Mobile Growth Hacking: How To Bring Down The Cost Of Installs And Get Value For Your Marketing Budget

A panel on  “Mobile Growth Hacking” took place at this year’s App Promotion Summit in Berlin– it was chaired by Johanna Brewer  Ceo & Co-Founder at Frestyl. The other participants that took part in the discussion are Keren Zemer ,Vp Business at Appsfire, Ville Heijari, Gm Europe at Playhaven, Chris Hanage, Gm Europe at Appflood, Cristina Constandache, Svp Sales & Marketing at Mobpartner.  Now we are able to share audio recordings of the panel discussion video as well as transcription of this lively discussion.
You can watch the video or listen to the audio recordings and podcast recording here:

Mobile Growth Hacking Video

Mobile Growth Hacking: How To Bring Down The Cost Of Installs And Get Value For Your Marketing Budget from App Promotion Summit on Vimeo.

Mobile Growth Hacking Audio/ Podcast

Mobile Growth Hacking Transcript

Johanna: All right guys, I know we’re already entering into some super lively discussions, but we want to take it and bring it all together. So, if you guys can try to wrap it up, that would be fantastic. So before we jump into our questions, I want to get the panel introduced because we have some different people on the stage that you’ve seen talking so far, but I think you guys know the drill.
So, if we will slowly quiet down, we will allow the panelists to introduce themselves. So, I’m going to let you take it away after I steamrolled over everybody. If you like to introduce yourself and especially talk a little bit about why you’re here on the panel here today, that’d be fantastic.
Cristina: Hi, I’m Cristina, I’m the SVP marketing for MobPartner. I’ve been attending quite a lot of these panels and discussions. One of the reasons that we tend to keep on participating is we touch on all the different subjects that have been discussed today. Because we tend to work with advertisers that do everything from Facebook to re-targeting to looking for acquisitions for games, so we work with all of these different sources of traffic, so we try to present a certain type of performance.
Chris: My name is Chris Hanage, I’m the European general manager for Appflood. Appflood is a gorgeous ad network which is incredibly transparent with fantastic payouts and that’s all I really want to say now.
Keren: Hi everybody, I’m Keren Zemer from Appsfire. It’s a 360 growth engine for app developers  basically we cover all angles from user acquisition, discovery, monetization, and engagement. We’re here to show you some of the experiences that we’ve had as an app, Appsfire with over ten million downloads and how we got there and how we took the practices that help us grow and we implemented them into tools that we currently give to developers for free for monetization, for engagement, like I said, I’m here to share this experience with you.
Ville: My name is Ville Heijari I’m the European general manager for Playhaven and Playhaven is a business engine for mobile developers. So basically what you would use, create your game engine so that you are able to focus on creativity and game design, game development or use a third-party engine like unity. We provide a business engine which you can then plug in to do cover, cover the end to end cycle of acquisition, engagement, retention and monetization.
Johanna: Great, okay, so if you guys are ready, I would love to get some people up from the audience and asking some questions. Otherwise, I have some prepared. Can I get any brave souls up here to go first? Scanning, scanning. There we go. Okay. Who do we have?
Jen: My name is Jen Hafenkam, I’m from Eventum and my question is, how can I figure out the reason why app users uninstall our app. Because we have a lot of them.
Johanna: This one is going to be great for the panel, so. After the fact those advertisements of how we can get people to install the app, the question is how do we figure out why people are uninstalling and are you working on that?
Chris: Yeah, I can try to take that. One other themes that have been really common this morning has been billed really, really good stuff. I’m not saying for a second that’s not the case. I think that’s where we have to start. The second, as the industry grows up is just measure, measure, measure and measure again. I think that’s the other thing to add.
So what I would be doing would be to figure out when people are uninstalling. Is there a particular event, is there a particular bottleneck. Perhaps it’s when you’re asking users for money. Is there a particular user experience or something that means for people this isn’t for me. So really, just trying to identify whether or not there’s a something there that’s causing it.  I think that’s probably my initial thoughts.
Villa: I can’t stress enough the importance of quality in the first time user experience anywhere from 60 to 85 percent of the people who download apps won’t launch them second time. So it’s absolutely crucial to capture the emotion on that first use.
Keren: is what are you telling your users the first place to make them to install your app and what happen afterwards, and why exactly are they uninstalling it. Don’t forget all the time to improve, improve, improve your app. Don’t just say okay got whatever one million users. You need to constantly improve, bring additional features and then be in communication with your users. So create a dialogue with them, remind them that you are there, that there is actually a person behind it,  that it’s not just a company. Collect the feedback and make them engage and active.
Johanna: So, anybody else for the next question? I think this is pretty interesting what we’re talking about, how do we actually engage and keep going and look at users from the first moment they enter to say OK, why did you leave, even if you abandon me.
Millena: Millena Frachevo from Fragen Games we were talking about the acquisition costs and ECPA, what exactly, after how many months can I expect positive return on investment? Because, for example, for the browser we have, we develop browser games, we defined fixed CPA and we have prediction written on investment. For example, six months for some of our products and we don’t have any experience about return on investment for mobile marketing products. Could you recommend me, for example, when we should stop investing money. If I buy a user two dollars, how many months can I expect that this user spent money in the game and a positive return on investment?
Johanna: This is a pretty interesting question, I think. We’re thinking we should be calculated some kind of lifetime value for the customer in order to determine the ROI and actually press out the ads we’re spending. But how does that change over time and when do I say stop and when do I say okay? If it’s a lifetime value, do we wait until they die to make a change? Probably not.
Chris: I think we could actually weave into that. Are these free-to-play games? I think there are whole heaps of studies done in terms of the different breakdowns of play versus free to play when people are grinding away, when people become paid users, etcetera. I think the really concise and totally knowledgeable answer to your question is we don’t know and the only people that can know by yourselves. I think one of the comments in one of the presentations  this morning is start small and test, is really one of the most important things, nobody knows apart from you because nobody has the data.
So I will be looking to make a test get some insight from the data you get back. Anything more than six months is probably pushing it and being quite optimistic. We’ve seen a number of very large public companies come to huge problems because they spent a lot of money based on lifetime values that just went through. So I wouldn’t be looking to extend it much into the future. Just test and get a figure from the data.
Villa: If you’re coming from the online space, from browser gaming space already, you have the existing brand, you have the users, the behavior patterns from there, not all but some will translate into the mobile space as well. I would argue that in mobile, you have a lower attention span, so forth. So you would need to be able to also make sure those people keep coming back to the game. I wouldn’t say that it equals, mobile equals shorter lifetime value than online browser games. But you definitely need to make provisions to have extra attention to bringing those people back to the game.
Johanna: Keren and I were talking about this before, and we’ve been talking about this on stage, doing covert analysis to figure this out, looking at groups of users rather than saying the lifetime value of any customer. Maybe you guys can talk about this, part of the issue is that it’s changing over time, and you really need to be looking at not your entire ad portfolio and user portfolio, but specific new groups as you interact and on-board them in different ways.
Keren: I think it’s really not a game of quantities, but what are the type of users that you bring and whether they bring value to you. And then you need to measure all of them. It’s not just about spending endless money, do it like Chris was saying, test every case, try different networks and see which audience fits the best to your game. Because there is no right or wrong answer here. Might be you thought the QS is the key market for you. Suddenly you realize that it doesn’t work there. Then start with Euro, less competitive, makes more sense for you.
Christina: Everybody is saying more or less the same thing you should test and test more, but given that you already have an existing user database online, it might be an idea to start testing those users as well. Because on that database you already have some patterns and you understand the behavior. Pick segment, think about what you’re trying achieve, and push the mobile games to them and see how they react. At least you have a starting point. It won’t be the most accurate data in the world, but you have a starting point and with more test that you do on new mobile users, you should be able to figure out where to go next.
Johanna: Okay, so we’ve got a question down here.
Ivan: Hello, I’m Ivan Goeich, Berlin Miligen. I was wondering if you had any insight to the relation of how big an app, specifically a game is and how likely it is that it will be uninstalled and also in the relation between the loading time for a first startup and how likely it is that it will be the last startup.
Johanna: This is pretty interesting. We talked about this in some of the other discussions, but do you guys have any hard data or intuitions about the size of the app and the likelihood to download? We all know that bigger is worse, but is there tipping point? And also about loading times, initial loading screens, especially with games,  that’s a big concern, especially for a lot of 3-D stuff they’re doing on platforms. But do you have any hard data or intuition that you want to share with us?
Chris: I think for many territories, anything that puts friction or time delay between user pressing a button and getting a rewarding experience is going to obviously going to have an effect on performance. On the one hand, quicker payout for the user is obviously important. However where I do see a very strong correlation is on a territory basis. There are some territories where people don’t have such easy access to Wi-Fi, where data is really, really expensive where we see really, really strong correlations.
I’m thinking specifically of China, Papaya whom I work for is a Chinese company they taken a number of Western games over to China and those which have been very, very large sometimes make it more interesting in terms of getting distribution. Data can be expensive, not everyone has access to Wi-Fi and therefore it can be; keep it small and deliver levels in packs further down the experience is probably the best way forward. In my experience it’s more a territory thing.
Ville: If you have a niche product, what are the expectations from the players. If I’m going to download a triple A massively multiplayer online strategy game, my expectation level is very different if I am downloading Tetris. I don’t want to play Tetris now, but I can wait two and a half minutes to get my hands on that thing that I really, really want because my friends play it too.
Keren: The thing is that the users may not know when you’re promoting it in the mass channel, the user doesn’t necessarily know that this is a large game 3-D whatever, that it’s going to take time, that you promote it to the masses, you’re missing this and not focusing, it’s a very big game. And then you’re loosing these guys, so like above 100 mega, it can be a problem.
Cristina: We tested, and it’s not even about 100 mega, anything over 30, 40, maximum 50 is an issue as well. But as Chris said, it really depends on the territory and how easy it is to access the Internet, the downloading time.
Villa: So the important lesson here is that if you’re making a 300 MB game don’t go cost-per-click.
Johanna: It seems managing expectations is a big point. If you’re downloading a premier AAA amazing game, you’re going to wait a little while longer. But that’s a lot about the conversation you’re having. Go through advertising and general marketing and positioning, what am I expecting, so how long am I willing to actually wait for it?
But one of the things I  wanted to follow-up on that I think can get a little practical, so how do you guys actually do the regioning stuff? Where are you looking for this data? Are you going on intuition are you doing marketing studies? How are you breaking down it down by country, because as we, the service I have, we’ve moved from different countries, here in Germany, what we’ve seen on the ground is no one is installing anything even less than 5 megs over 3G. Just talking to users directly, I won’t do it. Is there any good source for this data?  Or is it something you have to find yourself?
Cristina: I think it’s just a matter of testing. In our position, because we work with many players it’s just a matter of testing and learning from the experience and then learning from similar experiences with other partners and trying to share the data with them. If someone comes to you saying I want to promote my game in China, and it’s 100 MB, I say I’m sorry I cannot make it work. It’s a matter of managing their expectations as well.
Chris: I think you’ll see very quickly on the CPI basis. If you’re doing the CPI you will very quickly find your apps blocked by everybody because they know they don’t generate much  revenue.
Johanna: So do we have any other questions in the audience?
Andrea: Andrea from Digital Surgeries. I have a more meta question, to the perspective on business intelligence because what we learn now is that engagement is definitely the area we should concentrate on. But we have also learned that there are many ways to engage with the user; if it is social media, campaigns, analytics what we study. So what I see is that you should consolidate on this data, like you said and what is your perspective on where it’s going. Do you know best practice versions where you really analyze the data on a central place, consolidated. What is your perspective on that?
Johanna: So we have a lot of providers to choose from, we’ve seen up on the slides on the breakdowns of who to pick. I have a feeling the answer is going to be it’s complicated. But I think the question is how are we picking between the providers using and is there a best practice centralizing that information. So we could see a lot of this could get fragmented fast and in order to have it all in one place, what do you guys recommend for making that decision?
Chris: I think what is most important is to work out what to measure. And only you can do that. What is it that’s actually driving the revenue or the lifetime user value or whatever within your own app. Work out for five headline KPR values are and measure those. To some extent one can ignore many other things.
However I do think there is an opportunity there is an opportunity to pull together all these different analytics into one dashboard. If I look at the number of dashboards one has to check every morning just to get a picture of what’s going on, it’s a lot.
So you’re right, it is complicated. I think simplifying what one measures is a way around that at the moment.  But I wouldn’t be particularly surprised if we were all using one dashboard in the way Google analytics for desktop in three or four years time.
Villa: So this is the difference between essentially metrics and analytics. So some of us are still staring at metrics like these are my sessions, this is my DAU, this is the money that’s coming in. But we are not really looking at the behavior patterns and analyzing what is happening there behind the scenes. And what in our app, what the consumer behavior is and what is driving that behavior.
Keren: In terms of choosing the provider, because you’ve already covered the analytics part, I think it’s just a matter of working with a partner that first of all gives you access to several types of traffic, so that you can test several different types of traffic, where you can measure, where you have constant feedback, and where you can also test and start small, where the conditions to start spending millions for a test campaign is not going to work necessarily. So you have to work with somebody you can trust that allows you to take your time but test at your own pace and with whom you can share feedback that is then heard. It’s common sense but it’s what works best at the end of the day.
Cristina: I think one more angle to your question is actually where you do this engagement with your users. And since the app is a mobile app, from our experience we see the best to engage through the app, not necessarily to the Facebook page or on twitter. Once you do it on the app itself, so create a dialogue on your app like a feedback forum, or any other form of engagement and ask questions to your users, let them answer, then you can collect this information, you can analyze it and it truly helps you improve your app and create these dialogues with your users.
Johanna: I think what we’re going to hear a lot is, there Is no magic silver bullet for your metrics or analytics saying here’s what you should do. It’s really about how you interpret them. It’s about how you make the decisions. I think we have two really good recommendations here, one is keep it simple, reduce to very few things, five KPIs that you can compare across all these ad networks, across all the strategies that you’re using. It’s easy to say this one gives me this and that one gives me that, but trying to simplify is important and also creating a feedback loop to have that dialogue with your users. Because you going just gathering information, and not thinking about how to respond, you find yourself at a dead end.
Ville: And again establishing those challenges of being able to change things on the fly on the server side and so on. So that when you have your analytics and you’re trying out different things, you don’t have the issue like a new game build every day. So basically you can change things server-side and change different campaign parameters and try different user segments and try different things with different user cohorts. That’s super important. It’s going to save you a lot of work and trouble in the future.
Johanna: I think we’ve got a new mike out there
Tivo: Hello, I’m Tivo. I’ve got a question regarding retargeting. The presentation was about possibilities we had at the moment. My question is the following, how do you set up retargeting with dynamic creatives at the moment for at marketing?
Johanna: Do you want to elaborate a little more on that?
Tivo: This we have on the desktop we have cookies and I can send you the product, so the current solution if I’ve got this right is to integrate the SDK of a network that we have the ADA fade letter so that he can retarget. Is there a way to set a bridge between the tracking tool and the inventory of the network without integrating the network SDK?
Johann: Okay, so getting pretty specific. You’re pointing out something that is very cutting edge stuff I mean this is coming out right now, I don’t know if any of this panel is an expert in this stuff, but if not I’ll try to help field it. The question is how do we,  if we’re doing retargeting and I have my entire user base and that sometimes can be quite large, how do I match that up with my ad network without installing huge SDKs? How do I start performing this and potentially cross ad networks. Facebook is beginning to offer it, but if I wanted to do that across Facebook, Google and others, how do I manage it and keep that under control?
Cristina: I think it’s just a matter of if you install a third party solution, let’s take for example, we can set together what is your retargeting criteria. Let’s say the user entered the game three times in the last week or last month, so this third-party can count it, can give a direction that and can work according to that. So it doesn’t require an SDK specifically from the network, but if you have a third-party that provides this data, you can cross so it’s available.
Johanna: Thank you for that but we’re getting a little technical. Do we have any other tables. Here you go we’ve got one down here.
Daria: My name is Daria, I’m from Hundix, I just wanted to hear your opinion on the current key trends of the mobile advertising, I’ve heard a lot about retargeting, like last year I’ve heard about geo location, vita ads, so what you think are the key trends now?
Johanna: I’ll make it even nerdier, what is the coolest thing that is going on in mobile advertising for you guys?
Ville: I think some of the coolest stuff right now is about introducing new surprising things, and I’m seeing a couple of player books in mobile, so you get an interstitial ad and you can actually play a level of the game that is being advertised which then takes you to… Obviously that is highly engaging and complicated so you need big volumes and you need extra resources into creating the creative steps but that’s new, that’s fresh. Personally, I like video a lot as well.
Johanna: We were talking a lot about the video that has been going around, actually I think a lot of the new trends that are coming out are specifically about the technological innovation that is happening. I would not have imagined seeing interstitial video on a phone two years ago.
Chris: I’ve got an interesting one and a boring one. I’ll start with the boring one, I think were starting to see real transparency in the stream, I think we’re starting to see the lifting of black boxes of analytics. It’s really quite dull, but it’s fascinating in terms how it tracks back to ROI. I think for the first time we are beginning to be able to see how we’re spending money and what difference that’s making to our bottom line. Because at the end of the day we are all here because to make money.
I think the next thing which is much more fascinating is seeing what effect the increase in the data pipes of the mobile will have on the creators. You mentioned the video, 4G is obviously launching around the world. As 4G becomes the standard the sort of stuff one can have on a mobile is just going to be amazing.  And you mentioned having video, you mentioned having players, we haven’t begun to see where that is going to go. I think that’s just going to be gorgeous.
Cristina:  Take it one step further, imagine what is going to happen when cross device tracking kicks in. You’re going to be able to follow the user with different creators from one device to another to end the cycle. When that comes in, probably, hopefully.
Chris: And scary, because there is nowhere to hide.
Keren: I think the values are more dear, in terms of advertising. Before you know we used to use these banners, I think the industry is going towards being more respectful to the users, don’t put things that will cause accidental clicks and the advertiser will have to pay for that, don’t put ads that are intrusive to the user, don’t waste the, you design a beautiful app, and you have an ugly ad in there, so don’t do that. So this is something that’s important. Don’t waste inventory know exactly which breakout session you’re going to put the ad. Cap it so the user won’t see an add seven times in an hour. So all these elements are really, really critical and becoming more and more popular.
Cristina: That’s a very good point you made, everyone is talking about analytics and metrics and figures and numbers and formulas, but no one is talking about the impact of the creatives, and a call to action. And we seriously am, so I have to say that, lots of banners, we’ve downloaded free games now, which is only that all games are free and all the apps are free.
So the user, why would I download one game over another if there is not a call to action with regards to something happening. And it goes for everything. It goes for hotels, it goes for others, it goes for absolutely everything. Stop limiting yourself to the banner, you have to download the app.
Talk about something that is happening in app, talk about something that the user will only get by downloading your app and not going online and not going to move it aside, and not going somewhere else. If we manage to preach enough about this to all the advertisers, I think the ROI and the formulas will become a lot clearer and the results are going to be much, much better.
Johanna: To wrap up, what we’re seeing here today, especially from the first two sessions, it’s about quality, it’s about creativity. And we’re all talking about measuring but unless you’re putting significant value out there, there is so much competition. It’s not a land grab anymore, it’s creating quality dialogues with specific users to understand and really build, not just a valuable product now, but a valuable ongoing experience from the first time they hear about you until they finally join your service.
So I think we can all get to talking together about what is going to be happening over some lunch and what the most creative things that we are all saying. So I’d like to thank the panelists again, give them a big hand.
Thanks to the moderator and panellists for a great discussion you can find out more about App Promotion Summit here 

By signing up you agree to our privacy policy. You can opt out anytime.