Tabarak Paracha
Today, we will be discussing the success story of HyperBeard, a hyper-casual mobile game publishing studio, and how Tenjin is helping them with multiple use-cases. To discuss this success story, we invited the co-founder and CEO of Hyperbeard, Alex Kozachenko.
For a summary of the case study, you can download the PDF below. If you’d like to dive deeper into what Alex shared with us, feel free to keep reading.
Alex, would you like to introduce yourself?
“Sure. My name is Alex Kozachenko. I’m the CEO of HyperBeard. We’re the largest mobile game developer and publisher in Mexico, with over 210 million downloads on Apple AppStore and Google PlayStore. Some of our great titles include Campfire Cat Cafe (launched May 2023), Adorable Home e Pocket Love. We specialize in cute casual idle games targeting a young, adult, female demographic.
Personally, I specialize in business, product and marketing strategy. But, you know, to run a company you need to do a little bit of everything.”
Could you tell us a little more about HyperBeard, how you guys started, and with what kind of games?
“We started in around 2016. Our first big game was called KleptoCats, which went viral. That was our introduction to both; a successful game, and also one that we self-published. That taught us a lot. We didn’t really start off saying we will make cute casual games for a female audience.It just kind of snowballed into that as we got good in that space, and at targeting that particular audience. We kind of slowly went into that direction. In 2018, we started publishing as well which brought in developers with cute casual titles over to us. So this created a business model that we are currently using today.
All in, we have published 30+ titles in the last 10 years. Almost all have been cute or cute adjacent. Our biggest game, Adorable Home, got 52 million downloads. Our second biggest game, Pocket Love, that was released last year (2022) got 25 million+ downloads in the first year.
We are about 65 people globally, mostly in Mexico. We also have people scattered in South America, Europe, and East Asia as well.”
What marketing or product analytics tools have you guys used?
“We like to keep things fairly simple. We used to use Unity (UGS) Analytics, but as most people know they just changed their pricing so now we switched to Firebase Google Analytics. We are not doing a ton, mostly looking at Day 1, Day 7, and Day 30 Retention, Sessions per user and then setting up funnels that are big on pushing people into the game to get strong Day 1 retention, and the precursor to that is the onboarding experience.
As our MMP, we use Tenjin – they are our primary one. And we also use Singular here and there. We also use ironSource LevelPlay as our ad mediator. At various points in the past, we have built pieces of internal BI, but to be honest, deep engineering is not really our strength. We’re much more about content, single player experiences, etc. And so, one nice thing about Tenjin is that they have been able to supplement our internal efforts and our BI is basically the Crescimento FullStack product that they offer.”
Speaking of Tenjin and Growth FullStack, are there any specific features that have been most valuable for HyperBeard?
“We love using Cofre de dados. Historically, we were just using whatever was available on the dashboard with whatever MMP we were using. But obviously, that’s pretty limited. The dashboard for Tenjin is great, it provides a lot of insights, but obviously you also want to expand beyond that. So, DataVault allows us to do that. It’s great for creative performance tracking, and also one thing that we do is a portfolio LTV analysis. We’re what you call a boutique publisher. There used to be a lot of these, now there aren’t so many. Now mostly publishers just look for really high LTV titles that they can get lower CPI’s on and buy at scale. For us, by staying within the same kind of themes, we’ve been able to drive a ton of organics year over year. So we’re what is known as a ‘boutique publisher,’ we specialize in a particular type of game, for a particular type of audience. Because of that we have a lot of users between our games and we really struggled to measure the financial impact of that. We know what happens, but in terms of what percentage or what kind of value – we had no clue. So, thanks to Tenjin, we’re now garnering insights that we can extrapolate for the entire user base and factor into our UA strategies, which allows us to afford higher CPI’s and grow at scale.”
You use aggregate-based LTV in Tenjin, which is a unique metric that we have. I wanted to understand how you think this is different from the solutions that are provided by others in the industry?
“It was kind of interesting when we were switching over from another MMP. Obviously, user-level tracking is kind of what people tend to think is the gold standard, but the problem with user-level tracking is that the entire ecosystem is moving to a more privacy-centric regulatory environment. And you get less and less data. And the users that you are actually getting data on are not exactly representative of the entire populace. And so you have these holes in your data. You’re not sure if it is measuring accurately, what it is measuring, or what the holes are, or how much is missing. There’s no real source of truth to tie it back to because those are the aggregate numbers you see. The amount you get paid on a daily basis, or the number of users you see that are actually downloading the game – these are your true sources of truth. And so, for us it was really frustrating because not only were we dealing with losing user-level tracking for users who are not giving their IDFA, which is a large percentage of users on iOS. But also, all of our games are for a mixed audience, which means they’re perfectly safe to play for a children audience, which we define as 15 and under. We have an age gate in front of all of our games, but the problem is once you do that you have no tracking on them at all. They are not able to provide any personally identifiable information without parental consent. So with aggregate-based LTV, we can extrapolate what the revenue is across all users and tie it back to a real source of truth. So we have a lot more confidence in the data itself.”
How have your KPI’s changed since you started using Tenjin and Growth FullStack for a while?
“One thing we have been wanting to do is expand to other geographies. We know that there is a demand for our games in other geos, that are relatively low LTV such as Brazil, and the rest of South America, and also countries in South East Asia, such as Vietnam and Philippines. These are areas that we do want to eventually target with paid growth but historically it’s been very difficult because the pricing models that a lot of other MMPs offer are based on your downloads. So you’re not going to pay 3 cents a download for someone in Vietnam when their LTV is probably around 3 cents. So, that’s one thing that we’re looking forward to doing with Tenjin. We haven’t rolled this out yet, but we have been able to increase our ad spend by approximately 300% over the last quarter which has been awesome. This has been mostly in our games Pocket Love, Adorable Home and to some degree Campfire Cat Cafe. We have also been able to see our ROAS pick up 15 – 20%, which is also how we were able to increase our ad spend.”
You mentioned earlier that you love using DataVault, which is our data warehouse as a service. Can you explain a little more in detail about how you guys have been using it?
“We use DataVault to pull all of the information into Growth FullStack so we can analyze it and then look at our LTV across all of our apps, both Android and iOS, and then also aggregate between the two. We also look at this data at a campaign level. That allows us to understand where users are coming from in our portfolio as a whole. Are they coming in from Pocket Love and going to Adorable Home, and then going somewhere else? Or are they coming in through one of our smaller games? Some of the games we have that we really don’t prioritize but they can be a channel for users to move around in our portfolio. And that gives us insights into where we should be spending our UA efforts, how much we can spend on users in every geo, and it’s just super helpful. That’s what we’re really focused on with DataVault – this portfolio LTV UA approach. It’s something we’re still rolling out.
We just switched over to Tenjin this year, so we only have a limited amount of data. But obviously if we’re looking at retention across the portfolio, as opposed to retention within one particular game (most of our games are around 5% Day 30 retention – which is not great but also perfectly average in the space that we’re working in). But if we’re looking at Day 90 retention across the portfolio, if that number can be closer to 5% or 10% we can scale a lot more because we can just contract users across as they go.”
In tandem with DataVault, you’re also using Growth FullStack, which is our product for building automated data pipelines and custom dashboards. How does Growth FullStack come into all this?
“We don’t really have a ton of data engineers that know how to use Tableau and other tools. Growth FullStack has allowed us to visualize everything. To sort through tables and pull in the data that we need and then hone in to the specific areas that we want to analyze at the time. For now that is mostly the US, mostly tier 1 english speaking countries. Eventually, we want to look at where the opportunity for UA is the greatest as a percentage basis, because it may not be in the US. The US is our primary market, that is where we get a lot of revenue, but that’s also because the US has a lot of spending power, and also there’s a lot of people in the US. But we may find that in Malaysia, we have a portfolio LTV bump of 200% ROAS and that’s where we should really be spending a lot of time and effort. Because even though the users on a per capita basis aren’t as valuable, they are extremely more profitable. So those are the kind of insights we’re looking at, and Growth FullStack is able to let us visualize all of that.”
I believe that the LTV Portfolio, AKA the Cross-promo LTV, use case has been one of the biggest use cases we have been working with you on. Can you tell us a little more about how we have been able to solve this for you?
“The nice thing has been that Tenjin has been able to do custom deadwork to pull in all the data that goes into DataVault, and then pull it into Growth FullStack for us to help us visualize. Because, like I said, those aren’t resources that we historically have in-house. We worked with various people in the past, but it’s hard to piece all that together. So to have it in one place, and to have a company that is well-versed in the tool, and that can help visualize it is super important for us.
To provide some specific numbers, for example, right now we only have about 90 days of the Portfolio analysis. But what we’re seeing in some games, such as Adorable Home, is a 35% uplift. That means that if we are buying a user into Adorable Home, to break even they don’t need to hit a 100% ROAS, they actually need to hit 35% less than that because they are going to make up for that 35% across the portfolio as they move. So then, that allows us to scale. So instead of buying users at $1, we can buy them at $1.30, which unlocks a whole other level. We see that across all of our games. So we don’t need to be releasing games that are multi-million dollar hits. We can release games that additively add. And as long as they are bringing in new users, which all of our games have done in the past – then they are adding to that overall portfolio valley.”
Do you have any advice for other teams that are considering using Tenjin or Growth FullStack?
“I would definitely say that you want to have a clear idea of the challenges that you’re trying to overcome, before you jump in. Because effectively you can build anything. I would like to approach it in an iterative manner, like doing a biweekly analysis of what you have and then what new questions pop up, and if it makes sense to build that in. What I see with Tenjin is that they’ll work with you and help you build it as part of their Enterprise package. And so, that’s been super helpful for us. It’s really accelerated our ability to get this information and to start utilizing it. You definitely want to think before you move, but also it doesn’t have to be perfect. It’s not going to be perfect to start, you’re going to be iterating over time just like you do with every mobile game or product, and that’s totally okay. Get the insights, ask for more information and keep pushing.”
