to get the latest episodes from other content providers (HBO, CW and others).

  13. Aron says:

    If I had to get just one metric here, I’d prefer it be aggregate hours watched on the vertical.

    While Kimmy Schmidt has a fine future and was a good product, I am very surprised and substantially disbelieving that it could match House of Cards in it’s first season, particularly in the first few days of launch. The vast majority of people were probably completely unaware of it before it launched. Then again, perhaps the vast majority of people were unaware of WHEN House would launch. *shrug*.

    Pretty sure Netflix in negotiations will simply go ‘Oh that’s what they say about the viewership? That’s quite interesting.’ and then carry on.

  14. Steve S. says:

    I’m not sure for whose benefit this (flawed) data is being prepared. It can’t be for the producers of original content since they can easily measure what Netflix is paying them as opposed to commercial TV networks and/or Amazon, Hulu and other OTT providers. When Netflix was the only game in town, it was difficult for content providers to ascertain whether Netflix was paying them a reasonable license fee since they had (and still have) little idea of how many people are watching their content on Netflix. However, with Amazon, Hulu and others (mainly overseas) competing for such content, the issue now changes to one of supply and demand and also one of prestige. Hulu is 1\10th the size of Netflix but outbid them for the rights to CSI. That says nothing about the number of viewers who will watch CSI but does tell me that OTTs believe that their subscription numbers will grow as unique content, whether original or exclusive to a territory, grows. Why else are viewers willing to pay twice as much money for HBO NOW as opposed to Netflix? Because with the exception of limited content licensed to Amazon Prime and to certain overseas OTTs, all of HBO’s content is unique to its viewers and is of exceptional high quality. Why else is Netflix trying to emulate HBO? This isn’t about viewership of one particular program (although having a Game of Thrones or House of Cards is helpful). This is all about how the mix of high quality and popular programming in its entirety will drive up subscriptions.

  15. Tim Buttner says:

    I can confirm that I watch all Netflix content through an AppleTV, and thus would not be included within this data. I watched House of Cards and Daredevil the day they premiered. All viewing was done through the AppleTV so once they’re are able to include the data from TV viewers then I think all these numbers will increase.

  16. Steve S says:

    What a flawed system. 2500 viewers out of 44 million. No television viewers. And no “credit” for watching a program more than 30 days after its release. I have many programs in my Netflix queue, including Bloodline, Daredevil and Marco Polo that I intend to watch but clearly not in the first 30 days. Yet Luth doesn’t include these numbers, instead using the tradiditional TV model of day of broadcast plus 7, extended out to date of release plus 30. Since Netflix does not have advertisers, why does it matter when someone watches a program so long as they watch it at some point. Maybe a better time period would be date of release plus 365 days, although even that period may not capture viewers in new foreign markets. The Luth measurement system ignores the rationale behind Netflix, which is that you can watch any movie or TV program at ANY time during its licensing period. While some people may binge-watch a new show, there is no penalty for those of us with lives to watch a program 45, 90, 365 or even 1000 days after it is released. It’s still part of the overall content package. Under the Luth system, the HBO NOW programming would have the same flaws since presumably it should not count viewers of any “old” programming, like The Wire or The Sopranos. However, it is clear that this programming has considerable value; otherwise why would Amazon have paid millions of dollars to the rights to these programs.

    • DanZee says:

      As for only measuring tablet viewers, this was the only thing it could measure, so it measured it. Probably by reading the computer’s log. Nielsen uses 2500-person viewer blocks to measure network shows as well. That’s the industry standard to represent the US, so that’s in line with accepted metrics. As for not measuring after 30 days, you have to have some cut off point. Advertisers run limited period ad campaigns so the fact that Daredevil may attract 10 million viewers in 2 years isn’t important for an advertiser that only signed on for 30 or 60 days. The bottom line is Netflx will release actual viewership numbers when it feels it has the numbers to start accepting paying advertisers. This survey provides some insight into what’s popular on Netflix, although older, more conservative folks are probably watching on TVs while younger Millennials are probably watching more on their phones and tablets.

  17. nerdrage says:

    Those graphs seem plausible. Daredevil got a lot of publicity online, huge amount of sampling. Marco Polo got slammed, nose dives to zero. Bloodline low but comparatively steady. Kimmy Schmidt appeals to the Parks & Rec crowd by being somewhat broadcast-y but with more of a crazy edge.

  18. Jacques Strappe says:

    Not including television viewing of Netflix whether by smart television or streaming devices would appear to render this data almost irrelevant since I assume (maybe incorrectly) that vast bulk of Netflix programming is viewed on a television. At least that how I binge watched all 13 episodes of Daredevil within 2 days of its release. I know many other people who did similarly and 98% of us watch Netflix on a big screen HD television. A series like Daredevil lends itself to a large television set and home theater system. House of Cards and Unbearable Kimmy Schmidt, not so much.

  19. Like most things, it’ll work…until it doesn’t. Or not.

    • nerdrage says:

      Streaming TV will work, long term. Netflix is doing the right thing: use its revenue to create a wide range of original shows and see what its audience likes; and expand globally to lock in a massive audience with first mover advantage.

More Digital News from Variety