May 16, 2018
Sergey Galyonkin was just trying to fix a problem at work when we accidentally revolutionized the way we understand video game sales. We uncover the fascinating story behind Steam Spy, the people who use it, and the insights it gives us.
Hosted by @dannyodwyer
Funded by 4,197 Patrons.
Sergey: My godmother, she used to work for a huge computer center, you know like a secret type of building, you know, so you can't get in unless you get a y'know pass or something. But because I was a kid, they would let me in with her. I was, I don't remember like, seven or eight. And she let me, she would take me to you know to her job and she would let me play with computers. And they didn't have many games, it was you know they were mostly to do with statistics and stuff like that, but they had Tetris and they had Kingdom Euphoria. And back then I totally hated Tetris. I didn't play it much, but I mostly played Kingdom Euphoria, which was a text based strategy game.
Danny: Text based strategies appealed to Sergey. From a young age he enjoyed solving problems. He'd spend hours making small games on a programmable calculator. You see, the Soviet Union in the 70s and 80s had restricted access to most type of electronics. So the computers available to consumers was limited to Soviet manufactured machines, or expensive black market imports from the West.
Sergey: I didn't play many video games until like maybe age of nine or ten. Because we didn't have any. We had only like you know those old Soviet arcades. But then the Z Spectrum came to our country and it was a revelation. It actually was the first mass computer in Soviet Union. Not just in Ukraine, in whole Soviet Union. And I bought the first one, not I bought it, my father bought it for me. And I actually assembled the second one myself. Because you could buy you know the scheme, you could buy everything, you know separately. And just solder it. And it was fairly easy back then and I saved a bunch of money, do it.
Danny: Using his ZX Spectrum, Sergey would create games for himself. He didn't enjoy programming in BASIC, he found the code too restrictive. So instead he opted to program using Assembly Language. His love of programming continued through his teens and when it was time to go to university, he chose to study Computer Integrated Systems, with a focus on Neural Networks. Ukraine has always been ahead of the curve when it came to developing algorithms. For instance, the first Neural Networks used to detect fake dollar bills were prototyped in Ukraine. Sergey continued his education and worked a bunch of jobs. He did page layouts at a local newspaper, he spent some time at a game studio, focusing on edutainment. Eventually he'd find himself moving to Kiev and taking up a job at a games distributor responsible for selling games for some of the biggest publishers in the world. What were some of the popular games in the Ukraine around that time? Any stand out in particular?
Sergey: Well, I mean, it's the usual, except for S.T.A.L.K.E.R. We were not distributing S.T.A.L.K.E.R. S.T.A.L.K.E.R. was a different company. But you thought about S.T.A.L.K.E.R., right? That was the most popular game in Ukraine and I guess it's the only, see a lot of people, I guess playin' it. From our products I would say World of Warcraft was the most popular game ever. I mean, it was selling like hot cakes. That was just literally crazy. You know? We couldn't get enough of it, y'know? Into stores. That was unbelievable.
Danny: Was there any games that were very popular in the West, that just were not popular at all in the Ukraine?
Sergey: A lot of like, intellectual properties that are not familiar to Ukrainians were not selling well. Like 50 Cents video games that, y'know nobody, knew about 50 Cent back then in Ukraine. So didn't really sell well. Also was an awful game, to be honest.
Danny: Not many copies of Blood on the Sand sold in Kiev?
Sergey: Yeah, yeah.
Danny: Sergey's greatest love was programming. He'd continued to code during his spare time. But there was something about the distribution business that excited him. Again, he was problem solving. Learning how customers made decisions and using data science to find answers. Well, that and simply watching people.
Sergey: I enjoyed it immensely. Because you learn a lot about how people behave and how people consume games, by just doing a little distribution. And I sometimes, I would just spend like half a day in a store, one of our partner stores, just talking to people and trying to understand how they behave, you know how they're looking and products on the shelves, how are they buying, how they're making decisions to buy, and that helped a lot because, I mean, I like looking at stats and the numbers, but unless you talk to people it's sometimes really hard to understand how they actually think, y'know?
Danny: Sergey would eventually take what he learned in distribution and bring it back to the world of development. He spent two years at Nival Interactive, creators of the Blitzkrieg series and the developers of Heroes of Might and Magic V. He enjoyed the job and life was good. Sergey was married now, he had children. But something bubbling under the surface in Ukrainian society was about to come to the boil. A few days after Valentines Day in 2014, the Ukrainian revolution would see rioters clash with police throughout the capital city. The tragic shooting of unarmed protestors would lead to the ousting of Viktor Yanukovych, the Russian invasion of Crimea, and the eventual war in Donbass which continues today. A frozen conflict taking place on an area half the size of the country. A proxy war where Russian funded proto-states fight Ukrainian government forces, thousands dead on either side.
Sergey: I was in Kiev at the time. My family was still in Lugansk, so we had to move them out of the war zone. And, yeah. But me and my kids and my wife were in Kiev.
Danny: Was it a difficult decision to leave during the war?
Sergey: Well, not really. I mean, when people are shooting outside of your apartment, it's kinda like a natural decision. So, yeah, no. The moment they started shooting, y'know, in my area, I just packed my family and we left. A lot of people don't realize how, how the stuff affects game developers as well. I mean a friend of mine he was still living in Lugansk when the war started. And he would drive to his office and he would like he would hear bullets just flying past his car when he would drive to his office. And it continued for like maybe a week until he's like I'm crazy. There's a war going on and I'm going to a job making video games. So he left after that. But I mean, because it happened all of a sudden and you know you see it in the movies and you expect it to be like in the movies but it's not. It just, y'know, it's a new type of war. You don't see a lot of tanks just rolling in. You don't see like, you don't see the front lines. It just, it's just, people start shooting. So he left and a lot of people did around the same time.
Danny: The conflict led to an exodus of Ukrainian Game Development. 4A Games, developers of the Metro series, relocated their studio to Malta. Sergey and his family left for the Mediterranean island of Cyprus. The reason was simple, it was the closest country him and his family could move to without requiring visas. As it happens it was also one of the 20 or so global locations that developers Wargaming had offices. The Belarusian developer responsible for the wildly popular World of Tanks.
Sergey: Yeah, Wargaming is an amazing company. It's huge and Wargaming is really different from any other companies I've ever worked for. And I've worked for Eastern European companies, not just for the Western companies. Its culture is really something. It's a conflict-driven company. Yes, you're expected to shout at other people in discussions. You're expected to disagree. You know like every time I go to a meeting with my friends at Epic, it's usually I agree with you, I respect your opinion, but in Wargaming you would start with the but part, y'know? You would not do any formalities. You would say well, this idea is incorrect because this and this and this and I don't like this because this. And it really saved a lot of time in discussions, because people know that everyone respects everyone, otherwise you would not be working, y'know? At the company. If you don't respect other people. And that let people express opinions kinda in a more aggressive way. We're getting also, it's really interesting because, the core gaming audience, people that don't usually play video games. So you look at people that play World of Tanks or World of Warships, they are over 40, most of them have families and kids and sometimes they have grandchildren, y'know? And they don't know much about other video games. And they don't consider World of Tanks or World of Warships to be video games. They just consider it to be y'know their hobby. Like they would consider fishing to be a hobby. And that is both amazing and really demanding. Because you know it's a different audience, gamers are used to certain rules in video games and gamers are used to change. And gamers are used to a lot of stuff being taken away. Like people do not complain when Call of Duty releases a new game every single year. You essentially have to re-buy it and they take away all of your progress, when you buy the new Call of Duty, right?
Sergey: Well imagine doing that to a bunch of 60s years old people, you know? Every year. They would probably not like it, right? On the other hand, you hear a lot about in online gaming. And while World of Tanks players are not, not the most pleasant bunch, they are way more polite than your average kids in Call of Duty. So that, likewas never a huge problem in World of Tanks, every time people come and talk about we are free to play game, you're supposed to have a toxic audience. Well, not really, I mean if you're 60 years old you probably know how to behave yourself, right?
Danny: Sergey worked as a Senior Industry Analyst at Wargaming. Helping the team find in-roads into different markets. Aside from their core Wargames, Wargaming published games from other studios and even worked on experimental games, under different brands. Think mobile games about managing a coffee shop. It was varied work that Sergey found interesting. In the spring of 2015, like so many others in the international development community, Sergey took the annual pilgrimage to the Gamers Developers Conference in San Francisco. Here he attended panels, networked with other analysts, and met old friends. One panel he attended was presented by Kyle Orland, a journalist for the technology website Ars Technica. Kyle had created a program that could pull user data from Steam and using it he was able to calculate video game sales. He called it Steam Gauge.
Kyle Orland talking at a conference: I'm Kyle Orland, I'm Senior Gaming Editor at Ars Technica, and this is Analyzing the Steam Marketplace, using publicly derived sales estimates. Now I've been covering the game business for a little over a decade and anyone covering this industry, or following it, one major annoyance is the lack of reliable specific data about sales of games. Now it's not like this in most other entertainment media. It's just not a problem. Nielsen, for instance, provides ratings literally overnight for TV shows and makes the headline numbers very public in publications like Variety. Theaters and studios provide box office estimates every weekend for movies. There's billboard charts for music, there's The New York Times Bestseller list every week for books, et cetera, et cetera. So what do we have for games? For games we have this. This is what NPD, a US tracking firm sends to the media every month. It's a top 10 list based on their sampling of US retail outlets and now electronic sales. If you pay a lot of money you can get more details than this. You can get every game that they track and actual sales numbers, but people who get those numbers are contractually prevented from sharing them publicly. And NPD is pretty strict about enforcing it. You get occasional leaks.
Danny: Back in Cyprus a few weeks later, Sergey was doing market analysis for Master of Orion: Conquer the Stars. Wargaming was publishing the game and Sergey was trying to determine market data around 4X Strategy Games. However, his VPN was down and he didn't have access to any of his data. It was then that he remembered Kyle's talk.
Sergey: Well it was end of March, 2015 I was still working for Wargaming and the funny story behind Steam Spy that my VPN was down and the office was closed for an extended holiday. And I needed to look up some numbers and I didn't have access to my data and I like, well I need this data, because I have nothing else to do. And I was just came from GDC and I remember the presentation by Kyle Orland from Ars Technica, about Steam Gauge. And I said well, how hard would it be to recreate that? And he didn't give any y'know instructions or anything how to do that, but I mean you have internet it's fairly easy. So I spent couple of evenings writing it and by Monday I had all my data, I wrote my documents, required for the office, so by the end of Sunday and I was like, I was stuck with essentially Steam Spy. Without any interface. And I was like, well maybe I should just add interface and open that up to everyone.
Danny: Sergey added that interface, gave it a web presence, and shared it with the folks who listened to his video games Podcast. Right away he saw indie developers flooding to it. This tool, something he was calling Steam Spy, was democratizing data in a way the PC market had never seen before. What Steam Spy was doing was incredibly clever. The Steam marketplace was the biggest online retailer for PC game sales and by default user profiles were public. Sergey's algorithm would poll data from between 60-70,000 profiles a day and using that extrapolate total game sales. It didn't poll every single person on Steam, but with enough data points his algorithm could get to within a few percentage points of accuracy. When NPD produced its top 10 charts, all that that was highlighting was which games were the most popular. But Steam Spy, with its repository of data, was far more powerful. For instance, you could look at trends and see how must more games sold when they went on sale. Or you could use the data to see how popular baseball games were in Portugal. Unlike NPD which just told you a specific thing, if you had an unanswered question about PC games sales, Steam Spy could help you get to the answer. Sergey had developed a tool for market researchers in the video games industry, but it seemed everyone wanted to play with it. It wasn't long before the games press started posting articles using data they had gathered from Steam Spy. Reddit was full of threads about games that were secretly incredibly popular. But it wasn't just hobbyists using it. Indie devs now had access to a powerful market research tool. And even large publishers were using Steam Spy. Were you at all worried that, I mean you were just using the Steam API, right? To pull this stuff?
Sergey: Yeah, yeah, I was, I checked the rules. I mean I'm not a lawyer or anything, but I read the Uler, I actually read it. And I didn't find y'know that I'm breaking anything. They changed the Uler after that. But back when it, I launched it, I was not breaking any laws. And I guessed well, I mean, anyone can estimate anyone's sales, right? That's why we have a lot of research companies. And you have super data, you have Usuy, you have NPD. They all do an estimate and they all the publicize them y'know, online and it is completely legal. Anyone is allowed to do that. As long as you're not stealing someone's, y'know financial information, you are allowed to do estimates.
Danny: And you weren't surfacing any individual's information, were you?
Sergey: No, of course not. No, European laws about user privacy are way more stricter than American laws about user privacy. So all information from the beginning was already itemized. I was never storing anything that is, can be used to identify a user. Well, but coincidentally, it was mostly y'know gaming journalists, small indie developers, gamists, y'know, game enthusiasts, trying to understand how the market works. I was, after started adding more and more professional tools, into Steam Spy, like Cross Audience research, playtime distribution, and stuff that I felt is useful to me. And I've seen that the audience has shifted towards more professionals. And it's been, it's been interesting talking to people that actually use Steam Spy, at different conferences. Intel uses Steam Spy. Tencent uses Steam Spy. Electronic Arts uses Steam Spy. Ubisoft, Activision, you name it, I don't know a single gaming company that does not use Steam Spy right now. It became a tool that a lot of people in the gaming industry use, because it's not great, but it's good enough. And if you look into any other tools available, you know like SuperData Arcade is an amazing tool. App Annie is an amazing tool. But the precision is actually way worse than Steam Spy's precision. And accuracy is way worse than Steam Spy's accuracy. And people still use it, because having information that might be 50% off is still better than having no information.
Danny: One of the things that Steam Spy did great was validating the market. For instance you could use the tool to see if fans of a certain genre bought lots of games in that genre. So, for instance Sergey found that MoBA players rarely played more than one MoBA. So during the height of DoTA2's popularity, when every developer under the sun was trying to make the next big MoBA, they were trying to sell to an audience that largely didn't want one.
Sergey: On the other hand, you look at Survival Games, like DayZ and you see that people that enjoy survival games actually buy a lot of survival games. And that you know that makes it safe to launch a new survival game, like Conan Exiles for example. Y'know you look at the market, you realize well people will buy your game and you make leap of faith. People looking into trends obviously and it's harder to do with Steam Spy unfortunately, I'm using different tools myself, when looking for trends, but Steam Spy is decent at this. So you could look into what's growing y'know how games are changing what people are playing now verus what people were playing last year. If you look into audience for playing on battle grounds, you'll see that while some of them are coming from so that's good, a lot of them are, haven't never played anything before. So they are newcomers to the genre and it means that a lot of them will not leave the game because that's the only game they ever played or played in recent years. And that makes it really hard to compete with and Fortnite on the market, unless you're willing to do something radically different. And that's why I believe it's, a lot of innovation is gonna come from, y'know. People doing Battle Royale but in an unexpected way.
Danny: I'm European. I grew up in Ireland, I lived in London for a few years, eventually found myself in California and now live in the woods on the East Coast. And one of the things I've enjoyed throughout my life, moving from country to country, is understanding the preferences of different people in different parts of the world. As it turns out, Steam Spy is really good at highlighting the types of games that certain countries like. I asked Sergey, what were some of the most interesting geographical trends that he came across.
Sergey: Well my favorite part is the German admiration of anything that has similation in it. Like the farming simulator, anything that has to do with simulation, really. They will play it. Farming simulator is a phenomenon. And it was developed in Switzerland, but is mostly played in Germany. And you talk to anyone in America and the fact that they have a trolleybus simulator they have a trash garbage trash simulator. And people buy it and people play it and that's just crazy, but that's, that's how people in Germany particularly like to spend their time, y'know. Japan, back then was obsessed with zombies. Anything with zombies would sell really well in Japan.
Danny: Was there any stuff that was very popular in America that just was not popular in Europe or vice versa that you kind of saw?
Sergey: Well America is such a huge market and when Steam Spy started, was still the biggest gaming market in the world. So everything that is popular in America was pretty much popular everywhere else. So they have a, well back then they used to like royalgames and open world games. Not as much, like French people do not enjoy open world games as much as Americans. But French video gaming companies like PBSoft it's selling games they make recently, right? They only make y'know open world games.
Danny: Steam Spy was cracking open the sales data of thousands of games. As somebody who worked in the games press, I couldn't imagine this was something that publishers were particularly happy about. The gaming audience is savvy. It cares about consumer rights and it's quick to react when publishers do things that take advantage of them. Steam publishes some data themselves, like concurrent live players. But the amount of data that Steam Spy was surfacing was on a whole other level. I had to imagine that publishers must have been lobbying Valve to do something to lock out Steam Spy. I asked Sergey if he had ever talked to Valve during any of this. I just wanted to know, what did they think of it all?
Sergey: I used to, when I worked at Nivall, I used to work with them, because we published games on Steam and when worked at Wargaming,
Sergey: We also published some games on Steam. And they used to reply fairly quickly. But every time I would mention, well I would not write from my corporate email, of course I would write from a personal email, every time I would write about Steam Spy, they would just shut down. They would, I mean it would just literally, shut up and not reply to any of my emails or any of my communications. And I have couple of friends working there, not on Steam, on the Dotter team and it's the same situation. Every time we discuss something, you know like, gaming related or something like that launch plans or something like that, they talk, anytime I mention Steam Spy, they just shut up. I guess it might be an uncomfortable topic for them.
Danny: Why do you think that is?
Sergey: Well, I feel like Valve is a company that has no leadership. It has no management structure. So there's no one to make a decision. And they only make a decision when everyone agrees to that decision, or everyone on a team agrees to that decision. And there is no consensus about Steam Spy, I guess. And no one is senior enough, like in any other company you would have a head of whatever, head of Steam, come up and say, well that's my decision, we'll shut it down or we will let it go and everybody will, okay! I might disagree with that, but I will, y'know. I can live with that. Any time they make any decision, you will sit and wonder why did they make this decision? Every time they make something new, it feels like a compromise. Y'know what I mean? It doesn't feel like they are making any bold, unusual decisions and it's, to me it has been a probably the biggest disadvantage in the last several years, because they stopped experimenting, they stopped doing something really unusual or bold. Like I mean the trading card game in 2018, really?
Danny: It's difficult to measure the effect that Steam Spy was having on the games industry. He heard anecdotally about games that were funded through market research derived from Steam Spy. He saw publishers like SEGA bring many of their classic games to PC once they saw there was market for them on Steam. But one of the big trends that Sergey noticed was how his tool allowed indie developers to more accurately price their games.
Sergey: I feel especially if you're a young developer it's really hard to put a price tag on your game. You always feel like you haven't made everything you wanted to. You haven't achieved everything you wanted to with this title. So if you're releasing your first game and you feel like well, maybe I should just price it 9.99 because that's a no brainer. But actually your game is worth maybe, y'know 29.99, because if you look at the last games at that price points when they were released they were priced higher, so maybe you should price your games higher. Maybe your game is unique and it has no competition and it has no comparison points. And if it has no comparison points, maybe you should price it higher, because it's something unique that people are willing to pay more money for. People are trained to expect triple A quality from $60 titles and for $50 titles even, but you go below 50, you go to 40 to 30, and people expect it to be an indie game, maybe rougher on the edges, y'know, maybe y'know, better graphics than y'know, $5 game, but they expect it to be an indie title. They are willing to forgive a lot of quirks if the title is actually fun. This is the biggest fear of any game developer I believe. You're making something, you're sitting in a pretty much in a dark room, talking to no one but other fellow developers, from the same company and you always think well, maybe I'm not relevant anymore. Maybe people don't want to play city simulators and I've just spent four years of my life developing one. Maybe people want something to play something different. And maybe I should just under price it and put it for 9.99 and hope that well, maybe if I don't make a lot of money at least people will play it, y'know?
Danny: Steam Spy ran for three years, helping indie devs price their games, helping large publishers do market research, helping journalists find sales figures, helping redditors prove their point. That was until a few weeks ago, when Valve flipped a switch. On April 10, 2018 Valve pushed an Update to every user's Profile Privacy Settings Page. Up until now if you created an account, your game ownership data was public by default. People could set this to private, but most didn't bother. Steam's update flipped this entirely. Not only would new accounts be automatically set to private, but it switched every account on the system to private, too. Without this data Steam Spy could not work. And Sergey quickly announced that the service was dead. At the time the update went live, the EU had just pushed through a new regulation on data security. GDPR or The General Data Protection Regulation was created to add new protections to user's personal data. As soon as it came through, online services around the world were changing their End User License Agreements to be in line with the law. Some services were having to push updates to get in line. One game, Monday Night Combat, would eventually have to shut down, as making the required changes to their backend would cost more than the game was bringing in. Everyone assumed that this was just Steam doing the same, falling in line. But after a few days, Sergey realized it had nothing to do with it.
Sergey: Well it's not really related to GDPR, the latest change was not related to GDPR, because GDPR requires companies to do a bunch of changes to appoint a person responsible for User Privacy to change default settings, to change privacy settings, for underage people, under 18, and Valve did nothing. Like that. Valve still displays your friend list, your achievements, your groups, your screenshots, are publicly on your page. The only thing they hid were games. And GDPR actually does not require that. GDPR requires to hide everything else, that is still displayed. I don't believe it was linked to GDPR at all. I thought that it was like that when they made the change. But after looking into it, I don't think it was related to GDPR.
Danny: So if that's the case, then it must have been related to what you were doing, right, because is there anything else that's happening, that people are pulling from game data?
Sergey: Well, I don't know, I mean, it's on one hand it's nice to think that Steam Spy was so disruptive they decided to shut it down. But it's really easy for them to shut it down. They just have to drop an email to me and I will stop it. I guess, bunch of companies are doing similar stuff to what Steam Spy does. Only keeping it to themselves. Or I've heard of other companies that charges like a thousand bucks per month for accessing the service that does this, similar to Steam Spy. Has a little bit more options, but mostly similar. And maybe they were unhappy about those guys and the only way they saw to shut it down was just shut it down completely, so no one could use it. I guess that's, that's one way to do it. But yesterday they shut, well they didn't shut down, but they made some changes, rendering the Store API useless as well. And the Store API is the API that provides information about the game price, game developer, like the basic stuff. Like genre and so on and a lot of sites were using that and it's now unavailable to them and I mean, what they did, they improved the store's privacy, or what? It just feels really odd to me.
Danny: Without access to games lists and with the Store API changes, Steam Spy was unable to poll the data it required. This was a seemingly insurmountable problem, but Sergey, Sergey likes to solve problems. And in this case he used machines to solve the problem for him.
Sergey: I no longer rely on information provided by an APT at all, I use a bunch of other parameters. As it happens I have an unfinished PhD in machine learning and topic my thesis was using unrelated, using loosely related information to predict economical outcomes. And that's what I'm pretty much using for the new algorithm of Steam Spy. My algorithm that I developed when I was still thinking about taking a science pass. And it works more or less.
Danny: And this is probably like maybe it's a stupid question to ask because it's incredibly complex, but what is the machine learning doing to try and figure this out, if it's not pulling from statistics or from data and creating statistics out of it, how are you coming to these numbers?
Sergey: Well, the thing is that, it is kind of hard to explain. It takes a really huge sample of data like I would say, maybe 15 million data points, and it goes through processing trying to filter out the data that is proven to be irrelevant and trying to amplify the data that is more or less relevant. Then it feeds it into a Neural network. And that Neural network does its magic. And the problem with Neural networks is, Neural networks tend to over feed. Neural networks are great for recognizing images, but are really bad for predicting outcomes that are outside of what they are recognizing. So, if you feed an image of a man to a Neural network and say, it's a man and you also feed an image of a dog to a Neural and say, it's a dog, Neural network will be able to distinguish between this man and this dog, but it's going to be really hard for the Neural network to, if it sees a woman. It will not understand if it's a, y'know if it's a man or a dog, because it does not fit into any of those categories. And in case of our Steam Spy, we're trying to predict well the game is, the Game A has 10,000 owners, the Game B has 20,000 owners, Game C doesn't have 10, doesn't have 20, it might have 30, it might have 40, please do an, predict that and Neural networks are really, really bad at it. But that was my PhD, testing this. Is preparing the data in a way that lets Neural networks actually work with this type of tasks. And it works more or less. It's not perfect, I'm not, I'm still not happy with it, but it is, it works. Yeah, based off of what I've heard from developers and I have a sample of maybe 100 games, y'know that provided me with actual data, it seems that for most of them, for maybe 95% of them, that used Steam Spy, it was within 10%. Give or take. So actually pretty good. For some of them, it is violently inaccurate. The last 5% I mean I've heard about a game that was the difference was 15 times. That was just staggering to me. But for everything else it seems to work.
Danny: Steam Spy started while Sergey was working for Wargaming in Cyprus, but during the intervening years he moved around quite a bit. In early 2016, him and his family swapped Nicosia for Berlin as he became the Head of Publishing for Eastern Europe for an American company in the online shooter space. This company was responsible for some of the biggest shooters in the early 2000s, but they were struggling to find audiences for their suite of online games. One of those games was a third person MoBA called Paragon that would eventually shut down. Another was a remake of their classic arena shooter, perhaps you've heard of it, Unreal Tournament. And the third was a survivalcraft game that had been in development for the best part of a decade. It had sold well on launch, but the game was designed to be very malleable. With Sergey and Steam Spy's help, the team looked at the market research data and decided to take a swing at putting in a Battle Royale-style game mode. Seeing as Sergey was working with the headquarters in America so much, he would eventually move him and his family to North Carolina, to become Director of Publishing Strategy. The American company was of course, Epic. And the game was Fortnite.
Sergey: Yeah, I was part of the team. I was part of making the decision and obviously we were looking at Steam Spy data to see how the genre is evolving. And with talking about Fortnite, original of the Wolf Fortnite, that's the reason I joined Epic. I visited Epic several years ago, they showed me Fortnite and I was blown away. I mean, that was a game that you could make into anything. It is so flexible, it is, I mean, well it didn't have Battle Royale mode, but it had several PBB modes back then. Experimental PBB modes and people you saw 50-versus-50, right? It is actually, well the idea for them all. You know, two teams building castles and fighting each other, was actually back then, in the original Fortnite. Obviously not 50-50, versus, smaller teams. But still. And Fortnite to me felt like a, y'know like a mold, you could make it into anything.
Danny: And I mean even when you talk about Fortnite, it's like we don't know 'cause it's on the Epic, Epic launch, right? So we don't know how many people are playing Fortnite, we don't know how many people are playing World of Tanks, actually now that you mention it, either. So your games have been surprisingly hidden behind this.
Sergey: Well, I'd have to, I mean have access to all the data, but somebody else could. Both of them have APIs that you can access. For World of Tanks, there's bunch of services, statistics services for World of Tanks. And there are several services for Fornite statistics, as well. So you can see the numbers. Actually, it's just Epic is a company that doesn't like to brag about numbers and when we publish numbers we, we've felt some pushback from, y'know from the gaming audience, because they felt like, well, we just were viewing them, gamers, as numbers not as people. And we are really sensitive about that. I mean we're trying, we're always trying to do the right by the gaming audience. So we decided to do it less. It not completely stop it, but just do it less often. After I was, I decided, I actually decided to shut Steam Spy down after all those changes, because I didn't feel like continuing. We also had a huge outage at Fortnite at work and I felt like, well I don't have enough time to, y'know do my day job. I also like to sleep sometimes. This didn't leave a lot of time for Steam Spy, but I thought I've received maybe, 200 emails from people using Steam Spy, asking for me to continue and I felt like, well I mean, yes it makes sense to do so, y'know, people really like it. And that's when I heard all those amazing stories about y'know peoples, companies starting a publishing business because they now were able to see the statistics for game that offered for publishing company getting small indie company from barely getting financing from the German government, because they were able to prove that well, the gamethat they were trying to make is gonna sell. And it did. It was really good. So I felt well, it provides a lot of fire to the market and I like that. And I'm not doing it for money or anything, I mean, at my current day job, I am well provided for. It's not that. It's, it's, the fact that I believe that informational asymmetry, asymmetry of information is unethical, in any business transaction. And Steam Spy is designed to remove informational asymmetry from business transactions or from any discussions. The gaming publisher, the big gaming publisher, have access to more information than a small gaming publisher or a small developer. Then if you're trying to sign a contract with a small developer, you can abuse your power. You have access to more information to get a better deal. That is not gonna be beneficial to the developer. And we've heard these stories about that so many times, y'know even before Steam Spy, like publishers abusing power or big developers abusing small developers. And having this removed actually helps the market whole.
Danny: And do you feel like you're doing a service to the world of video games?
Sergey: I feel like I'm doing more good than harm. In this case, yeah.
Danny: My sincere thanks to Sergey for talking to us this week. You can learn more about Steam Spy and look up all your favorite games by visiting SteamSpy.com. You can also throw Sergey a few bucks a month for his efforts, by heading over to Patreon.com/SteamSpy. Thanks for listening to this first episode of noclip. We hope you enjoyed our first story. If you have any feedback or tips you can hit me up on Twitter @dannyodwyer. Or send us an email, firstname.lastname@example.org. Oh, and hey, if you liked the show, maybe subscribe, tell a friend, or leave us a review on iTunes. If you enjoyed this Podcast but you feel like your eyes are missing out, a friendly reminder, if you want to watch some high-quality video game documentaries for free, head over to YouTube.com/Noclipvideo. We recently traveled to Amsterdam to tell the story of Horizon Zero Dawn. And to Canada, where we filmed a documentary series on Warframe. All of our work is crowdfunded, so if you like what we're making, please consider becoming a patron of noclip. We have bunches of fun rewards, including early access to this Podcast, behind-the-scenes videos and much, much more. Head over to Patreon.com/Noclip to learn more. We'll be back with Episode Two in just a few weeks and we'll be focusing on a game. One of my favorite games, in fact. A game from my childhood. And the creative team who left Lionhead to make its spiritual successor. Whatever happened to Theme Hospital? Find out in our next show. Thanks again, see you then.