BPS 287: Screenwriting for Netflix’s Algorithm with Stephen Follows

Alex Ferrari 2:18
I like to welcome back to the show returning champion Steven Follows. How are you doing Steven?

Stephen Follows 4:25
I'm doing very well. You say that every time and I'm starting to believe you.

Alex Ferrari 4:28
Well, that's what I'm trying to do. Little by little I'm building I'm building you up, sir. I'm building you up.

Stephen Follows 4:34
I'm delighted to be back and your audience are always awesome as well. Every time I'm on the podcast, people reach out to me and say, Hey, I heard you. I'm part of the tribe. I heard your analysis podcast. Hey, I've got every single one so far has been like really polite, but also with a really interesting question or perspective. And yeah, you've got a great tribe. So I'm always happy to come back.

Alex Ferrari 4:53
Thank you, man. I appreciate that. And last time, we were on the show. We we Did that diehard episode, which was fairly controversial, sir?

Stephen Follows 5:04
Oh, was it? Did you get pushback?

Alex Ferrari 5:08
No. Well, a couple people couple people I got a couple of tribe members like really Alex an entire episode about diehard and I'm like, Yes, it is. But you know funny enough is that when I am when I talk to people now about diehard because now I'm you know i'm i'm an evangelist I go you know, you know Diehard's a Christmas movie and they'll push back up but no, no, no, no, no, I have data. I have proof that Diehard it so I appreciate you doing the hard work on that. And so now at parties, or at least zoom meetings nowadays. I get to, I get to say no, no, I have to get data. Here's the link and I send them to our interview and people just like amazing and I just

Stephen Follows 6:00
that's always the thing you want to hear at a party when you're having a conversation with someone when they go No, no, I have the data.

Alex Ferrari 6:05
Exactly. It's just your life of the party. You are the life of the party without without question, and then I just released a list of the Top 12 screenplays of unconventional Christmas movies and of course Diehards on the top of that list but I had to leave them weapon on there. What else did we have on there? Lethal weapon? Gremlins, Gremlins two

Stephen Follows 6:29
good night.

Alex Ferrari 6:30
Which one?

Stephen Follows 6:31
Is it? What's the one with 497 with

Alex Ferrari 6:34
oh oh no luck is good night. Yes long kiss goodnight. Long kiss goodnight is on there as well. A bunch of a bunch of Shane Black a bunch of Shane a bunch of Shane Black episode whatchu ma call it screenplays because he's he just loves absolutely loves writing. I mean I could argue Iron Man three but I prefer not.

Stephen Follows 6:57
You know Disney do to do list that and on Disney plus under Christmas movies?

Alex Ferrari 7:01
Oh, yeah.

Stephen Follows 7:03
I genuinely don't know if they're doing it to try and stir up controversy or they genuinely believe it.

Alex Ferrari 7:09
And before we start on our current interview, I have to I'm gonna list off the list of Christmas. Christmas. unconventional Christmas movies. Diehard, Lethal Weapon, Kiss Kiss Bang Bang. Gremlins Gremlins to Batman Returns.

Stephen Follows 7:25
Oh yeah,

Alex Ferrari 7:26
Eyes Wide Shut.

Stephen Follows 7:30
You really went there?

Alex Ferrari 7:31
I went there. So yes. Edward Scissorhands.

Stephen Follows 7:34
Mm.

Alex Ferrari 7:35
Long kiss Good night. Bad Santa. Black Christmas. And Krampus. Yes, yes.

Stephen Follows 7:44
That is quite a list.

Alex Ferrari 7:45
That is that that was it took me a minute to put it together. But I had to. I had to I had to. I had to, I had to give them love. So Stephen, man, I mean, I'm always so impressed with everything you do. I love you just an insane, insane human being in the way and the same way you call me insane for what I do. I it's a mutual admiration because I can never do what you did. In 20 lifetimes. I don't think I'll be able to ever do what you do. But I was. I was I don't know where I saw it. Because you kept you kept it to yourself. I have to I have to keep this after I have to give you props for this. You told me that you were working on something big. And I'm like, What is it? He's like, I can't man I gotta keep it. I gotta keep it quiet. I'm like, Alright, fine. I do the same thing. That's fine. And then I think I saw a pop up somewhere, like a few weeks later. And this thing called VOD clickstream showed up on like, what and I clicked on on like, what is this? And I? And I didn't, honestly, I didn't connect that it was yours for a second to like, because I literally had no idea what this was. And then I went to the about the team. And I'm like, oh, some of this is this is Stevenson. I emailed you right away. I'm like, What is this? What's going on? So can you tell you TF but yeah, so I did actually, I think it's like What the f? So can you tell the audience what VOD clickstream is?

Stephen Follows 9:16
Yeah. Definitely, I'll give you the simple pitch. And then I'll make it more complicated and nuanced, because it's got some weird sort of qualities to it. But the simple pitch is that I've got access to a huge data set, which reveals what people have been watching on Netflix, over a three and a half year period. So this is Yeah, I know. It's something I've been chasing for a while and we can talk, you know, in a minute about the history of the whole thing, but it's been something I've been chasing for a while. And it's it feels it's like almost more of a mission for me than than just a stats project. Because I don't like that. We don't know what's going on on this.

Alex Ferrari 9:54
And I think it matters,

Unknown Speaker 9:55
I think and I think it matters, you know for your audience and my audience was very similar, like studios must have a better idea than the average filmmaker. And we don't have the kind of openness and transparency that we have from theatrical or in other areas. And crucially, everyone's experience of s5 is so different. My wife and I share a Netflix account, but we have our own profiles. But there's so different. And we accidentally log on with hers. And we go, omg it's all pink, white, Sandra Bullock and everything. Like, why is everything so sad and exploding. And what that means is that even to people who, like live together, still can't get a sense from their own experience. Whereas when you go to the theater, and you see if it's full, you see lions, you know, you hear about it, and we have the same shared experience. So because of that, because of the essar is so highly personalized. You can't get any clues how things are doing. And they started to release a little bit more data in the last few, like months and stuff. But compared to what we're used to getting on the box, isn't even that's not enough, like how filmmakers supposed to know what to do what people want to watch, like, what is this new realm that is dominating, so much of the value chain is only going to dominate more of it? And obviously, COVID? And I just don't know how else we're supposed to know do this. So this is an answer to that. I guess it's not perfect, but it's, it's pretty unusual. And I think really powerful. And we've only just begun really, this isn't a project where I've launched a finished thing. Here's a report, go read it. It's like, okay, the work begins there.

Alex Ferrari 11:25
So this is essentially the Holy Grail. This is this, this is Eldorado for you, as far as data is, as far as data is concerned. No. Neither is the holy grail Eldorado. Neither is the holy grail or El Dorado for that matter.

Stephen Follows 11:41
I get the Wi Fi is terrible.

Alex Ferrari 11:43
It's horrible. It really is. I've been there. It's not it's not pleasant.

Stephen Follows 11:46
Is that why you left?

Alex Ferrari 11:48
Yeah I just left. I mean, I grabbed a couple things along the way.

Stephen Follows 11:50
But. But yeah, so it's like, it's been really exciting. And the volume of data and the complexity of the data is, it's an order of magnitude, much bigger than I'm used to dealing with. So it's not just me, I've had some help from some amazing data scientists, and most of them, I mean, two of them have PhDs in theoretical physics, you know, they, they deal with things like dark matter, and, and whether the standard model of the universe as we understand it is correct or not. And then I start talking to them about like, what we know about Netflix, and wow, we know more about dark matter. We know more about the origins of the universe that we know about what a film performed on Netflix. And I was like, aha, do you want to like, join the team? Let's figure something out.

Alex Ferrari 12:38
You're Indian. You're Indiana Jones, and you put it together a team?

Stephen Follows 12:44
Almost all physicists like we're the team is more I think it has more theoretical physicists than people who are not a theoretical physicists. And by the way, I'm one of them not. And so it's been kind of bonkers, because not only are they very talented data scientists, but also they're used to dealing with abstract ideas and abstract numbers. And actually, you need to deal with that. I mean, we'll talk about this later on, I'm sure some of the ideas in how you analyze this get quite abstracted quite quickly. Because it's not as simple as like, you know, box office, you say, How much money did it make? Even that's a bit flat, because you don't know if it was lots of kids or a few adults or peak time or off the web. But generally, it's comparative, you account for inflation, you can sort of sort out with this data, it's so much more complicated than that to try and get straightforward, simple answers. And so that's why they were so they just had all the right training for it. And it was just a joy to work with those just incredibly smart, talented people. And sort of see what we can do something interesting for the film, community.

Alex Ferrari 13:44
I have a I have a theory, see, but I feel that the reason why people take you seriously, it's not only because of your work ethic and your talent, but I truly believe it's your it's your it's your accent, because everything you say, I mean, you I mean, it's, it sounds so legit. Like if you if you would like Listen, my friend, I have some land to sell you some swamp land to sell you in Florida, I'd be continue.

Stephen Follows 14:09
But you know, you have a voice for excitement. Let me tell you something exciting. I'll be like.

Alex Ferrari 14:15
So if we joined forces, Steven, we could roll the war

Stephen Follows 14:18
Or we could ruin each other.

Alex Ferrari 14:20
This is very true. Very true, sir. Very, very true. Okay, so how, okay, so you're essentially going inside? The algorithm of Netflix is exactly like that.

Stephen Follows 14:33
Not quite that. So this, it doesn't come from Netflix themselves. There's no data breach. We haven't scraped there. We haven't taken it from them in any sense. What's happened is it utilizes this sort of type of data called clickstream data. And what clickstream data is, is that people have volunteered they've signed up have opted in to install plugins and services and things like that in their browser and other things like that. The the these, let's say plugins are really useful, you know, they've maybe they're a really good translator tool or they just do a certain thing really well. And they're free for users. And the deal is that in return for that, they the users agree that their anonymized history, they clickstream, all the clicks they made essentially, can be sent to a server and put into a big bucket. And that sort of firehose of millions and millions of people are there. anonymized history allows us to see what the journey they made around the web. And so the actual raw clickstream, which I don't have, which is the full, like epic amounts of data, you can imagine millions of people clicking constantly around the world, that is so valuable to so many people in so many different ways, you know, you could get a sense of how popular something is, before the the quarterly reports come out, you could see how people are buying things on Amazon, all that, what I wanted with a tiny slice of it, and I just wanted, I actually wanted all the streamers. But Netflix was, for various reasons, the best one to go for. And I've been chasing these guys for a while. And I was like, because I've known about this for a few years, and I've said, Look, just give me access to the Netflix slice netflix.com because it can be really instructive and very useful for filmmakers. And because of the nature of the clickstream industry, it's a small industry that makes highly expensive content data. And so they were quoting massive figures, like five figures a month, a massive high end, and it was just impossible. And then so I've been talking to them every six months or so I catch up and go, hey, you've you've suddenly sort of decided that your data is worth far less than it was. And they'd like no. And, and then around the summer 2019, there was a sort of big shift in the clickstream data world where there was a there was a sort of a perfect storm of a few different things. Like some of the browser browser, the big browsers changed some of their rules about what their plugins and extensions could do and what data they could share. Generally, people were getting tighter and privacy and so things that they were happy to share in the past, they were less happy to share now. And just all these sort of things came together. And so the clickstream industry transformed and sort of what a lot of their business models, they had imploded and some of the companies are still around doing other things. But basically, it kind of that version of it kind of ended in the summer of 2019. And so towards the end of last year and the beginning of this year. So beginning of 2020, I went back to them and said, Look, you've got this now static data set, and I can't offer you money. I can't there's no, I don't know if the values in that. And I can't do much with it. But I I know that it's fascinating and for filmmakers, and could be very instructive. Please, can you basically give me the the Netflix data, so I worked out a deal with them, which didn't, which was possible to do. And then, so then they gave me about two thirds of a billion data points. So first of all, it was it's just the volume is like, it's just and you know, they get a give me a sample, you can only have a million rows in Excel, before it crashes. And before it didn't work load any more rows. And they gave me a sample of the data. And it was like day one, hour one. And Excel was like it wasn't ours day one. And but an Excel went can't look at anymore. And so that volume is amazing, because it's really granular. And so what I ended up with was, these are anonymize users. So each user has a randomly generated ID, which resets over a certain period. And I know what country they're in. And I know what URL they clicked on exactly the time and the day. And that's pretty much it, there's a bit of metadata. And that's pretty much it. And so, in of itself, a click isn't doesn't mean anything. But when you add them together, you can infer meaning. So you could say, this person clicked on the Netflix link, that is the watch page for a bit of content and the content 22 minutes long, they waited 20 minutes, and then they clicked on the next one, well, that you can reasonably assume that they viewed it right. And you also can see what people have searched for and things like that. So so we have all this data, it has sort of three big limitations. The first is historical. So our data starts in the beginning of 2016 and ends in the summer of 2019. So it's like three and a half years. It's a shame, it's not live, but everyone asks me, can it be live? And the answer is always, sadly not. But if it were live, I wouldn't be able to get access to it. So it's kind of it's this or nothing. And secondly, it's only desktop and laptop users, which netflix they are about 25% of their audience. And so we didn't know if that would have a skew or not like whether people watch fundamentally different content on their desktop and laptop than they do on other TV or tablets, whatever. So the first thing we did was that we went about recreating the stats that Netflix had announced during that period. So when they said birdbox got X number of views or was the number one film within the first Two weeks or whatever, whatever it was, like any data point that they said in a press release, we would go back to our data and try and recreate it. You know, we've performed the same analysis and time and time again, we were getting the same answers they were getting. So because of that analysis, I'm, I'm very, very confident that the big picture we have is a very, very good model of what they have. They've always been cases where it's slightly different or whatever. But fundamentally, considering we started with nothing, I think we were very happy with that.

Alex Ferrari 20:30
So then you don't know how many people actually watch Cobra Kai or Tiger case? No, exactly.

Stephen Follows 20:35
That's what's so interesting is, so what we have is we do have a number for how many people within these panel abusers watched it. But we don't know exactly how that scales up. So what we've had to do is, if we had every single click on netflix.com, then you'd have your viewing figures, right, you'd have a wrong number. But because we we have a fluctuating panel. And and we've had to account for like different factors, like First of all, over the course of these three and a half years, the size of Netflix's subscriber base has changed. It's basically grown and sponsored in different countries, that the number of people using these plugins and services has changed, gone up and down, and maybe they break into a new country or a tool gets taken off. And so that's changed. And, and then also how you compare a piece of content that, let's say, was only available for one year in 2016. How do you compare the performance of that film with another film that was out in 2019? or something? Or 2018? or whatever? How do you compare them because they weren't available at the same time. And so what we have to do is basically, normalize all of these views per day, per country per type. What that basically means is for every single country, we've said, on this particular day, he was the most watched film, and then comparing all the other films to that film. So like film number two, the second most popular film got 70% of the views, the first one did and the third one got 60% or whatever. And then that then gives us comparable things, because you can say within all the films that are available, how did each of them perform. And then that allows us to then create scores overall, over these three and a half year period. So this is where the scientists were really useful because they, you know, compare this content across time and space and different panel sizes the d

Alex Ferrari 22:18
And the dark matter and the ends of the universe. And so I got I got I got it. Okay, so alright, so

Stephen Follows 22:25
last lesson, the last limitation is that we don't have demographic information. We don't have IP addresses. We don't know age, gender, like we know what country they're in. But that's it. So okay, invitation.

Alex Ferrari 22:34
All right. So all right, so let's, um, let's, let's ask some some tough questions and see what you can do to help us because the reason why filmmakers are listening is like, we find this very fascinating how you're getting this data. But how does this help me? So, does Netflix have a longtail? Is that something that that you were able to come up against?

Stephen Follows 22:55
Yeah. That's exactly something we're able to have a look at. So, the longtail idea was was made sort of most famous by an article in wired in 2004. And it was this prediction based on the idea of growing digital platforms like Amazon, selling books and DVDs at the time. But the idea being that, previously, when you have a physical shop, you make most of your money from the top titles, top 10, top 100, whatever the ones you can have in the front of the store, right. And that's where you make your money. The concept of the long tail is that the way that Amazon, the future will make their money is actually through all of the other inventory, the other 100,000 titles, some of which they only sell one or two, every, every every year or whatever, but there's enough of them on total. And so it becomes about the misses, not the hits. And so this was an idea that was put out there and some people supported some people don't. and how it relates to us is that we already know that the box office doesn't really have a long tail, we know that three quarters of all of the money made in the box office goes to the top 50 Films each year. Like it's heavily heavily skewed towards these top movies. So the other however many, you know, seven 800 movies released that year are competing for the final quarter. And that is not great. Because it makes it very hard for us to to to compete because we if you're not big, you're nothing, right. So one of the first things we wanted to test was okay, well, we know that the movie industry is already top heavy already. Massive disproportionately supports the big films. But if we took on this long tail idea, maybe Netflix would be a place where lots of smaller movies would do well, everyone which is something different, but it doesn't matter overall because Netflix are happy. And maybe that's our Savior, you know, maybe it's a fairer space for us all to compete in and loads of tiny movies can equally survive. So that was a big thing for me to look at. And I gotta say it's disappointing, but not surprising news. So basically, net though the viewing patterns on Netflix are slightly more skewed towards big films than the box office. Which means that most people on Netflix are watching a small amount of massive bits of content, which was mostly in the US. It was mostly Disney films, like Disney, they had to deal with net.

Alex Ferrari 22:55
Yeah

Stephen Follows 22:57
Just now finished. But that accounted for a huge proportion of Netflix's views. And it really was a problem for Netflix when Disney ended that deal. I don't know the ins and outs of the the deals. But what I can tell you is that they lost their best performing content in a number of different grounds.

Alex Ferrari 25:43
Yeah in the office in the office as well got lost because it went over to isn't it on HBO, Max or somewhere else that it got?

Stephen Follows 25:50
So, yeah. Well, presumably, peacock? Because it's right. Right. Right. Yeah. And And so yeah, the office is, the office is a great example. Because the office allows us, we've got all the stats on the episodes of The Office. And because the off all episodes of The Office were available across our entire time period, it's actually really easy for us to compare the performance of different episodes, we actually don't have to worry about accounting for time and availability and stuff. And so we've actually used that as a good example to look at how might the nature of S-VOD viewing change the way we think about filmed content. What I mean is, right, so in TV, we're used to having seasons, you know, because of the way that they're funded and broadcast and just the way that it's evolved, we are used to having a piece of content. So having having a series that is got a beginning, middle and end, maybe it's got an arc across the season, and the beginning of the end of the season is significant. And then we wait and whatever. But that doesn't, there isn't time is less of a factor with things like Netflix, like, it's not irrelevant, but it's far less of it doesn't matter whether it's summer or winter, you just watch them, Benjamin. So when we looked at all the viewing figures of all the episodes of the office, we noticed a couple of really fascinating things. Which is, first of all, the most popular season was season four, not seasons One, two, which is kind of interesting. And I think it's about that was where it really hits stride and where people start watching it or maybe where they rewatch it as well. But we couldn't see, if you have a look at the chart of viewing figures across, you know, you're on the left hand side, you've got season one, episode one on the right hand one, you've got the last one in season eight or nine, whatever the last one was, and you have all the viewing figures as a line, sort of a line going up and down across across those two points. You can't see where the seasons begin and end. You know, they may make them as seasons, but people don't watch them as seasons. So it's much more like a podcast than a radio series. Right? So you might think a radio series has got a season and certainly in the UK, the BBC have, like, Okay, this will be six or maybe 10 episodes of a radio series, then there'll be a hiatus and then they'll come back. Whereas podcasts you just think are always going to continue right, you just it's a it's a long stream of content. It's like a soap opera rather than a miniseries. And that's how people watch the content. And so I don't know how long it will be. But it seems inevitable. based on the data we have, that when people start making more content for Netflix, they're going to move more and more to this sort of soap world where they're always making them like a churn. Like even the most expensive ones it from the way people watch them, it makes sense to just drop a new episode every two weeks, forever, than it does to quickly go and make 10 of them. I mean, the economics, the production costs might be different, you might want to throw more in one location, but the production costs are not a big concern, if you get a Netflix hit. So maybe we're gonna start seeing seasons of indefinite length, and maybe break breaking down like how long episodes are like, I've just been watching them. There's some brilliant comedians called Auntie Donna, Australian comedians who've got a Netflix series that's just come out, which is incredible. But they some of their episodes are like 17 minutes long. And it's great because every second is great, but they don't have to stretch it to 20 to 30 minutes, whatever it would have to be for TV. So

Alex Ferrari 29:19
you know so so with with all this information are you seeing because I've been reading a lot that Netflix, you know, is infamous for just canceling shows. Some of the people's favorite shows just get cancelled and they're like, you know what, screw you We don't care, because we're gonna get we're gonna put out 20 new shows this month. And, and they generally don't go past three, four seasons. You know, I mean, I think Frank Grace and Frankie is one of the longest running shows on on Netflix. Orange is the new black got cancelled, ended eventually and, and they don't seem to care about letting things go on and on and on and on. Because they just rather just start playing thing from scratch. And I think it's because mostly because of the talent costs and

Stephen Follows 30:04
and that's gonna say I don't, I have no inside track on to to Netflix and I the data doesn't give me all of what I'm saying here. So some of this is filling in the gaps or my opinion. Sure. But I, I would say that the cancellation for most of these things comes down to exactly what you'd expect, which is number one cost. And number two talent, which is related to cost because they are they asking for more money each season and crucially do they want to still want to do. And obviously Netflix are going to cancel shows that they don't think of performing. But they, they could do with more content, almost always. And if you think about it, what they actually really want it, they obviously want content that everybody watches, that's amazing, that'd be great. But one of the other things that's actually really important for their business model is content that's important to some people like really important. So let's say that hypothetically, you and I both have a Netflix account. And let's say that you watch loads of different TV shows every every month, you watch 30 different shows, if I watched just two shows every month, but both of us pay the same fee. Those two shows that I watch are more valuable to Netflix, because if they cancelled those two, or maybe even just one of them, maybe I would leave. But if they canceled 10 of the 30 to you watch now you probably watch the other 20 and maybe some other ones. So the model that they're having to use here is not just not just the number of people watching you, but it's how valuable they are to that particular sort of audience. Yeah, yeah, exactly. So that's what I mean, it's a whole different business model where it's on television, you're saying how many people are watching it? And what demographic are they in? Like that's, that's what's driving content on television. And what's driving content on Netflix is different.

Alex Ferrari 31:54
completely different. In so what I'm have to ask the question, what does the lowly independent filmmaker, how does their stats work? I mean, obviously, you said that they're mostly skewed towards the big movies or big stars. I mean, I saw an interview or an article discussing why Adam Sandler, is one of the biggest stars on Netflix. And that's in like people like Why does he keep making these movies? Why? Why does Netflix keep giving him money? Why I mean, like, I personally a fan of Adam, so I love his stuff. Not everything but most of his stuff. And the thing that they said was in the article was was really interesting. And it made a whole lot of sense was the reason why Adam Sandler is given this these kind of movies and these kind of deals is because when you're scanning through an S VOD platform, there's so much content that when you see something familiar, when you see an Adam Sandler movie, you know what you're going to get? Like there's there's no mystery about like, I think you just run this Halloween, they just released the holly something or other how Halloween Holly Halloween or something like that, which was a huge, huge hit. They're going to do a sequel to it, because so many people watched it. And it's the same. You know, it's the same stuff Adam Sandler has been doing since Billy Madison, and Happy Gilmore. But because of people's comfortability with, they know what they're going to get people watch and watch and watch.

Stephen Follows 33:36
We see a lot of that. I mean, there's a lot of brands that do very well in a brand, Adam Sandler being a brand here that does very well on Netflix, and I think that some of it is down to is absolutely what you're saying. I think with him, there's also not that much competition, you know, there, there aren't many substitutes. What's the Adam Sandler substitute? Well, Kevin James is in most of Adam Sanders films.

Alex Ferrari 33:57
So it says David Spade.

Stephen Follows 34:00
There isn't a lot of competition. And I think that there's something you touched on there, which is incredibly important, which is that the way that people invest in the time they're spending watching stuff on our squad is a lot more about relaxed time and not making a decision and some of that, and I think that speaks to why Adam Sandler is popular, but also why that's the same films people are watching. And they're watching the same TV shows again, and again, rather than watching new ones. And I think that actually doesn't help independent filmmakers, because we're making stuff that doesn't have famous people doesn't have existing brands. And more often than not, is trying to challenge something I'm not suggesting we're all trying to pass on a message or communicating but it's not the same. It's not the kind of Sacher and stuff you might get from Transformers or a Disney movie or Adam Sandler where you're like, Okay, I'm just gonna go with the flow. Most independent filmmakers are making something a bit spiky than that. And I think that doesn't suit most of the way people watch Netflix. And so I think when you your question about what is a story mean that Netflix is what we've learned so far from Netflix, it applies to the other platforms and continues broadly. In the future, I'd say that it's not great for selling independent content. Because first of all, what these platforms like Netflix one is the MCU, they want the Marvel Cinematic Universe, because they people will watch it, they also can do one deal and get loads of it. And that'll get most of their views. They don't do individual deals, the audience aren't watching that content, the other company or the independent content. So there's not much of a drive for that. And so I think that's not great. However, what I would say is, in the same way, right now, I wouldn't invest in any of the companies that own theaters, I would still invest in the concept of theater going, because I think people go on dates, they see their mates, it's a cheap, for it's the most the cheapest, most social form of going out with the lowest effort. And I think that what's happening right now, I mean, I just happened, I was really a few days ago that Netflix have spent 2 billion pounds in the UK alone on content this year on production. So independent filmmakers might have many more routes to being employed than they would have had previously as if their previous routes were big movies or TV shows, they've now have a whole new realm they could compete in I don't know how well I don't know how fair that system is, I don't know. But certainly, there's got to be more and more entry level people, because there's just more concurrent more content being made.

Alex Ferrari 36:26
Oh, yeah, here in Mexico here in New Mexico, they're, they're expanding, Netflix just got approval from the state to expand their studios, they they're building out a massive studio complex. In New Mexico to hire

Stephen Follows 36:39
people, it's going to create, of course, below the line. And so as filmmakers like that, and also there are some interesting things Netflix is, it's very easy to think of them as a studio. And they're actually fundamentally not, they are a technology company. And they bring a lot of different values into what they're doing. I mean, I would argue that they are one of the the most forefront of HR, in the film industry, human resources, like they actually are able, normally when you if you work on, if you're below the line crew member, and you work on six different independent productions in a year, you can expect to have six different relationships, and no concurrent, no sort of handover really beyond a mild relationship in the sense that if something bad happens, you just try your luck again on the next one. Whereas here, because there's a continuity of people being the higher end, you know, Netflix care whether there's a complaint about somebody, and this is great for things like sexual harassment or unfair treatment or discrimination. I'm not saying they're going to solve everything, but there is a continuity there. I mean, some of the studios have tried that Warner have been doing that for a bit and Disney to some degree, but no one to the extent net and Netflix are doing this. So they are doing some things very differently. And as individuals, it might be a good thing, as people buying and selling content the way we used to doing it. I just can't see it being better than it was. Because it's also an oligopoly. You know, I don't I'm not suggesting they're acting in any way, duplicitous. But when you have five or six, possible, maybe even, let's say, three or four, as you know, hold them. So Apple aren't buying existing content. So let's say that it's Amazon and Netflix, let's say that they're the only two that could you could sell your content to in any big way. That's not going to engender, you know, fair prices. And you're doing a single deal in perpetuity for the world. Maybe, maybe. And so that's, that's a simple ad sale. If you get a good price, then that's, that's amazing. But will you get a good price? And I also think there's some worrying practices. I don't think any of them are illegal, but I don't like that as well, I can say. So for example, I was talking to a lawyer recently, who, who's sort of looked over a lot of deals to one of the big streamers. I won't say which one. And this lawyer said, Look, one of the problems is that part of the terms and conditions of the deal between the distributor and the platform, is that the distributor is not allowed to tell the filmmakers how their film is performing. There has to be some sense of aggregation of the numbers and you know, yeah, so it's horrible. That not only is that horrible in a human sense, but it's also terrible for that deal. And it also it stifles long term growth, like how can you have a sustainable career unless you get feedback, and your feedback can't be we did a deal, but I can't tell you anymore. And that brings us full circle back to the VOD clickstream because that's what we're trying to get a little sliver of light in a dark room. Like it's not like we can illuminate everything. But we're trying to understand these things that filmmakers need this feedback loop that needs to happen with the audience.

Alex Ferrari 39:42
Now do you know our American audiences, streaming a lot of international shows because I personally, I've watched a bunch of international shows recently because they've been popping up on my on my, my, my feed, so I'm like, Oh, that looks interesting. Oh, that looks interesting. And sometimes they'll they'll pitch me Something I'm like, Yeah, no, no, thank you. I need, you know. And it just depends like, you know, I'll watch subtitled movies, but not normally, because I want to relax when I'm watching movies, unless I'm watching it for cinematic purposes. But I'm just chilling, I don't want to read. I just want to say you

Stephen Follows 40:20
You don't want to be challenged. Like there are some movies that I would, in a heartbeat recommend to other people that I've only ever seen once and might see again once in the future, but only to introduce it to someone else or because you know, some bizarre circumstances yet there are bad movies that I will acknowledge that a bad thing I've seen the Meg twice,

Alex Ferrari 40:39
right? It's no.

Stephen Follows 40:42
That's the wrong ratio.

Alex Ferrari 40:43
It is it is the right ratio. And you know, what the Meg I, you know, I, I watched the mag as well. And it's just a, it's a popcorn movie. It's, you know, it's there. It's there. That's the reason why just the reason why my wife and I just sat down and watched all four Lethal weapons in a row. Because we watched the first one because I hadn't seen the first one forever. And I'm like, Oh, my God, that's so brilliant. Well, we have to watch two, we have to watch three, well, let's just let's just go, let's make it the fall four. And in four days, we watched for all four of them. And we're like, what's next? Let's watch Tango and cash. You know, like, like, I haven't seen that in 20 years. So it's like I'm going but it actually says exactly what you've been saying is, I'm doing that because I know that I'm comfortable. Those are comfortable viewing habits. And I'm like, oh, let me go revisit that again. Because I haven't seen that in forever. I remember here and there, but I haven't seen it. So

Stephen Follows 41:38
that is a response to being overwhelmed with content. Because you know that there are so many movies been out there. And if you had to create a list, how would you find brilliant movies you hadn't seen? It would take you seconds? IMDb score, meta score, sure one Best Screenplay. And there's loads of movies that you'd be like, wow, that's sure heard. That's amazing. I've not seen it. But that all takes a lot more effort and commitment than most people are willing to give. And this is something that I think filmmakers really independent, because really need to either embrace or realize you're not going to embrace it and then find other routes. Both are valid, like I'm not actually saying make popcorn movies, I'm just saying you can't make challenging movies, and expect them to then survive in a mainstream environment world world, because that's not how people watch that content. Right? It's just fundamentally and I think that the growth of content, or sorry, the evolution of content, and the growth of platforms, are massively interlinked. And the best example I can give you is outside of the film world, but it's kind of makes a lot of sense, which is the rise of Kindle, the Kindle e-reader was a massive part of the success of 50 Shades of Grey, and 50 Shades of Grey was a massive part of success of the Kindle, because you could be on the train reading something, reading basically soft porn, and no one would know. And both of those two things sort of coincide with the same sort of time. And it's not that everyone reading stuff on the Kindle was porn, but it did mean that you could read private things. And the same with the rise of sort of portable devices, and podcasting, you know, these things are interlinked, right. And so what we're seeing, what we're starting to understand with this was, is that people don't watch content in a curated way, the way that they might when they go to a certain type of theater, or they go to like, you know, they could draft house or they or they watch when they buy a blu ray or the Criterion Collection, or, you know, the considered in a centerfire way. That's not what people largely doing on these big platforms. They're sitting down watching stuff that is comfortable. But it's easy to understand that one challenge that they can pause when there's someone at the door, or they want a cup of coffee or something that is out and sat through and through.

Alex Ferrari 43:45
No, there's no, there's no question. I mean, and the other thing is, like you're saying movies that challenge you, you should also you can make movies that challenge you, but you got to do it on a budget. If you you know, if you if you have any hopes of recouping that money, like you can't make a two or $3 million, you know, indie film that five people want to watch. She's just irresponsible.

Stephen Follows 44:05
So there's more, you know, the more it just makes sense, the more you spend, the more you got to recoup. But I and I totally agree with that. I think the other thing that I know you've been screaming at people for since way before Netflix, but it's even more the case now, which is you have to know your distribution route before you make it. I'm not saying do the deal, because I appreciate that it's very hard to walk into a room and say I haven't met you don't know me, you don't know my movie and I haven't made it but can I do appreciate those kind of pre sales can happen. But you can't hope that you're just going to throw it in with the straight with the with the sort of stream of content and including the streamers and it will get swept up and it will rise to the surface. It just from the data I've had I've seen here. That just doesn't happen. That's just not the case. You can't write a book and expect it to be on the front page of Amazon or in the in the front of the book shops, right we know. And yet filmmakers still think if it's good enough, it'll break through and I do worry somewhat that the way the S VOD platforms are working now through no fault of theirs, they're just chasing them, you know, subscribers. And the bottom line is that it isn't. It doesn't reward films The way that the previous system would, to some degree, you know, maybe we'll see fewer breakouts, maybe it will be that the where you really break out is on a much smaller platform like for example, film festivals, whether they're physical or online. Or maybe it's niche sites like shudder or something like that. I don't know, I don't have a site. But that is nowhere near the volume that Netflix does or Netflix competiting appears. So there is, we all know that there's huge amounts up in the air partly it was happening anyway. And then COVID accelerated things. Now, it hasn't landed yet. But we don't yet know what this model will be for independent filmmakers. I am absolutely confident independent film will exist? Because it's not it's never been supply and demand. It's always been supply. Right? Where can I find the demand. And that's been part of the joy of like, movies, they're not been made. Some of the best movies have been made, because they want to be made rather than because they I know, I know, everybody have a deal in place. But things are gonna get tougher until we figure out what they are. But if I have, it's never been easy. And you look at some, you know, the crash of cinema tickets in the 1950s, you look at the crash of DVD, and you look at the uncertainty of aswad. And all this stuff, they will find a way, but just don't know what that is yet. And it's not the one it's not the easy one. It's in front of us. You know. And so you were asking earlier on about TV, because we have we have data for movies for tv and for comedy specials. And for TV. It's it's a, it's the same pattern in a different format. So what we saw in movies is that the most watched movies by a huge degree are the big famous ones. And when it comes to television, what we tend to find is that it is the big shows, but also it's the more familiar shows. So if you go on if you're in the US and you go on Netflix, there's content from many different countries you could choose to watch. But what do people watch? They watch it from their own country, you know, and And

Alex Ferrari 47:14
Generally speaking

Stephen Follows 47:16
yeah, yeah, exactly. And, and certainly, if you look at the, like, the top shows, like the top 50 shows are almost all produced in the US. And you have to even the top 500 most watched shows, not episodes shows, on Netflix over this period, almost three quarters of them were us produced. The UK does very well. But that's largely down to the Great British baking show and things like things. I'm David Adams,

Alex Ferrari 47:42
I would have to say I've seen both. But

Stephen Follows 47:46
what I think is so interesting is that first of all, I was as proud as I create these, these Brits are surviving and competing against that as I owe notes to shows. Like, what this means it's though there's an interesting, like thought process that goes on here, if you're being rational, you would say it would make more sense. Rather than trying to make three films or trying to make one film that competes in three areas. It's quite good, it's quite scary, it's got some effects, it makes more sense based on this data alone to compress all of your resources and that includes time and money and passion and whatever into one thing you know do one thing incredibly well and because of the power law and sub nature that if you go from being the second most the third most popular to the second most popular will mean so much more for you than going from fourth to fifth from from fifth to fourth and that it would argue that it's better to make something that's extremely one thing and you see this I mean we're in we're recording this before Christmas and and in look at how many Hallmark Christmas shows there are

Alex Ferrari 48:53
OMG

Stephen Follows 48:54
good well made or enjoyable I'm gonna show one or two but what they are is feel good Christmas like they are absolutely that

Alex Ferrari 49:02
are there there's a formula and again the comfortability factor for a specific demographic of people. That's why those films generally have a Mario Lopez or, or a Deem Cane or a face that people feel comfortable with because they remember them from you know, they're just comfortable they've watched their films or watched a TV shows over the years. And the watch it because it's like, Oh, you know what, I want to feel good. I want to feel good, Christmassy. And, oh, great. This is a new movie. And there you go. And all of us like I knew that no, that Mario Lopez Christmas movie exploded on Hallmark apparently. Because people love Mario Lopez because you know, it's later but

Stephen Follows 49:46
and he's his this thing is this thought that we I haven't heard expressed very much. I'm sure it's not a brand new thought. But in the last 10 years of being an independent filmmaker and working with independent filmmakers and chatting to them. I've heard people talk about oh my god, we have to hire People actors who've got a bigger social media following or whatever people, people have often complained to, they have to weigh up talent and appropriateness for the role and the wonder, and the fame on the other. What I haven't heard many people talk about, but I would argue is perhaps the battle that we're going for in the next five years, is in familiarity, not fame, but How comfortable are people with that person? So it's like, you know, one of the reasons that George Bush got in over Al Gore was that people were happy to have a drink with George Bush. It wasn't about politics to some for some people. And I noticed because I know, some people that voted for him who actually, I think their politics was slightly strange. And they were like, Yeah, I just don't like the word. I'd have a drink with bush. And so when you think about actors, it's not so much their fame, although obviously that's not a bad thing. And it's not so much their talent, although

Alex Ferrari 50:50
Do you feel comfortable?

Stephen Follows 50:52
Do you think your audience would go Yeah, okay. Without thinking about it. You know, and that's why you look at actors like I mean, almost every one of Adam Sanders movies as a comedy to the point to which people have been watching on uncut gems and been appalled. Whereas there are other actors who you just don't know what their movie is going to be because they play such a wide spectrum of carrier

Alex Ferrari 51:15
Tom. Hey, Tom Hanks is Tom Hanks like you he'll play everything and he's definitely not Adams. Yeah, it's his brand but that's fine and but you also feel comfortable within Tom Hanks or with Meryl Streep. Meryl Streep plays everything. She's going to be in a musical this month on on Netflix. But yet, you know, she she also was on HBO, Max doing another film with Steven Soderbergh. And you know, she she does everything, but that's her brand and you feel comfortable. And

Stephen Follows 51:40
I started this a while ago, I started on my blog, like how broad the act the the roles that actors have played across genres. And I found Adam Sandler was the most siloed. He did, most of his films have been in one genre. And the the actor, I only looked at sort of a couple of 100 really big actors. But the actor I saw that had the broadest as in like, had the least siloed in one genre was Ron Perlman. Yeah. Any you take, there's no one genre that accounts for more than a third of the roles he's done. So he's done some comedy, but he hasn't done mostly comedy. He's done some kids stuff. He's done some horror, easterns fantasy. And so Ron Perlman is an example here, who I think is a terrific actor. Yeah, like a good bloke. He is a perhaps you metric slightly less attractive to hire because he doesn't have that whatever we're going to call it comfortability effect. Whereas someone like Adam Sandler, who I would I'd rather see Ron Perlman take on Ron Perlman take on some certain drama roles at random center. But Adam Sandler would be more comfortable watch for more people. So I don't know what we'll do with answers.

Alex Ferrari 52:43
It's, it's very interesting the way this whole, this whole thing is, but I'm really I'm really happy that you've doing what you've done with the with VOD clickstream. And I'm, I'm just impressed. Like I always am with everything you do, man, you're insane for what you do. And I know that you're going to be digging through that data and continuing to grow, and you just started to go through that. And that's a good it's, it's not exactly what's going on. But man, it's, it's more than we had before. And it's definitely a direction to aim at, it might not be pinpoint. But man, it's better than, you know, like, Hey, I'm gonna go throw a football into a stadium, I have no idea where it's gonna go. Now at least you'll get it on the field. And maybe you can even get it within a few yards. You know, maybe that's the goal.

Stephen Follows 53:33
And also, you know, filmmakers should use all of these data points, and all of these things they hear and then they know themselves and they talk, they hear on your podcast, interviews. All of these are things you need to weigh up yourself and weigh them against everything else. No one person or one system can tell you what to do. And I'm just glad that we have at least one set of signals about SVOD, that doesn't come from the PR department.

Alex Ferrari 53:56
Yeah. Well, and I appreciate you fighting the good fight, sir. And getting this information out to the filmmakers. Where can where can people go and get this info.

Stephen Follows 54:04
So it's VOD clickstream.com. It's entirely free you, if you want to read more than the beginning of the articles, you can just sign up, but it's free. But that's the reason we put that barrier in is that we have got forums that anyone can join. And we wanted to make sure that there was a there was some effort you had to put in and that effort is signing up and accepting your email address. And what that means is that we have forums where people can post suggestions because we're still working out what to do with all of this data. You know, some of it, we have plenty of ideas, and we're churning away at them. But then there's some deeper things that we don't know what to look at. Yeah. And might the best suggestions for the research I've done over the years have come from audiences, I guess, if I was to think of the sort of most exciting things I've studied there, almost all of them come from audience suggestions. So that's what we're looking to have is like, what have you always wanted to know by S-VOD? I can't give you an immediate answer, and I might not be able to answer it at all, but probably I'm the best shot Most people have. And I'd be delighted to follow those threads and suggestions that we've had from people.

Alex Ferrari 55:05
What man, I appreciate everything you do. Steven, thank you so much. We have to we have to come back on the show and talk about our 12 unconventional Christmas movies and do another episode next year. But I appreciate everything you do, brother thanks again for coming on and, and sharing very valuable knowledge with with the tribe. So thanks again.

Stephen Follows 55:27
Thanks for inviting me. It's always a pleasure to be here.