ABOUT: LinkedIn  | Twitter | Supersonic


ABOUT ROCKETSHIP HQ: Website | LinkedIn  | Twitter | YouTube

KEY HIGHLIGHTS

🤔 Why hitting #1 on iTunes is not a good goal – and what is a better goal instead

📈 Given that iTunes has deemphasized the top charts, how impactful is a top rank on the app stores?

📉 How much has the k-factor associated with a top rank dropped in the current app store compared to the previous version of app stores?

🎮 How Supersonic decides which games to promote.

💰 How Supersonic runs early marketability tests and soft launches to validate a game’s long term profitability.

🚀 When Supersonic launch a game, the key risk is that they may be paying too much for a top rank position.

👀 How Supersonic looks at the allocating installs between organics, paid and cross promo during a launch.

🧮 What metrics Supersonic’s teams are looking at during a launch – and what their launch cadence looks like.

🗺️ How Supersonic evaluate different geos during a launch.

🙋‍♂️ Do Android charts behave differently from iOS charts?

👨‍💻 How Supersonic’s UA teams are structured so as to drive post-launch growth.

KEY QUOTES

There is a formula for marketing the right title

If you pick the right title, and if you are making the right choices and correctly optimising both creatives and CPI, you should get a significant amount of installs and then find yourself at number one or two or three or even seven or whatever it gets. 

Quality is critical

Today, the main way to get a bigger portion of organics is the quality of the game, and less because of its place in the charts. I would say that today, the K factor purely because of the charts, wouldn’t be more than 5% or something similar. The rest, 20-25% or whatever you get, depends on the game. I assume it’s coming from virality and not from its place in the charts. It used to be 20, 30, 40, 50% from organics,  that’s not the case anymore. 

How to evaluate a game for potential success

We’re looking at the game itself for the retention, play time, game characteristics and potential to add rewarded video. And we estimate the potential LTV of the game. And then we see if the game can generate significant profits and scale. 

What to optimize for

What we actually do is we are not trying to estimate what will take us to the top 10. We’re trying to estimate and to make sure that we are not spending money that is not needed. So to answer that first of all, once we decide to go into global launch, we are optimising everything toward profit, not toward the placing the charts, maybe it will be number 40. Maybe it will be number one. 

If it is number one, my risk is that it could have been number one for a much better higher margin, and pay too much.

It is important to be responsive

Once we are optimising the UA on a daily basis, we are looking at day 0… many times day 1, day 3, but a lot of day 0 because if there is change in trends, we want to react as fast as possible. If monetization drops because X Y Z, we don’t want to wait until day 3 cohort and find out that we lost money for 4 days when we are already ready to react on day 0 once we see the day 0 was lower than what we expected.

Introducing the title growth manager

In Supersonic, we have a title growth manager, who is the owner of the title for both supply and demand, ad monetization and user acquisition. He is the full owner of the title, so his responsibility is to take this title that was picked, designed and improved in our gaming hub. 

The job of the title owner / growth manager is to take this title, which has the great KPIs and to make it a multi million dollar hit.

FULL TRANSCRIPT BELOW:

Shamanth: I’m very excited to welcome Nadav Ashkenazi to the Mobile User Acquisition Show. Nadav, excited to have you on the show. 

Nadav: Thanks so much for hosting me and I’m super excited to be here as well.

Shamanth: Indeed. Excited to have you because certainly you guys’ studio has really hit the top charts on iTunes lately. But I’ve also certainly worked with your team much before the games were on the horizon. And we are going to dive into and talk about your journey, your process in getting multiple games at the top 10 on iTunes today. You guys have had five games that hit number one on iTunes in the US. This is even though it’s been just a few months since the studio has launched, what would you say have been the key factors that contributed to this?

Nadav: Yeah, so maybe I’ll start and say that hitting the number one is not the goal for us. I’m more looking at it as a result. Hitting number one mainly means that you managed to get a significant amount of installs in probably a shorter time window in specific countries like the US. 

I think that the key for succeeding and scaling your titles and getting to number one – the result is in two different steps, the first step will be probably to pick the right title to launch. This is something that we do and focus a lot, along with the marketability potential of the title. And will we be able to scale this game significantly, of course profitably once it’s out. So this will be the first step. 

And the second side will be once the game is being published, and it’s in production: how sophisticated are you in both the creative and the CPI optimization, two important factors that impact the amount of installs that you get.

If you pick the right title, and if you are doing the right choices and optimise correctly both creative and CPI, I think that you should get a significant amount of installs and then find yourself number one or two or three or even seven or whatever it gets.

But I think this is the key.

Shamanth: Yeah, I like what you said about how being number one is not necessarily the goal. I also have perspective from four or five years ago when you could basically buy burst installs, you can hit number one, it would cost you a lot of money, but you could get that and a lot of games got burnt doing that. There were certainly horror stories of companies that essentially lost a lot of money on it. So I definitely like what you said about it, about just that being almost a by product. 

And certainly we’ll dive into some of the other aspects of what you said about picking the right title and making sure that it’s a foundation for launch. We’ll dive into that. But also, I would like to spend a little bit of time around just the top charts and the dynamics of how they work because iTunes has de-emphasized top ranks in favour of editorial content. And, in fact, I remember iTunes before 2017, or 2016, the top ranks are fairly clear and up front and centre. Now I have to just browse through, look through it. It’s not very clear and obvious as to where the top charts are. How impactful is the top 10 ranking or the top 1 ranking today compared to what it was before?

Nadav: Definitely lower. Let’s look at it from the organic impact, which is probably one of the most important aspects of topping the charts – it used to be a very big reason for growth in organics. And you said that lately in the last few years, Apple has the top 3 in a much less visible location. And today when I analyze my organic installs and my K factor, I actually don’t think that the place in the chart has a significant impact. 

For sure, it has some impact because for sure some of the users are finding these top three and maybe installing it accordingly. But definitely not as often as it used to be. And I look today at organics K factor, I look at it mostly from the virality perspective, meaning what we see is that the games that have good scores, that have good retention, that have good playtime, have a better K factor than others regardless of their place in the chart. It’s more of a friend bringing the friend or friend playing and saying, “look at this awesome game. why don’t you play it?” . Or they are sitting in class in school and he played, and all the other classmates see the game he played. I think that

today, the main ability to get a bigger portion of organics is mainly due to the quality of the game and less to the place in the charts.

I still believe that there is impact. When we climb in the charts, we see some impact on organics, but not something significant. It’s not the biggest motivation for organics.

Shamanth: Yeah. And to the extent you’re comfortable sharing, would you have a ballpark estimate of how much it has decreased compared to the previous app store? 

Nadav:

I would say that today, the K factor purely for the charts, won’t be more than 5% or some similar. The rest 20-25%, or whatever you get depends on the game. I assume it’s coming from virality and not from the place in the charts. It used to be 20, 30, 40, 50% from organics,  it’s not the case anymore. 

Shamanth: Yeah, those are certainly numbers I’ve seen, and I’ve certainly seen games that hit like the top three in the past hit like, a hundred thousand just organic installs. And while that is definitely not out of the question today and certainly it sounds like that those numbers have been significantly diminished now. You did mention that the top ranks are not an end in itself, because eventually you’re looking at the long term marketability, the long term profitability of the game itself. You briefly mentioned that you evaluate the game before the launch. What does that evaluation process look like? What does the soft launch process look like? And you said picking the right game is important. How do you decide which game you push into the top 10?

Nadav: Yeah, so I would say our sorting process has a few stages. Their very first initial stage is just intuition. We are looking at the concept – probably it’s not even a game yet. Or maybe it’s already a ready game or maybe only a prototype will be only a video or maybe only an idea on a dock. But the first filtering is being done manually by intuition of our publishing managers or game designers. And it’s mainly according to first market trends. So we’re pretty much – and again, we have the experience of being in the industry for 10 years and we have a pretty broad experience overview definitely now when we are testing so many games a day, what works better and what works less. 

And many, many times we are able to probably kill or pivot an idea even before spending $1 just according to the look, to the feel, to the alignment with the current market trends, etc. This would be probably the first sorting, we’re sorting probably half of the ideas pitched to us, and picking the right half that in our eyes has much better probability to succeed and scale according to market trends and what we see in the market. 

The next step, what we’re doing is we usually start a Facebook test. And just to feel both the CPI from one end, and the retention and play time and other potential motivation factors on the other end. What we’re trying to do is we’re looking at the overall picture – we understand how marketable the game is, and it’s an early indication, because we’re looking on Facebook there’s a limited aspect where we’re understanding how marketable the game is. 

We’re using our own algorithm to convert these results from the initial test into a potential future CPI on a scale that we believe we can achieve. And on the other hand,

we’re looking at the game itself for the retention and the play time and game characteristics and potential to add rewarded video. And we are kind of estimating the potential LTV of the game. And then we are seeing if  ever, the game that in our eyes can generate significant profits and scale. 

If yes, we are moving to the next phase, which is the soft launch. And in the soft launch, we’re actually converting assumptions into facts. We don’t need to assume what will be LTV anymore, we will know now what will be the LTV. We don’t need to assume marketability based on one test we are doing, we’re probably going to soft launch a few different channels and to look at a few different creatives and much more data. 

The soft launch, it’s kind of the final validation for us that our estimation based on the previous stages was right. And this game actually can be profitable in scale. If it is, we are moving into a full global launch worldwide, Android as well, etc. And our goal of course, is not to be surprised in the soft launch. My conversion rate that I’m aiming to have is 100%. Because every game that they move to soft launch and eventually found out, it’s LTV or CPI are not proper like I expected, meaning that my analysis wasn’t solid enough, I need to rethink about my analysis, because it’s a waste of time, focus, money, not only for me also for the developers that we’re working with, if it’s an external publisher of the app, or vice versa, and we want to be as accurate as possible, as early as possible.

Shamanth: So you are iterating on the game, you’re looking at the metrics, and you’re gradually proving out – actually, you’re running ads to the game to see what the metrics are, to see if this can be profitable at scale. And once you pass that soft launch phase, you realise, okay, we do want to go big. How do you estimate what sort of download volume you need to hit going out of the gate to basically hit the top 10 or top three? How do you come up on those estimates?

Nadav:

What we actually do is we are not trying to estimate what will take us to the top 10. We’re trying to estimate and to make sure that we are not spending money that is not needed. So to answer that first of all, once we decide to go into global launch, we are optimising everything toward profit, not toward the placing the charts, maybe it will be number 40. Maybe it will be number one. If it is number one, my risk is that it could have been number one for a much better higher margin, and pay too much.

So what we usually do is first of all, after soft launch, we analyse the data, we understand the ARPU and understand the CPI, we understand the potential. Then we try to understand, okay, let’s assume that we are going live uncapped with this CPI. We don’t want to get more than 100k installs a day because if you’re already the top one, you’re already maximized, a lower CPI still will put you in the top of the charts, you’ll still have a high scale but better margins. So we try to understand whether it’s going to be too much or not. If we don’t think it’s going to be too much. We’ll continue ahead as planned, optimising towards our target margin. And if you find yourself getting too many installs we are lowering the CPI even further. If not, we’re keeping it as is.

Shamanth: Yeah. And I think that’s an important consideration that you could just get too many installs, you waste your money. I did some work on Words With Friends at one point and I was talking to the team. Words With Friends was in the top three for 45 days. And one reason they were able to do that was basically the process you talked about. They were like, how do we make sure we don’t overspend? Because they were like, “look, we have so much cross promotion that we can go on. And if we just send 500,000 users today, maybe the top one, but we don’t need 500,000. We just need like, hundred thousand – we just need the right amount.” 

I think that’s so important to note. And like I said, Words With Friends they were like, top three for like 45 days because of that. They could keep that momentum for much longer. Once you decide, okay, this is how many installs I want to maximise my profit – that can come from paid user acquisition, can come from cross promo, can come from organics, you said you know, it’s far diminished, but it’s still there. How do you figure out – how does this breakup happen among these 3 sources?

Nadav: Yeah, so first of all, already in the soft launch we try to understand… based on other games, we try to understand the K factor and organic installs we are getting… and what should we aim for in terms of paid installs. And cross promo, we look at it as part of the paid installs. Regardless of if you’re paying money for this or not, it goes through their regular paid ad units for paid campaigns. And so what we are doing is, we understand, okay, we want to get and we believe that 100K is definitely doable for this game. And let’s say we’ll have 20, 25K coming from organic 75K… and then we understand, okay, we always start with the strongest channel, both on social and SDK networks. We usually use the soft launch period in order also to optimise the creative per channel. Then when we go into the global launch, we’ll probably already have the right bid, the right creative, set up for all UA channels, so we can get to the very calculated goal we put in and in hitting the margin goal, which is the most important challenging thing to do in this phase.

Shamanth: Right? And when you say hitting the margin goal, where you’re getting like 100K installs a day, you only have D1 or maybe D3 data. So, are you having D1, D3 targets based on your soft launch? How are we looking at that?

Nadav: So in the soft launch, we actually accumulate at least day 7. Okay, some of them day 14 or more. And in hyper casual, specifically, day 7 and day 14 gives you quite a good indication of your LTV and it will be a bit better or a bit worse, but once you understand that you are stable data for day 7 let say you have some good understanding of your day 30. But yes, it’s a guess. And we try to do it as accurately as possible based on our data. And then what we are doing is

once we are optimising the UA on a daily basis, we are looking at day 0… many times day 1, day 3, but a lot of day 0 because if there is changae in trends, we want to react as fast as possible. If monetization drops because X Y Z, we don’t want to wait until day 3 cohort and find out that we lost money for 4 days when we are already ready to react on day 0 once we see the day 0 was lower than what we expected.

Shamanth: Yeah. And you know, when you say you do need to react quickly, I imagine that’s even more important when you’re at the top of the charts are getting hundreds of thousands of users. What does the cadence of check ins look like at that point of time? Are you like meeting every three hours, every six hours? What is the check in process at the time? 

Nadav: So I would say it’s really not 24/7 but yeah, 18×7 we have a real time system. We are connected and we have alerts. And we’re looking carefully for changing trends. And probably we will catch it, hopefully on the same day it happens… Sometimes the day after because on the same day, it’s hard to notice it. Because in the middle of the day, it’s hard to understand the full day 0 ARPU that you will have. The morning after, when you have the full day complete, you can easily see the differences that you had from day 0 yesterday compared to other days. And if we see something that doesn’t look good, we sometimes do the acting in the morning after and then you know, it can waste one day from unexpected activity.

Shamanth: Right. And how do you assess “Oh, if day 0 is good, how do you assess the duration for which you should push a game in the charts during the launch phase?

Nadav: So well actually from day 0 until the day we stop the UA, we are always optimising for the margin. If the game will stay up in the chart, say if we take Sort It 3D, the game that we launch in the beginning of February as an example, already three or four months in the top 50 in the US. This was not the goal for us, we’re optimising this UA just like working on anything else. If the game is marketable, and we are doing a good job on the optimization of the CPI and UA and creative, it will stay at the top of the charts for a very long while.

Shamanth: Sure. Makes sense. And you know when we say charting, typically we are referring to the US iTunes charts. How do you think about some of the other geos? Do you just have separate planning or strategy for different geos? Do you have geos by tiers? How do you think about that?

Nadav: So there are a few geos that we are looking at independently. US of course, China, which is a very big market, Japan, which is a very unique and high potential market, and I would say the UK is the same. And these will be countries that will typically look at the actuals per country. Then we have the other tier who, let’s call it a tier-1 group of countries from Australia to New Zealand to Germany, to Korea and many, many, many other places. When we’re optimising according to the tier 1, we do it a bit differently than we have the tier 2, tier 3, but we always do it globally and we put as much attention at the potential of the country for us in terms of revenue and profit. 

Shamanth: Sure. That makes sense. And do you notice that Android charts behave very differently from iTunes when you’re launching a game?

Nadav: Not as much. All of our games got to the top of iOS, got to the top of Android as well. Some games again, there is different behaviour between Android and iOS. Some games are going to be very, very successful and Android and only a bit successful on iOS or vice versa. Mainly because the type of users is different, and the type of execution. Sometimes games that work very, very well on iPhone won’t work well, or as well, for a device and maybe the game is a better fit for iPhone, or less fit for Android. So there are definitely differences between Android and iOS per game, but I would say that in general, which reached the Android chart, the iOS chart the same we are optimising both of them the same.

Shamanth: Right. After you guys have launched, you guys decide to go big, and you’re still optimising for profits and margins at that stage. Let’s assume once an app is out of the launch phase, and you’re like, maybe, we don’t need to push that hard. You don’t need to get to 100K installs every day. It’s maybe a couple of weeks, in what you might call the post launch phase. What are the key UA strategies you look at for your games?

Nadav: Yeah, so one of the interesting things that we’re doing in the post launch phase first of all,

in Supersonic, we have a title growth manager, which is the owner of the title that is owning both supply and demand, both ad monetization and user acquisition. He is the full owner of the title, so his responsibility is to take this title that was picked and designed and improved in our gaming hub and will serve out of many that were tested saying “hey, this title has the potential to be a big hit.” 

The job of the title owner / growth manager is to take this title, which has the great KPIs and to make it a multi million dollar hit.

He will be owning everything related to the UA and monetization of his title with a team that supports him of course – from analysts to operations, etc. But what we will do is, and I think that this is what is nice about it being the same person because he has the full picture…. from one and you can see the LTV performance of the title, he can see that a CPM, the impression per user, per ad unit. He can affect the waterfall, he can discuss with different ad networks from Facebook to AdMob to Unity etc. how to optimise and squeeze the lemon on the LTV side. 

On the other hand, is also managing all the UA side and on social networks, SDK networks, and optimising the bids and the creative for each channel. I think the fact that it’s the same person owning the full business operations of this title is a very, very strong ability for us because he can close the loop much faster and be much more efficient. So if he sees that by improve something on the waterfall, the ARPDAU and eCPMs climb, he knows automatically how to increase their CPI to keep the margin and to get even more scale, which we can, thanks to the higher LTV or vice versa. The fact that you can close the loop so quickly is one of our key elements for managing several hits in the top of the charts.

Shamanth: Just also to dig in. You said there’s one person in charge of UA and monetization? Is this person managing all the channels or are the channels divided between a couple of people? How does it work?

Nadav: He is the owner of all of the title. He has operations that he can allocate. He can say, “ok I need your support from the operational perspective on the social and your support on the SDK, and it can really differ between titles, because some titles are stronger there, others are stronger there. But he’s the owner of everything – all the UA channels, every dollar spent on this title, and by this title is under his hat. And he can allocate the resources accordingly between the team, according to the potential of the title and how well it works in each channel.

Shamanth: Yeah. And that seems like such a well oiled machine for not just for scaling the game that’s proven but everything you talked about is so impressive to me, from the ideation stage to proving it out to soft launch to going big, and of course, continuing to scale. You know, this is definitely one of the things I find very, very impressive about how you guys have executed and it’s certainly not surprising that you guys have hit the top ranks for five of your games in just a few months. Nadav, this has been incredibly instructive. Thank you for pulling back the curtain behind the hit games that you guys make. It’s been a pleasure having you on the show. To wrap could you tell our listeners how they can find out more about you and the games that you guys make.

Nadav: Yeah, so they can login to our website supersonic.com. They can contact us if they want to work with us and to get our support of publishing their games, feedback on the game, UA and everything. We are here, we are planning to keep going. We launched another game yesterday and we have another five launches for the next month. We expect all of them to top the charts because all of them have the right promise and KPIs. And we are always happy to work with talented game studios and we believe that Supersonic has a lot to offer. So we are here, you can contact us via our website. And we are looking forward to a great 2020 full of great games.

Shamanth: Excellent. Yeah, exciting times for you guys. I will look out for those five games. And we will certainly link to your website and how other game developers might want to contact you guys. We will link to all of that in the show notes. But for now, it’s been an honour having you. Thank you so much Nadav for being on the Mobile User Acquisition Show.

Nadav: Thank you very much. Thanks for having me.

A REQUEST BEFORE YOU GO

I have a very important favor to ask, which as those of you who know me know I don’t do often. If you get any pleasure or inspiration from this episode, could you PLEASE leave a review on your favorite podcasting platform – be it iTunes, Overcast, Spotify or wherever you get your podcast fix. This podcast is very much a labor of love – and each episode takes many many hours to put together. When you write a review, it will not only be a great deal of encouragement to us, but it will also support getting the word out about the Mobile User Acquisition Show.

Constructive criticism and suggestions for improvement are welcome, whether on podcasting platforms – or by email to shamanth at rocketshiphq.com. We read all reviews & I want to make this podcast better.

Thank you – and I look forward to seeing you with the next episode!