Maximize LTV and Understand Player Behavior

A gaming company needs data and analytics to understand player behavior, predict LTV, and innovate live ops. Understanding this will allow them to improve player experience and grow their community.

Transcript

Peggy: Analytics are table stakes. They tell you if your game is a hit or a miss.

So what do you need to know to continually deliver new features, updates, promotions, in-game events, improvements to your game? The list goes on and on, and how do you know when you are hitting it, when you need to dial your efforts up or down based on the metrics? Well, guess what? We're going to answer those questions, tough as they are, because we're going to do a deep dive, literally, into Dive.

Dive is a game analytics platform and service that helps companies like Gamefam maximize their LTV, understand player behavior, and we're going to get the inside track from its founder and CEO. He's an entrepreneur with a solid track record heading two successful companies in the online gaming and social casino space and one of these was House of Fun. He's active as an angel investor, an advisor, a board member, working with startups for the last five years and he has poured all that experience, that pure gold knowledge, into Dive. Elad Levy, welcome to The Hive. We're going to talk about The Dive on The Hive. How's that grab you?

Elad: Sounds great, thank you very much, Peggy. Love the intro.

Peggy: Now, you founded it based on first-hand experience in the industry. We talked about that. You've been out there and you sold a company to Playtika.

Elad: Working on a top grossing game studio is an amazing experience and having that front seat role is a privilege. I think I developed and redeveloped and redeveloped the same tools over and over several times until I got it right and we actually got acquired by Playtika. Playtika asked, "Look, we love your tools, can you bring them to the rest of the game studios?" But they were really tightly coupled into the game itself that it was very hard disconnecting. It's actually happening in a lot of game studios that they develop tools that are slightly coupled into a specific game and then they start growing and it's not that easy decoupling them out of an existing game and moving them to other games because every game is just, the dynamic is different and I think that in a way this is how Dive started. We wanted to kind of build a platform agnostic decoupled tool and then be able to connect that tool to different platforms, different games, different genres, and then over time Dive became like just the clients just started asking for stuff and we ended up developing the liveops dashboard, they asked for A/B testing we did that, segmentation and we just added more and more and more features into the platform and became what it is today.

Peggy: You talked about the live ops, the a/b testing. What does it now cover?

Elad: So first of all today we are running over 80 games with 50 million monthly active users which is insane. It started like as a boutique business that I bootstrapped myself with the CPO and today we are 20 people and processing like five billion events every month of data and it's insane, it's really... I don't even have how it got that. Sometimes I just woke up and I don't know how it got there, but yeah. It started with data as the building blocks for everything but then we started hooking up with marketing attributions and Google and Facebook and centralizing everything in one place and CRM and you know email and push notification and web push and then the live upstart part started for the calendar events of like orchestrating game events. Because we came originally from building a game in a game studio we know how the mindset of game developer works and what they need when it comes to data or live ops or marketing so we can really identify with our clients and hit on that spot of what is required and we got our first publisher Carry1st and you know all of a sudden we you know manage BI for like bigger and bigger and bigger companies and about a year ago Game Frame approached us and that was a beautiful challenge because Roblox is kind of like this whole metaverse buzzword. Voldex appeared and we launched with them a Minecraft game and I think we're the only one doing Minecraft analytics out there so that's really exciting as well.

Peggy: It's exciting that you can also grow with the market and go in these other directions as you are. Was that something you set out to do, Elad, or did it just happen?

Elad: I personally in House of Fun had like a third party analytics tool in the beginning, when we started. Like one of those cookie cutter SaaS dashboard. And we threw it away after a year because we realized that when you start scaling it's not useful because analysts need raw data and product managers need to manipulate that data and they need a lot of customization. So when we started Dive we said there's no way around it. I know, I know it sucks, I know it will be hard to scale but we must accompany the platform and customize it for every client, it's just inevitable. Salesforce understand that. SAP understand that. You know those are platforms that are being tailored for every business. I think that games, we understand it and that's why we have this approach of customizing our platform for every client, every game, every game studio. And because we started this way then if you throw at us a new platform or a new backend or a new bus or whatever, it's easy for us because the company is like... basically the company structure is half developers half data people, so we're accustomed to... it's not that when you ask us for a feature you need to wait now one year for it because we have huge backlog. Sometimes in like a couple of weeks we already delivered to production because we will build with customization in mind from day one.

Peggy: You're all about leveling the playing field for game studios, you know, allowing them to compete with the major names, without building an in-house data team. That's sort of it.

Elad: That is correct.

Peggy: You're delivering a ton of valuable insights without a ton of tools. It's customized. It's a dashboard. Tell me a little bit about what you see, what you extract, what you really offer the games studios and how they can apply it.

Elad: I mean there are two things that we focus on data. One is insight and the other is action. When it comes to insight it's about centralizing everything in one place and that's the data warehouse that we built for every game, for every client. So marketing, crash reports, gameplay, game economy, inflation, retention, engagement, session length, revenue, everything in one place. So insight is the first thing that we do and obviously we start small, we start with the basic stuff. You know the basic KPI that you wake up every morning and you want to see the DAU and retention and engagement and session length and play time and revenue ARPDAU, ARPPU, that's like the basic stuff. But then over time we start evolving that insight into LTV, marketing, prediction, stuff that usually is done in big companies but for us... someone called us BI in your pocket team. Like bring that enterprise BI to smaller game studios and it's a huge help. So that's the insight part that we really focus on and the second part is the action. And the action is... the term data driven games is a very... it's something that you know they throw around very often but i don't know if lots of people really understand what it means. And how to really take insight from that data and put it back into action. I'm not talking about pricing. A lot of people are obsessed about pricing and I'm not talking about that. I'm talking about, let's say that you segment a portion of your players and you want to offer them specific content for example and then that content you wanna a/b test if it's working or not. You cannot treat the same way someone who plays five hours a day and someone who plays five minutes a day. So when you segment those players and you want to offer them different experiences, this is something that has to be connected to data. The same custom approach that we tailor in our reporting for insight we do the same for the live ops part, for the action part, so it's fun, you know. We have segmentation based on, I know one game is based on trophies, the other game is based on matchmaking and winning, and it goes on and on and on. That was our base assumption from day one. Everything in the games industry is an exception and must be treated as such and therefore we have to tailor that experience for every game and every game studio. And that's just the way we work in everything. We had cases we had to redevelop our entire SDK because a game studio had this really special build pipeline for their game and QA so we redeveloped the whole thing just for them and it's fine. For us it's fine, this is like normal stuff. This is... our day-to-day is like that.

Peggy: And it's always learning, it's always changing, and it's great that you can, that it is so flexible and customizable. That you can move with that. And in prep you told me something interesting and I made a note of it. I said, "no, I gotta ask him what he really means by this", because you said you deal with the ugly side of data. What is that? What did you mean by that?

Elad: Let's say that there are two major players in a data team. One will be the engineers. They are the ones that collect, centralize, normalize, clean all the ugly data. That's a lot of hard work. They are the ones that wake up at 2 a.m on Sunday, when something fails, when the database is too heavy. This is like the data engineering part and I call it the ugly side of data because it's really horrible day-to-day work. And then there is the analysts or the data scientists that take that data when it's clean and shiny and perfect for querying and testing and analyzing and then they look for insight in that data. So this, the whole data engineering part is ugly because it's a lot of hard work, annoying, struggling, you know, with databases, late night, that. It doesn't matter what technology you have it will always reach a point when you start scaling that it will start failing. It's normal, it's part of the process. It's like any game company, you know. You have more traffic, you have more data, you have more everything, and it starts failing. So, that's when it starts getting ugly because if you made the wrong decision along the way of the way you built that infrastructure then you will pay for it.

Peggy: And I hear you thinking from experience here, Elad, for some reason.

Elad: Yeah, of course. So many nights, so many nights waking up because the CEO wants to wake up in the morning and he wants to see a report and that report can never fail, but you have no idea how many times it fails even in huge companies. That's what we do best that. The whole ugly part, this is what we do best. We developed our own monitoring system for controlling this whole thing and we developed a ton of proprietary tools to control the processes of collecting the data and aggregating it and normalizing it.

Peggy: I'd like to have a little bit of your perspective as an angel investor, a VC and put yourself in those shoes. Where and how does this pay off for an early stage or VC backed game studio?

Elad: Where does it pay off? It pays off because they can focus on the game and get amazing insight. Like one of our clients and he said, "Dude, you're basically offering BI as if it was Zynga for a fraction of the price, you know, and that's amazing. I'm used to all the dashboards because I'm coming from top grossing game studio and you offer all that for, you know, with a fraction, with great support, and everything." Because we also, we create a Slack or a Discord channel for every game and every game studio, which is insane, but we do that because if there's a lot of back and forth with analysts or product managers, and to facilitate that, we we do that kind of support, that kind of personalized support.

Peggy: I've interviewed a number of VCs and there's another angle to this. I'd love to have you weigh in on this because some would tell me, you know, when I see companies watching this, taking care of this, thinking about this. You know, how do I acquire more effectively? How do I find a way to get literally more value extracted out of the data with a company like yours, you know? It's thinking about that, thinking about the shortcuts, the workarounds, the ways to do this intelligently and effectively, and they love to see that.

Elad: We ended up making intros sometimes to VCs because at some point I'm speaking with managing partners and they say, "Hey!" I mean this is actually how we spoke with Pollen as well, because, I mean, hey, you're on top of the data. You see everything. You see the gameplay, you see the marketing, you see the LTV. If there is a good opportunity to invest, let us know and we actually do that. So we do intros to VCs or to Pollen for UA loans or stuff like that, because sometimes I see the KPIs and I say, "Man, I would put my own money there." It's too good to be true, you know. Some game studios are like that. And it takes time. It takes sometimes a year, sometimes two years, to polish KPIs. Live ops is something that comes in a later stage when you start scaling but in the beginning the insight is important. Insight on the marketing, on the game economy, on the crashes, on the retention and engagement and stuff like that, but then once the company starts scaling then this is where LTV has the most important weight. And this is where a good live ops helps because if you segment them you offer them good pricing, good content, you do good a/b testing and you really build enough in-game events to keep them engaged. Then this whole thing becomes a whole different game and yes, we are, therefore leveling the field for those game studios to have a chance. We actually promote a lot of self-publishing, but by game studios as well because I don't know. It's kind of a classic, conservative approach you can say that I have.

Peggy: That they're self-published...

Elad: Yeah, I think so, because when... I mean I think it's more accessible today, or more possible these days, than it used to be in the past. You know with a small investment you can somehow start to get the ball rolling, hook up with Pollen for example for the UA loans and then the whole thing starts just like you know feeding itself and the cash flow just starts growing by itself. So we kind of like complete that part in the puzzle of, you know, the data part, the live ops part, helping to maximize LTV and getting the right insight in the beginning.

Peggy: What are some top tips to get the most out of what you offer, what Dive does and offers? How do I get the most out of this? Because it sounds like a, you know, a fantastic service, but you have to understand it to get the most out of it.

Elad: Exactly, exactly, and this is why, I mean, the chemistry with early stage VC backed game studios is so good because the people there are ex top grossing game studios. The VCs usually invest in those kind of teams, so they know what they want. They know how to approach data, they know how to... the same way they know how to lead a vision for a product, they know how to communicate. Because data serves the product manager the same way tech serves the product managers when it develops a feature. So it's very important for us when we start engaging to have someone who drives the car, because someone needs to ask those questions. Data, it's something we repeatedly say in the company, data starts with questions. Out of the people that completed this mission how many ended up converting? Out of the ones that we sent email to dormant users how many I know logged into the game? There's always a question behind it. We actually in Playtika used to work this way. We built a product spec document and at the end of the document we would write the question around the feature line. Because developing or deploying a feature without asking questions about that feature, whether it worked or not, is just wasting time. You cannot look at DAU and ARPDAU and stuff like that. You can obviously but it's too generic to understand because you maybe launched a feature but at the same time the UA guy launched a new marketing campaign and everything gets skewed. You have to ask really specific questions about we launched this bonus feature in the game, how many people used it? And out of the people that used it, did they repeatedly used it? And whether it... how did it affect them? And it's not only coming from data it's also coming from reviews, whether it's story views or emails and customer support that they say, "Hey, that feature sucks." You know. Data tells one type of story and the community tells another type of story and tapping to that into that community is critical.

Peggy: The questions are almost as important as the actions but say someone's listening in right? They're seeing our show. They're saying, "Hey! Yeah, how do I make the biggest positive difference in my gaming company right now?"

Maybe something about how to frame those questions.

Elad: So, it's one thing like if you want to, I don't know, you need to jump the revenue really fast so you can make some sort of like a limited time offer for the weekend or like some huge event and it's another thing of doing something that is more deep in the game, because the progression system shows that people churn in day 60. And when you start going deep into that you find out that people reach, we call it, an economy wall, like in the progression system they try to upgrade something or some item and they get frustrated and they just churn because they cannot do that. So that's kind of like advanced insight that you find out with with good data and quality data over time. It's the same thing with you know and I know if it's a match three game then then there will be like an economy wall somewhere because by the end of the game, those games rely heavily on human being making them, which is fine, that's what makes it so fun and we obviously make mistakes when we design game economy for a game and we assume something that is based on play time. We assume that over time the player will get more skilful in the game and will be able to pass this level or that level. Sometimes we are wrong because some people are amazing after one hour of playing a game and sometimes you would play like five hours and you will still stuck and it happens. It's kind of a balance between the pleasure and frustration, you know.

It's too frustrating, you delete it. It's too easy, you delete it. You have to create that balance that plays with your head. And it's really hard. It's really hard and you need quality data to track it properly. So we do a lot of that stuff, you know. Like game economy analysis and progression system analysis.

Peggy: And data will tell us that. We don't want an easy win but it can't be too hard. And then of course there are player behaviors and your data goes into that as well. You know some people are very competitive or they love to, you know, they love the thrill of this, you know. Some people want a challenge. Some people are not the ones who love a challenge and the data of course will tell you that. And asking the right questions will also direct you. So I want to wrap it up with a couple of final rapid fire questions.

Shifting to rapid fire questions here we go.

Metaverse - what do you think of? What's a must do? What's a no-go?

Elad: It's here to stay. That's for sure. It's a new niche. I don't really treat it, personally, as like a revolution, because anyone who played World of Warcraft 20 years ago will tell you that was an amazing metaverse. That was an amazing era. You would socialize with people, you would travel. People would play for like they stayed awake all night like 10/12 hours just meeting other people and connecting with them. I think that eventually the dust will settle and we will be back to quality content because quality content is the thing that wins.

Peggy: Most eye-watering opportunity on the horizon that gaming companies should know or pursue.

Elad: There are actually a couple of things that we can speak about here. First of all games in let's say you call it the Roblox metaverse, because the Web3 metaverse is a different metaverse than the Roblox metaverse. Roblox metaverse, it's kind of a new but not so new platform. So discovery is easy. It's not saturated like mobile so you know marketing, user acquisition is rather cheap. It's easy to launch a game. It's easy to develop a game and the marketing is rather cost effective so you can get like traffic fast and scale really fast. So that's a really nice market that many companies are tapping into, I mean. And then on the other hand mobile, that will be fun. But mobile actually declined I think for the first time because people are you know trying other platforms. It's natural, they're shifting to, I don't know, Web3, Roblox, Minecraft, other platforms and even web, you know. And that's fine, and I think it would be nice because the ones that will prevail in that industry, the quality games, will enjoy great UA opportunities because you know when the big boys start getting out of platforms or invest less money marketing there's less competition on the bidding so UA goes down, like CPI goes down, it's natural.

Peggy: And Dive was one of the first companies to provide game analytics for Roblox games as well, right?

Elad: Correct, yes yes, that's correct.

Roblox is a beautiful challenge because there is no attribution. You're flying completely blind. All the marketing is done inside Roblox. There is no attribution. You cannot attribute the campaign. You cannot connect the user and the campaign. So we developed a lot of like workarounds to work around that and calculate the lifetime value in Roblox and we reached the point that within like just a few days we can predict the lifetime value of a player and it's really easy to then refocus your marketing campaigns to make them ROI positive.

Peggy: So I'm glad you're bringing up the topic of work around. That's my last question for you. Because here we are, you know. We are halfway through and more of 2022 and want to know what's up ahead, you know. Where will game companies struggle? Where will they succeed?

Elad: I'm a big fan of quality content and I think that they will prevail over any platform.

Nintendo Switch, Playstation, you build a quality game, you build quality content, and people will follow. And then if you re-engage with them with more content or follow up... I mean you release tomorrow a season of Breaking Bad and everyone will watch it.

It's because it was just like a super cool idea, you know. Or like you revive Lost for another season... this is like stuff that you know the quality was just... the quality is just great. There's a good book, there's a good quality, people follow it.

Peggy: We have started a great conversation. I'd love to continue it and actually offline I will. I will indeed, because you have some insights here that are gold, around motivation, behavior, and just applying data to up your game, literally. But how can our audience connect with you? Continue the conversation, maybe follow you somewhere. What's the best way, Elad?

Elad: Yeah, so my LinkedIn is pretty easy to find - Elad Levy and then my email is elad at dive dot games. I still do the demos to this day. I love it. I enjoy it. I love speaking with... I sometimes find myself in GDC speaking with some garage indie game studio about data and I really enjoy it. It's coming from a different place. Like more of a passion and less of like let's make more money, you know.

Peggy: Yeah, I can hear the passion. That's why I really appreciate it, you take time, you told us, you gave us a story. I mean it's not just the top tips, this is strategy here. This was solid. I want to thank you so much for being on the show and sharing.

Elad: Thank you, Peggy. Great questions. I love your vibe.

Peggy: And of course, if you are a UA marketer or someone like Elad who helps gaming companies literally up their game, then hook up with me or Pollen VC on Twitter or LinkedIn and we will set you up with a show of your own. So until then, take care.

V/O: Stay up to date with our show schedule by signing up for alerts at pollen.vc. That's where you'll also find our suite of financial modeling tools to help you plan and manage your business growth.