I’ve spent a lot of time here at Birth.Movies.Death. talking about Netflix. The main reason is because Netflix is a true disruptor and the ripples from their success have implications that will affect every aspect of film. I’ve talked a lot about its effect on distribution and how while more content than ever is being consumed, the revenues made from that content are declining. I’ve talked about how it’s taken advantage of consolidation in the studio space to acquire high profile and high concept movies that allow Netflix to position itself as a destination for studio content long after most studios will have folded or merged with others.
One aspect I haven’t touched on is how Netflix has fundamentally changed which movies we choose to watch and how we make those choices. Netflix has always been critical of the theatrical model. It’s partly because they want consumer behavior to be about watching movies at home, but it’s also because it costs a ton of money to release a movie in theatres. Even with marketing and distribution costs being reduced in our digital age, it still requires hundreds of thousands of dollars to give a movie a meaningful release and millions of dollars once you’re releasing films outside the arthouse circuit and then tens of millions if not hundreds of millions to release tentpole movies all over the world. These marketing dollars, even with targeted digital plans, are incredibly inefficient. The hope is to cover as many people as possible and a tiny fraction of them will hear you and engage with your marketing. I remember working in an industry that mailed out coupons and expected only 2% of them to get used. Digital marketing has allowed us to better target specific demographics, but even with these advances, the same concepts apply – cast a wide net but catch only a few fish. It can be better than 2% these days, but it’s not a lot more than that. If you can get to 20%, you’d belong in the marketing hall of fame.
The main problem is the lack of data one has about the consumers they are marketing to and this is where Netflix has a HUGE advantage. They have collected so much information about your viewing habits, they are much better than the theatrical model at giving consumers what they want—and the best part? They don’t have to spend any money to get this data. They just track who you are, where you live, and what you watch on their service which you pay them to use. I don’t have to get into the idea that acquiring data is driving much of the desires of the large corporations we engage with. We hear almost every day about Facebook and how they use and sell the enormous amount of data we give them for free. Basically, if you’re sharing your behaviors with your phone or your computer, chances are very good that you are giving a lot of companies information on what to sell to you.
Netflix isn’t as nefarious as Facebook. Netflix wants to collect your viewing behavior because they want to keep you watching and they’ve gotten great at it. Per past conversations I’ve had directly with Netflix, 80% or more of the transactions that happen on Netflix are through their recommendation engine. Let’s go back a second. So theatrical marketing hopes for activations from 1% to 20%, Netflix shows you a list of movies and 80% of what you watch on Netflix is from those choices. Now you understand why Netflix doesn’t believe in the theatrical model.
This is why Netflix was willing to give out one million dollars to anyone who could improve their recommendation engine 10%. This engine could be the most valuable thing to happen to entertainment since the TV was invented. When I worked in video stores, the most popular question people had when they came in was “What’s good?”. Netflix is answering that question every time you interact with it. This is why they kept their search function hidden for so long; they didn’t want people to look for content. They want to tell you what to watch. Netflix had deals in the past for new studio content that did significant box office, but most of their content when they were growing was not box office content. So the more they could get you to depend on their recommendations, the less content they needed that people would be aware of and therefore coming to their platform to search for.
It's also important to discuss the role that user ratings play in Netflix’s recommendation engine. Essentially, the movie must be rated well for it to exist in their recommendation ecosystem. First, they used a star system, but now it’s just thumbs up and thumbs down because users would react in similar ways to movies with 4.5 stars to movies with 3 stars. They are letting their customers pick which movies to pass on to others. While it’s true that Netflix gives their originals heightened placement, the film still must be liked by their customers for it to live beyond the initial marketing. Many don't realize this as they can see billboards around town for their movie or go with Netflix to a major festival for a premiere or see their movie on Netflix’s splash page, but the film must resonate with viewers or the movie won’t continue to get recommended.
This is how Netflix has fundamentally changed how we choose to watch what we watch. Instead of choosing specific content, we’re now content for Netflix to tell us what to watch. In a way, they have taken the definition of content from something specific to just general entertainment. Netflix not only uses their data to tell you what to watch, but they also know what gets the most views from their worldwide audience. A lot of people laughed when they signed Adam Sandler to a big production deal because the word on him was that his star had faded due to a decline in theatrical box office. However, Netflix knew he was still incredibly popular to their worldwide audience because his older films were still getting watched a lot on their service. They made a bunch of movies with him and according to Netflix, they did well. These movies weren’t considered good by the current criteria film criticism uses to evaluate films, but since Netflix heavily recommended them, they were well liked by Netflix’s worldwide audience and that’s what matters most to Netflix
This is another vital shift in how audiences first interact with movies. Film critics and their curation has always been at the forefront of a film’s marketing. The belief being that movies with good reviews will do well in theatrical release, so distribution companies would get critics to see the movies early to help drive traffic to their films. Netflix doesn’t need theatrical marketing because of their recommendation engine, so they don’t need critic support—and while this is a disadvantage for theatrical releases, it’s an advantage for Netflix.
I’ve been tracking movies and their performance for over a decade now via spreadsheets. I’ve collected some data on CinemaScore (a company that polls people at theaters all over the country right after they see a film and finds out why they watched the film and what they thought of it) and Rotten Tomatoes and their critics score that’s not surprising, but the results are still illuminating. Here’s some data on 1,125 movies since mid-2008 that have secured both a CinemaScore and a critics score on Rotten Tomatoes:
Again, it’s not unexpected, but look at these numbers. 90% of theatrically released films score a B- or above. My thinking here being that even if Rotten Tomatoes grades a C- as “Fresh”, an audience score of “A” or “B” is the same as “great” or “good”. This means that audiences like most movies and critics, understandably, are much more selective. Whatever you may think of CinemaScore and other exit polls, Netflix’s recommendation engine operates in a similar fashion as it’s designed to rate the film right after you see it on their platform. It’s likely that the Netflix worldwide audience will respond the same way as theatrical audiences do to similar types of content, so I believe it’s accurate to assume that if the masses like 90% of studio content released theatrically, they will also like studio content – or the equivalent – on Netflix.
Then what types of movies don’t do well on Netflix? If you look at the bottom of the CinemaScore scale (D’s and F’s), you see a lot of two types of movies: Horror/Sci-Fi films (Silent House, Devil Inside, Last Exorcism, Skyline, Blair Witch, Apollo 18, Chernobyl Diaries, Devil’s Due, Splice, It Comes at Night) and arty studio films packaged as mainstream star-driven, mid-budget films (The American, mother!, Killing Them Softly, Suburbicon, The Counselor). Genre films will have some value in moderation because even if a mainstream audience doesn’t like genre, there’s a built-in audience and they are much cheaper to make because you don’t need to pay for big casts.
It’s important to understand the low-end, because once you understand how these movies exist inside Netflix’s recommendation engine, you can understand which types of movies are likely to do well for Netflix and which ones will not. If you can’t easily find a movie on Netflix, it’s not because it’s being “buried,” it’s because it’s not being rated highly enough by their audience. The flip side is that the high end for Netflix isn’t full of bad movies or movies you can watch while you do other things, it’s that they are movies that are a lot like studio films – filled with recognizable stars and featuring high concept plots that only can happen in movies. That’s why they are quick to pick up movies that are in studio turnaround. These movies may not work for theatrical release because the critics won’t like them, but they will work for Netflix because they are studio-level films.
This is the change in basic assumptions that honestly, most of us haven’t caught up with yet. We’re so used to the criteria critics have built that we haven’t noticed that what critics define as “good” or “bad” no longer applies on a service where critics aren’t needed to drive traffic to films. Critics are still reviewing Netflix movies, but they do so primarily to drive traffic to their own sites, not to drive traffic to the film. Netflix is doing more and more traditional PR like sending out press releases and trailers and doing some mainstream advertising at movie theaters and on TV, and there are expensive outdoor campaigns in LA for their content. These holdovers from traditional theatrical marketing campaigns are still somewhat a mystery to me, but it is likely that Netflix is promoting themselves just as much, if not more, than the movies themselves. Netflix also premieres movies at high profile film festivals, but these are done to placate the talent involved so they will work with Netflix again. A festival premiere can equate to an entire theatrical release if the talent sees the movie in a theater with an audience ready to embrace it.
To summarize, Netflix is really the ultimate disruptor and their model has changed not just how movies are released and how we consume them, but how we know they exist. Any conversation about Netflix and their objectives, especially conversations that revolve around the quality of their library, must think about Netflix on their terms. They are not marketing their movies the way theatrical movies are marketed. They want to tell you what to watch instead of you looking for specific content, or having guest humans curate their favorites, or you browsing through possibilities. And their library is not filled with “bad” movies. They may be “bad” for the old models, but those models are antiquated with Netflix, so these terms are also antiquated. As with all change, there is both a heavy dose of positive change and a healthy dose of negative change, but this Pandora’s box is open and it’s never getting closed. These paradigm shifts are here to stay – at least until the next disruption comes along.