The Art of LiveOps

Rovio Entertainment: Michail Katkoff

November 21, 2019 James Gwertzman and Crystin Cox Season 1 Episode 11
The Art of LiveOps
Rovio Entertainment: Michail Katkoff
Show Notes Transcript

In today's episode we are joined by Michail Katkoff, the Head of Studio for Rovio Entertainment, a company best known for its Angry Birds franchise.  Michail has a lot of experience looking deeply at games and game design through the modern lens and will share his views of some industry trends and patterns across different games. 

Support the show

James Gwertzman (00:05):

Hello, I'm James Gwertzman.

Crystin Cox (00:06):

I'm Crystin Cox. Welcome to The Art of LiveOps podcast.

James Gwertzman (00:15):

Hey Crystin.

Crystin Cox (00:15):

Hey James. How's it going?

James Gwertzman (00:16):

It's going good. We're at that point of GDC now where we're kind of halfway over the hump and I think we're all ready for it to end.

Crystin Cox (00:23):

Yeah, it's been kind of exhausting, but we have another awesome interview today. So we've got Michail Katkoff.

Michail Katkoff (00:29):

I work at Rovio as a Head of Studio. 10 years in games now. Started off from Digital Chocolate making Facebook games, then transitions to mobile, first paid mobile than free-to-play mobile. Then all the good stuff. Worked with a lot of talent people, worked on great games, worked on games that failed. Live games, new games, you name it.

Crystin Cox (00:49):

So Michail has done a lot of contributions to Deconstruct of Fun, which is a very cool website where you can go. There's a lot of really deep analysis and insight into games, especially modern Mike live ops games, mobile games, and things like that.

Michail Katkoff (01:05):

I've been also writing this blog called Deconstruct of Fun for the last seven years.

James Gwertzman (01:09):


Michail Katkoff (01:09):

That is now a podcast, that is also a consulting company,

James Gwertzman (01:13):

Right. They're basically doing tear downs for games.

Crystin Cox (01:15):

Yes, exactly. And so, Michail's had a lot of experience looking deeply at games designs, especially in the modern lens. I think he's going have a lot to share with us about what he's seen in trends in the industry.

James Gwertzman (01:29):

Patterns across different games.

Crystin Cox (01:30):


James Gwertzman (01:31):

Awesome. Okay. Let's go right to it.

Crystin Cox (01:37):

I've been making games for 16 years as a designer and a game director focused on live ops.

James Gwertzman (01:42):

I founded PlayFab five years ago because I saw a huge gap in the kinds of access to live ops technologies game studios needed to be successful. We've put this podcast together because a lot of the information on how to do live ops effectively, just isn't out there. There's way more information about how to optimize your graphics pipeline or how to put together effective game design, than there is on how to do effective live ops.

Crystin Cox (02:02):

And since it's pretty tough to go around yourself and just find everyone that does live ops and ask them questions, we did it for you.

James Gwertzman (02:07):

We've picked together a list of some of the top practitioners in the industry, folks who we think are really pushing the boundaries and doing some cutting-edge work.

Crystin Cox (02:14):

And we interview a different one every week and ask them about their experiences running live games, doing live ops, and even having live ops disasters.

James Gwertzman (02:21):

Yeah. Train wrecks are the best, that's where you really learned how to do things effectively.

Crystin Cox (02:24):

So tune in, we have another interview for you today. Considering the work you have done on that blog, let's dive in to talking specifically about design for live ops.

Michail Katkoff (02:39):


Crystin Cox (02:39):

We've talked with a lot of people about the change that's happening, and I think a lot of times when we talk about it in the media, it's sort of positioned as a business change.

Michail Katkoff (02:48):


Crystin Cox (02:48):

But it's actually a really significant mindset change in the way games are designed. So can you talk a little bit about the changes that live ops have brought to game design specifically?

Michail Katkoff (02:58):

So, that's an interesting question, because I've only basically known live ops since the get go. So I haven't done console, I haven't done product that much. And even on the games that were paid, we're turning them free-to-play. But when it comes to live operations, people argue about the word. Some people think live operations is server-side changes, that you change the economies or push sales on or other garbage, it's not garbage, but it's the small stuff and they consider it small things, but they always focus on the big things, like what Zynga calls, the bold beats.

Michail Katkoff (03:35):

In my opinion, live ops is how you grow the game. And basically for that, you have three different channels. The first you have basically your data, and whatever tools you're using, you're getting that numbers in and numbers out, so you're trying to improve the numbers and analyze the data in the best way possible. The second part that you need to focus on, is the community. And that's as important, but they also give you a little bit of a skewed opinion on what should be done. So you have to always use the data as well, because they're just companies that are very community-focused and they serve the community and they might miss the 95% of, of the other player base. And the third part that I think is really important in live operation, is the gut feeling. You have to know your game, you have to know your product. You have to be in it. You have to be one of the biggest lovers and haters of your own game. So you have to also make calls based on your gut feeling, what needs to be done.

Michail Katkoff (04:32):

And those three together transform into the updates, into both client and server-side updates. And for that, the better tools you have, the better you can operate your game. The better you can operate the game, the more it grows.

Crystin Cox (04:47):

So you touched on it a little bit with this idea that there's this community, but one of the struggles is a lot of that community is often what we call the vocal minority and then the silent majority.

Michail Katkoff (04:58):


Crystin Cox (04:59):

So how do you like to actually figure out what the community is feeling? We sometimes say, what are satisfaction metrics that you like to use? Or how do you know when things are working?

Michail Katkoff (05:09):

Yeah. So in my experience the way I've worked with communities, is through a community manager. And what I expect them to do, is I expect them to be inside, whatever you choose, whether it's discord, whether it's Reddit, whether it's your own forums, but being inside of those and interacting, not even interacting, usually a good community manager also sets up the touch points inside the community. So they're not voicing into the community, but they also have the moderators, they're actually interacting with a larger subset of community, and those moderators come back with different type of feedback, and then the community manager kind of clarifies that feedback and brings it back to the game team. So every time you're doing an update, you're always doing something with the community.

Michail Katkoff (05:53):

Now, interesting part nowadays with the community, is that we have now influencers coming. So what happens is the influencer are actually the vocal moderators of the community, and they are skewing the opinions quite significantly. So depending of course in your game, if you're using influencers, you have to be active towards them as well, because they are both the sounding horns of the community, as well as the shifters of what the community is feeling. So that adds the new layer.

Crystin Cox (06:24):

So have you done a lot of work with influencers on your games?

Michail Katkoff (06:27):

Yeah. Yeah, I've done in two different ways. So as I said, the Deconstruct of Fun is also a consulting company and we have people who are experts in influencer marketing and they have worked with some of the biggest publishers in their influencer strategies. I can't name them, but they are big. And through that, I have quite detailed information on what's going on. And also, if people have been reading Deconstruct of Fun, we have more and more stuff on influencer marketing, and that's just due to the fact that we have people joining our group that have experience in influencer marketing.

Crystin Cox (07:03):


James Gwertzman (07:04):

I'd love to jump in and ask a little bit about taking the theme of influencer marketing further. Marketing itself has really changed.

Michail Katkoff (07:09):


James Gwertzman (07:10):

It used to be marketings entire job was simply to get you to do a purchase, and then it's job was over. And I think what we're really starting to see is marketing is becoming part of the entertainment experience itself.

Michail Katkoff (07:19):


James Gwertzman (07:19):

Like if fun begins, it's part of the marketing, whether it's watching a stream or watching an influencer play the game that that's actually part of the entertainment. And then as a result, the ecosystem has really grown in terms of the number of moving pieces game studios have to work with. Maybe we can talk a little bit about how you see the role of marketing and gaming having shifted.

Michail Katkoff (07:39):

Well, it has and it hasn't. It's almost like a dichotomy. On the other hand, it depends on the game type. You have your Match 3 games that are getting deeper and deeper into user acquisition, or hyper casual games that are purely driven by how well your creative performs, whether the creative is good or bad, is actually either making or breaking your game. So in those cases, they have doubled down on the previous elements. And then when you look at other games that we're seeing, everything from Fortnite to Brawl Stars, as well as any other competitive game, they're employing more and more influencers to drive their community and drive the growth of the game, just because the games are streamable and the power of influencers have grown.

Michail Katkoff (08:23):

I mean, YouTube is giant. And then in YouTube viewers, 70% are male, I think it's about 70, even more. So that, of course, puts a certain type of channels that you're interacting. So that's why maybe the games are skewing towards male, also have these influencers through Twitch or on YouTube for mobile that are influencing it.

Crystin Cox (08:46):

So much of advertisement around free-to-play games really comes down to optimizing the channels that you're getting your players through, right? You really want to find out a lot about, you don't just want sort of shotgun out. You want to actually-

Michail Katkoff (09:02):

That also sometimes works.

Crystin Cox (09:03):

Sometimes it does.

Michail Katkoff (09:03):

Depends on your game.

Crystin Cox (09:04):

Sometimes it does, but as you're optimizing your game, you really want to be sure you're picking up quality players.

Michail Katkoff (09:11):


Crystin Cox (09:11):

How have you seen influencers feed into that? Or are we seeing trends like people who come in through influencers or are generally better players or more engaged?

Michail Katkoff (09:20):

Yeah, depends. Depends. So for that, you have to understand your influencer is like any ad channel, so they will bring that type of audience. A lot of people will pay a lot of money to, whether it's Nick at Nite or Chief Pat or whatnot, because they have that Supercell audience.

Crystin Cox (09:37):


Michail Katkoff (09:38):

But they also have that Supercell audience. So whether you come in with one of your videos in the middle of his Clash Royale, Brawl Star streams, that's going to be least watched, some people will watch it, but again, you're positioning against the killers and the people following that channel are interested in Brawl Stars or Clash Royale, or both, or Clash of Clans. So you have to really analyze your influencers and the way I've seen some companies do it, and I think it's smart and Supercell has done it as well, is that they, they look at it longterm.

Michail Katkoff (10:09):

Supercell has been working with influencers since 2012, I think, since Chief Pat was what, 10? So they started early on and they've kind of cultured those connections, and they're using that community quite actively, even with play testing their games before they go to beta so that they influencer feel that they're part of this journey. And once the games launch and they go big, it's not about Supercell or any other big company, well, some of the companies who are coming in late, they have to pay a lot of money to the influencers to stream them. But if you approach it in a smart way, you should start with influencers that are already interested in your game, they have some kind of following. And then you kind of bring them in and start creating that relationship with that person, with that influencer, and as your game grows so will that influencers channel, and he or she will be part of your growth and they will be supporting you. So it's not that much of a relationship where you're paying somebody to do a couple of videos, but you're actually growing your presence.

James Gwertzman (11:15):

Cool. Yeah. So where do you think streaming is going? Do you think that this is a... Because in the beginning, you're right, it was all very organic and authentic, and I think as the role of pay-to-stream or pay-to-influences had increased, there's a danger that it might start to become less-authentic and players maybe get turned off. Do you see that happening?

Michail Katkoff (11:35):

Well, the thing is when it comes less authentic, it becomes less effective. So I think it balances itself very easily. We all see those channels that are clearly shooting a video just for the money. We're not interested in that. You smell and feel and see the inauthenticity, but you also see when something is authentic. When Ninja is playing Fortnite, that's authentic. And then people follow those influencers because of what they do. So there has to be a connection between the influencer, the game, as well as the audience of that influencer and what you're trying to bring forward.

James Gwertzman (12:11):


Crystin Cox (12:13):

Okay. So let's dive more into design. So one of the things that I've found working with designers who are moving from traditional product, box product, to live ops-focused games, is a struggle with understanding how to do, what I call, progression post-1000 hours. So do you have any general advice for designers who are having to start to contend with the idea of a game that gets played for months or a year?

Michail Katkoff (12:44):

Yeah. All the designers that I know are playing games that are being played for years. So the fact that if you're unable to know why somebody would play a game for a year, I don't think a good designer would, actually, any designer, would have problems of explaining why they have spent so much time in World of Warcraft or Magic or any other game, Hearthstone. They spend a lot of time. They don't spend a lot of time playing board games. So the fact that, it's not even a fact, so I would say in moderate world, the designers know what it takes to make those type of games. They just need to be, sometimes, reminded that we're making a game that has to retain for a year and more just like their favorite games.

Crystin Cox (13:29):

Oh, that's fascinating. I find a lot of designs for games that, especially games that attempt to move into the long tail, really struggle with this.

Michail Katkoff (13:37):


Crystin Cox (13:38):

A lot of really interesting design mistakes. I mean, if we want to dive in. I don't know if we want to get really deep into it, we don't have to do anything like where we're arguing, no.

Michail Katkoff (13:47):

Just call out somebody on a design mistake.

Crystin Cox (13:49):

As I see games like Destiny, Division 2, things like that come on, I see them really struggle with progression design and make a lot of really significant mistakes in the way that their progression is put together.

Michail Katkoff (14:01):

I mean I'm not an expert in console, but I think sometimes it's just, especially on console, it's the schedules that really bother them. So with those tight schedules and with those cost amounting, they sometimes have to launch a little bit earlier, just like in case [unfamiliar 00:00:14:14]. So they didn't have time to build out everything that they needed to build and they have so many bugs and issues that they have to focus on first. So I think that it's more on that track.

Michail Katkoff (14:25):

Oftentimes what you do see is that the designer, whoever's in charge of design, doesn't have experience in live operations. So they haven't been on a live game and when they come up with their live operations strategy, they don't understand how much work they're actually and what kind of content treadmill or content hell they're putting themself into. And that usually happens with designers who just don't have the experience. So they can design a game that retains, but the next time they do it, they do it in a far more efficient way.

Crystin Cox (14:56):

Right. Yeah, definitely. No, it's fascinating to see. I actually think that skill, the ability to really buckle down and do great systems design, is maybe more rare than you think it is in the industry.

Michail Katkoff (15:13):

Yeah, yeah. Well maybe I've just worked with good people.

Crystin Cox (15:16):

You're an optimist. Yeah

Michail Katkoff (15:16):


James Gwertzman (15:17):

Well, okay. So what are some kind of common mechanics? I think when you have a game designer, like you mentioned, who's new to live ops, who's not done live ops themselves and they're sort of getting started, where would you point them at? What are some common systems or mechanics that they might want to look at as getting started? What are the building blocks of a game that's going to be able to have that longterm progression?

Michail Katkoff (15:38):

Okay. So first he started off with the economy. So the economy has to be robust enough, or it just has to have enough sink so they keep on adding new and new elements. And for any designers looking for pointers or elements that they need to learn, I would say it's mandatory for them to play an RPG game. I'm on an RPG game. Just choose one. I don't care which one, choose one with Marvel, choose one with Star Wars, choose the Idle Heroes or whatnot, but they are so rich in the systems that you can just pick and choose and you won't run out of different systems to add.

Michail Katkoff (16:18):

But overall focus on the ones that add social gameplay, and focus on the ones that competitiveness. I don't know if you guys do the Peloton Bike, they have that at the hotel where I am now, and it's a stationary bike, but it's a stationary bike with a leaderboard with friends. That's it. It's the best stationary bike out there.

James Gwertzman (16:40):

Live ops comes to exercise.

Michail Katkoff (16:42):

Yeah, exactly.

James Gwertzman (16:43):

Well, you mentioned sinks just now.

Michail Katkoff (16:46):


James Gwertzman (16:46):

And actually, we haven't really had anyone talk about that concept yet. Could you go a little deeper into the [unclear 00:16:50] of sources and sinks and why those are important to longterm game design?

Michail Katkoff (16:53):

Yeah. So when we talk about game without sinks, we could say League of Legends, for example. There's no progression outside the core gameplay. So that means that there's really not a lot of sinks that you could play around with, not a lot of elements you could do. So your live ops is basically adding new characters and skins on top of those characters, and just hoping that, not hoping in that case, they know the players are playing with all the different characters, just deep enough that they have to learn everything and that creates actually entry barrier towards the end.

Michail Katkoff (17:25):

When it comes to sinks, well then we can think about game like RPG games. There's different curves of power progression for each character. It's not only that they're getting power by buy gathering the XP so that they can level up, but then you have to ascend that character. Then you have to... It's running out of my mind. But anyways, there's multiple different ways to level up the character, there just called in different names, and all those different ways have usually different type of sinks, different types of elements you need to gather. And oftentimes they are even, not only put under different resources, but they're only put off of different type of items that are gotten from different types of game modes.

Michail Katkoff (18:08):

So essentially what you're making, is you're making your game deeper by having the player to interact with various game modes, because there's so many elements that you need to gather in order to level up that character, and each of the pieces of that puzzle are gotten from different type of game plays. So player has clear incentive to play those different game modes. And once they're playing those different game, some them social, some of them competitive, some of them are just single-player, they're actually spending enormous amount of time in your game, and now it becomes a habit, and now it becomes their hobby, and now you have people retained.

James Gwertzman (18:42):

And then one of the common things we talk a lot about, you mentioned economy, economy is critical, and good economy design. Are there any classic mistakes you see in designing economy? Or if you make some mistakes in the economy, how can that really kind of screw up your game?

Michail Katkoff (18:55):

I mean, it screws up everything, honestly. Economy is the lifeblood of actually everything. So part of the economy is tuning, it depends on different type of games, what this difficulty is we're talking about, or there is how much you need to get different types of elements. And that's where the live ops comes in, because you're constantly moderating and seeing how people progress through the levels through the elements of the game, and sometimes if the goal is too hard to reach, they drop out. If the goal is too easy to reach, you see players churning through because they go in through the economy and that puts more pressure on you adding new elements or you're not monetizing so well. So you're constantly trying to find the perfect balance and with modern games, you know, the perfect balance is actually based on a cohort, on a type of a player, instead of the average best progress.

Michail Katkoff (19:43):

So economy is super important. I don't want to call out specific games that are failing in economy, but what some of them... Out of the recent ones that I would say that had issues with the economy, I would say Command and Conquer for sure that came in on mobile. So I think they had some issues with balancing the game, and then you can kind of see that there's a lot of duct tape put on top of the game. Duct tape in form of different type of artificial coolers, elements that kind of don't make sense for the game. And those are usually the reactions when you've done a mistake in the economy, then you start duct taping it with different types of features that just don't feel authentic, they feel punishing towards the player, and in the end, things kind of crumble because it doesn't feel good. It doesn't play good.

Crystin Cox (20:37):

So when that happens, though, especially if you have a game that's got a fairly sophisticated number of materials and different progression curves inside of it, inevitably, if a designer, especially maybe someone new, does make a mistake, what is some advice you have for recovering without getting that duct-tapey feel?

Michail Katkoff (20:55):

Yeah, so it depends. I mean, you can avoid some of the mistakes by interacting with the community early on and really testing some of the features. I know Small Giant tests some of their biggest features with a community, so that way when you're pushing... And you know, A/B tests are possible, but only at certain level, once you start adding big elements to the game, you can't really A/B test them anymore in competitive.

Michail Katkoff (21:20):

But, okay. So you make a mistake with the economy. I think that's a little bit less common these days because you can test it, you can run it through Unity through various tests, see how it performs.

Crystin Cox (21:39):

Do modeling.

Michail Katkoff (21:40):

Yeah. Do different type of modeling, do various things. But if you run through some problems, then just do a hotfix. If the drop rate is not high enough, it's slow enough, it's not even a hotfix, then you can just tune it through the backend.

Crystin Cox (21:54):

Right. Yeah. I always try to suggest that people use a multi-pointed economy, right?

Michail Katkoff (22:00):


Crystin Cox (22:00):

Like don't create a perfect pyramid where every item leads up to only one ultimate sink, so that you have multiple different points in that pyramid that you can balance that out against because then it becomes a lot easier to trade those against each other if you have to make adjustments.

Michail Katkoff (22:14):

For sure. For sure. I've done tons of mistake in live operation. And the first game I ever worked at, I remember we were adding, the game was called Army Attack, it was Facebook game, and we were adding this new map and we wanted the players to kind of get there, but we knew that they want a new map. So we asked them to build an airport cause it was like an army game.

Crystin Cox (22:34):


Michail Katkoff (22:35):

And we knew that we don't have the map ready, but we brought the airport a week earlier and we kind of said like, "Hey, just grind for these helmets." And there weren't enough drops and players got super pissed because they ended up paying to get the helmets and they got it and the map said like, "Well, we're actually bringing it next week even though you paid for it." So, yeah, those are kind of like the early days, back 10 years ago.

Crystin Cox (23:00):

Those are always fun. One of the first MMOs I worked on, we put a cap on the amount of gold that a player can have and it was not that high.

Michail Katkoff (23:11):

Yeah, yeah, yeah.

Crystin Cox (23:11):

And then like a couple of years in, players had done what you inevitably know that they're going to do, they had started using other rare materials as the real currency, and then we had to really like design-

Michail Katkoff (23:23):

Was this Ultima Online?

Crystin Cox (23:25):

No, it was Maple Story.

Michail Katkoff (23:25):

Oh, Maple Story.

Crystin Cox (23:26):

Shh, don't tell anyone that.

Michail Katkoff (23:27):

Yeah, yeah.

Crystin Cox (23:28):

Well, and not that anyone doesn't know that about Maple Story to some extent, but it's certainly not our finest hour.

Michail Katkoff (23:34):

So we're talking about the same thing here. We're talking about doing the mistake of not actually being able to even fathom what players would pay for and how much they would pay for.

Crystin Cox (23:47):

Yeah. Right.

Michail Katkoff (23:48):

And how fast they actually go through the content, like the true, true, true fans of your game. And you'd think like, "Come on, nobody plays this for three hours a day." And you're like, "No, they played for nine." And they're like, "What?"

Crystin Cox (24:01):

Scale was really challenging for us on Maple Story, which makes sense, right?

Michail Katkoff (24:04):


Crystin Cox (24:04):

Maple story was definitely sort of an organically-grown game, but then it was so large that the systems were not really built to manage that many people, which I think is a great piece of advice. Never design assuming you won't be successful, always leave yourself open for success.

Michail Katkoff (24:22):

Yes. Yeah. Yeah, yeah. And then always assume that the craziest thing from your players, because the thing is, it's punishing to raise the threshold, it's punishing to make the game more tougher, but you actually get positive feedback from the community, in a way, make it easier.

Crystin Cox (24:38):


Michail Katkoff (24:38):

So always start off by making things OP and then bringing it down, or making things too difficult and making them more-easy, rather than making things easy and making it more difficult towards.

Crystin Cox (24:49):

And then even better, if you start off really OP and there are some players who did it anyway, reward them, like really acknowledge that, and then make it easier, but make sure you reward the people who did it when it was really hard.

Michail Katkoff (25:02):

Yeah. 100%.

Crystin Cox (25:03):

It's always super fun.

Michail Katkoff (25:04):


Crystin Cox (25:07):

One of my favorite things in maple story was we had these statues, so when we first launched the game, it was very, very difficult to reach max level. Like it was supposed to be impossible, which also never do that, don't design things to be impossible, nothing's impossible, so it was very hard to get to max level. So when people did it before we readjusted the progression curve, they got to make statues of their characters that then lived in the town forever.

Michail Katkoff (25:31):

Nice. Yeah, yeah.

Crystin Cox (25:31):

So the first X number of people who got to level 200 had statues.

Michail Katkoff (25:35):

Oh, that awesome.

Crystin Cox (25:35):

Which is super fun.

Michail Katkoff (25:35):

That's an awesome reward for players.

Crystin Cox (25:37):

Yeah. But then you can still ease it up so that later players could get there easier if you want.

Michail Katkoff (25:43):

Of course, of course.

Crystin Cox (25:45):

Well, we've talked a lot to the other people that have been on about analytics because everyone loves analytics. Especially coming from the design perspective, I'd love to talk to you about analytics. I think that there's a lot of tension in the industry around the way designers interact with data.

Michail Katkoff (26:06):


Crystin Cox (26:06):

There's a lot of debate around data-informed or data-driven design. What are your thoughts about the relationship between designers and data?

Michail Katkoff (26:18):

That's interesting. It depends. It depends a lot on the top of the game that you're doing. If we're talking about Match 3 game, design better be deep into data or social casino. But yeah, it also depends on the type of the team that you have. Like, do you have a product manager on your team and so forth? Let's talk about, for example, network.

Crystin Cox (26:46):


Michail Katkoff (26:46):

In their game, Legendary, the way they structure it, their designers are creating the features, the product managers are then managing. So they're kind of thinking ahead and discussing it, and of course they are certain number or things because we talked about economy before, we've talked about live operations. Those are all the elements that drive the development of different features.

Crystin Cox (27:10):


Michail Katkoff (27:10):

And those features are then, usually, designed by the designers.

Crystin Cox (27:14):


Michail Katkoff (27:15):

I've had multiple experiences of designers designing and then product management designing, and I hope that every time we have a designer, I hope that they are the ones that are trying more creative approaches, because a product manager can always come in with a benchmark and say like, "Hey, let's just do this."

Crystin Cox (27:33):


Michail Katkoff (27:33):

So, in short, I'm bummed if designers are too driven by, I won't say numbers, but by benchmarks. They should be inspired to reach targets with their designs based on the numbers we're seeing, based on the issues in the game. They should be solving those through the design, but I like when they try something new.

Crystin Cox (27:52):

Right. I would like to say design brings psychology.

Michail Katkoff (27:56):


Crystin Cox (27:56):

Right? Product management brings data, they bring science, which you need. It is super important. But designers should bring psychology. They should understand the way humans think and act.

Michail Katkoff (28:08):

Yeah. Yeah. The thing is product managers do too. Behavioral economics is kind of like the key element.

Crystin Cox (28:13):

Yeah, but I think they think of it more. They tend to, of course, there's some product manager who's just fantastic and they do everything, but they tend to, I think, in my experience, think more about people as statistics, sort of as groups. And then designers tend to bring that more empathetic, personal touch, because sometimes you have to look at the averages, you have to look at the numbers to design, but every person who experiences your game experiences as an individual, and so their personal experience also has to be designed for.

Michail Katkoff (28:49):

That's interesting. So I kind of go to my background, even studying, back in the days. I studied and I did a lot of work with, so I was in a business school, but I also did work with students who were studying design, as well as with engineers. And it was kind of like this sort of... Well, now it's called game development because all three coincide, but back in that time, what I noticed is our ways to solve problems were inherently different for these different three groups.

Crystin Cox (29:18):


Michail Katkoff (29:18):

So the way business people solve it, and by business people, I mean mostly product managers and executives, it's we kind of zoom out and look at the market as a whole and then started breaking down. The benchmarks, we should be here, we should be there. Let's take a pile of this, let's take a [unclear 00:29:33] of that.

Michail Katkoff (29:34):

But with designers what was really interesting, is that when they started solving a problem, they started from a detail, from a minuscule thing. Like if you look at Hay Day, it's more like, how do you collect wheat? And that's where you start the game. And a product manager would never start from a single feature, a single prototype of, from this point, we're going to start building the game.

Crystin Cox (29:54):


Michail Katkoff (29:54):

They would start from, let's look at the farming category. What is FarmVille doing? What have they done before?

Crystin Cox (30:00):

They'd be top-down.

Michail Katkoff (30:00):


Crystin Cox (30:01):

Designers love bottom-up.

Michail Katkoff (30:02):

Yes. And then with engineering, on the other hand, they solve problems the way it's the most efficient way to solve it. Doesn't have to be always the right way, but it's the most efficient way. So kind of like balancing those three, that's where the sweet point is.

Crystin Cox (30:18):

Yeah. I think that's one of the things that's wonderful back in development, is it really is all three of those attitudes coming together to make something.

Michail Katkoff (30:25):

But it can be like, you go too much design you end up indie, you go too PM, you end up just being another incremental copy of something that is not any better. And you go too much engineering, and it's nothing.

Crystin Cox (30:42):

I've definitely known some games that are engineering-led.

Michail Katkoff (30:45):


Crystin Cox (30:45):

But I won't mention them.

Michail Katkoff (30:47):

It might be like a feature, like some kind of a very complex thing that they're doing. And it's not really a game. It's like, "Okay, I get it. The technology here is impressive, but."

Crystin Cox (30:57):

Technical showcase.

Michail Katkoff (30:58):

Yes, exactly. It's technical showcase of like, "Look at these shaders, look at the VFX, the animation." Like, is there any game here? Well, you know.

Crystin Cox (31:07):

I mean, there's a spectacle.

Michail Katkoff (31:08):

Yeah, exactly.

Crystin Cox (31:09):

Right. It's all different levels of each.

Michail Katkoff (31:11):

Yeah, yeah.

Crystin Cox (31:13):

So for people that are getting into the industry today, what would your advice be for them?

Michail Katkoff (31:20):

Ooh, my advice for people who are getting into the industry today? It depends on what they're doing, but I would say overall, it's just be open to things. Quite often what I see and then what I've personally experienced and what I have struggled, is in the beginning you think you know how things are done, and that goes for everything. And once you start getting into that mindset of, "This is how you do things." Then that tells me, usually, that you're a junior, and that limits your growth.

Michail Katkoff (31:56):

Instead, start having an open mind towards how things can be done and analyze it and find your way through being open to different ways of working. Different approaches, different type of games, and let it just organically come to you on how do you want to make your games.

Crystin Cox (32:17):

Cool. I think I already covered this, but I've been asking everybody to share live ops disasters. We kind of got a little bit of one from you, but do you have any other really memorable live ops disasters? Everyone loves to hear live ops disasters.

Michail Katkoff (32:30):

Yeah. Those are the ones. The large scale disasters are usually around sales and you're running, you're trying to impress somebody by just creating these spikes and you ended up spiking your game to hell or just creating this content treadmill of sales.

Crystin Cox (32:44):


Michail Katkoff (32:45):

But most of my live ops mistakes are just around of not really fathoming how engaged the players can be.

Crystin Cox (32:54):


Michail Katkoff (32:54):

And thinking you can get away with shit by just putting like, "Nobody will do it."

Crystin Cox (33:03):

Somebody will do it.

Michail Katkoff (33:03):

That's usually like, "But nobody will do it." And then a lot of people do it. And just not being able to fathom how much time they also invest into your game when you're balancing things from the get go and you end up balancing them way too low.

Crystin Cox (33:20):


Michail Katkoff (33:21):

Based on, whatever, eight sessions a day. And that's actually not true, actually they can get this through three sessions and so forth. Then you're spiking things up. So most of my learnings are on a high-level and yeah. Yeah. Those are the things.

Crystin Cox (33:36):

Yeah. It is hard. I think we have trouble fathoming just, one, how many of them there are. When it comes to players, it's not that it's difficult to fathom, it's just how many of them there really are. They're always going to outnumber you.

Michail Katkoff (33:47):

Exactly. Exactly. So other than that, like big disasters... I mean, I worked on disaster games, but they were just disastrous because either the core game was just not good or like my biggest mistakes, in terms of game development, have come through the aspect of just me thinking that I know how things should be done and I was just too cocky when designing the features and then driving through like, "Yeah, I got this." And then nope.

Crystin Cox (34:18):

You mentioned it a little bit, but gosh, I really wish we could get the industry to embrace the idea of price protection. We are very ready, at any moment, to devalue our work. We really love to put things on sale.

Michail Katkoff (34:32):

Yes. Yeah, yeah. And it's sad. It works when it's connected to the real world.

Crystin Cox (34:38):


Michail Katkoff (34:39):

Whether it's Christmas or Black Friday or whatnot, then it makes sense, but if it's Thursday, so we're running a sale cause we're behind our targets, then it's just not a good thing.

Crystin Cox (34:50):

Player are very savvy, right? Like if you say, "We're putting this on sale but there's no reason." They get it. That means it's worthless.

Michail Katkoff (34:58):

It's better then, usually, to earn that sale. I like what now is happening is like you level up and it's a level up package or stuff like that.

Crystin Cox (35:06):

Yep, yep. Absolutely.

Michail Katkoff (35:07):

So if you're like, "Okay, cool. I got this because of this and it's time and so forth."

Crystin Cox (35:10):

Yeah. Well, thank you so much for coming by.

Michail Katkoff (35:13):

Thank you. Thank you for having me.

Crystin Cox (35:14):

I think we're just about out of your time, but this was great. Really glad we got to do this talk.

Michail Katkoff (35:18):


Crystin Cox (35:21):

Thanks for listening to The Art of LiveOps podcast.

James Gwertzman (35:26):

If you liked what you heard, remember to rate, review, and subscribe so others can find us.

Crystin Cox (35:32):

And visit for more information on solutions for all your live ops needs.

James Gwertzman (35:37):

Thanks for tuning in.