Game designers may not care about it, but acquiring users is still one of the most difficult tasks in launching a free-to-play mobile game. The problem is that a new game will compete with 132,000 other active titles on Apple’s iTunes App Store. Advertising can help it stand out, but as ad costs rise, the risks are very real that a company may pay more to get new users than it can generate a return on.
If a company pays $3 each to get 100 users, it would be outstanding if 10 percent of them convert into paying players. To get a return on the advertising outlay, those players have to generate $30 over the lifetime of the game (a stat known as lifetime value). It can’t take forever to get those users, either.
Sponsored by VB
In the real world, the problem is that some companies are paying $8 to acquire a user. And mobile marketing firm Fiksu says that the cost of user acquisition rose 21 percent from November to December. At the Casual Connect Europe event, I moderated a session about the tips and tricks of user acquisition.
It’s no wonder that some companies are searching for alternatives to getting their games discovered. Some of those alternatives are shifty. Tapjoy found that out when Apple cracked down on its incentivized installs in 2011, and many turned to Android as a result. Last year, as Gree entered the U.S. market and bid up the cost of user acquisition dramatically, developers longed to be featured. Gree was accustomed to paying much more — $15 a user and advertising on TV — to get lucrative Japanese players. But in the rest of the world, consumers aren’t yet as crazy about paying for games. Alternatives have to be found, even for the likes of Gree. But the pressure on costs is rising as more brands move into the market without worrying about user acquisition costs.
Our panelists included Jussi Laakkonen (pictured far right), the chief executive of cross-promotion firm Applifier; Stefan Bielau (second from right), a freelance mobile consultant; Erlend Christoffersen (third from right), the head of user acquisition at mobile gamemaker Supercell; Eric Seufert (pictured third from left), who’s in charge of marketing and user acquisition at Helsinki’s Grey Area Labs, the publisher of Shadow Cities; Gilad Rotem (pictured second from left), the head of sales and product for InGaming; and Billy Shipp (pictured far left), the vice president of growth at Iddiction, the creator of the App-o-Day promotion platform.
Here’s an edited transcript of our conversation. They brought home the point that everyone should think about solving the tough problem of user acquisition.
Takahashi: Stefan, can you talk a bit more about some of the history here in user acquisition? What has worked in the past, and what is working now?
Bielau: The old craze was around incentivized installs — pay-per-click spends on mobile advertisement. Nowadays, the market has opened up a few more opportunities. Facebook came around with the App Center. You find a lot more performance-based models, breaking down your acquisition target to a cost per install, or deeper, to cost per acquisition. Those possibilities weren’t available two years ago. A lot has happened to open up the field for user acquisition. Stuff like mobile search engine optimization becomes more and more important due to the fact that the App Store is so crowded. That can still be triggered by some outside factors to improve your ranking position. I would say localization plays more and more into the marketing part. Before you saw games localization just amounting to translation work. Nowadays — you just gave the example of Japan — it’s important to know the local market when it comes to mobile app marketing.
Takahashi: Erlend, you’re fairly new here in user acquisition, with Supercell coming up with a big hit game, Clash of Clans. It almost seems like you guys wouldn’t need to spend money on user acquisition at all.
Christofferson: For Supercell, it’s all about the problem of building a great game. Only when you see that from your retention and engagement metrics that you’ve got in there — that’s when you can start thinking about marketing. We focus on creating engaged users. We believe an engaged user is more likely to tell his friends about our games, and more likely to monetize.
Takahashi: How do you go about choosing things like an ad network?
Christofferson: What’s important for us in choosing a network to work with is user experience and transparency and tracking. User experience and transparency go hand in hand. It’s important for us to know exactly where our games are promoted and how to share that through the user experience. Tracking is important: making sure that we have all the data we need and that it’s reliable. Preferably, we can see it broken down so we can then optimize a formula to find those most highly engaged users.
Takahashi: Jussi, why don’t you talk about paid versus non-paid user acquisition?
Laakkonen: The new way for games coverage, like everyone says, is a friend showing another friend a game. “Hey, you have to play this.” That’s the best way for a consumer to find a game. You want to have that happen, but you need to delve into paid acquisition, which is very different in the social and mobile markets.
The social market on Facebook exploded, all on the back of viral acquisition. Seeing what your friends play, whether it’s spam or whether it’s authentic. Facebook shut down spammy ways for acquiring users. Whereas [with] mobile, you started by taking out your phone and showing your friend.
I think mobile is actually relatively sophisticated nowadays in paid acquisition. There are people who can help you arrange your campaigns. There are supply-side platforms that help mediate those campaigns. The one thing we have not cracked yet in mobile is the social angle. Not the Facebook-style spam that we all hate, and we’re still kind of afraid of. Every time we add an app on Facebook, we’re like, “Will this game spam me?” It’s always a question. But that’s something that still remains to be cracked. Is there a way where I don’t need to pull out my phone, but I’m somehow able to share and compete with my friends? We’re working to revolutionize the way people do that.
Takahashi: Can you talk about Everyplay for us?
Laakkonen: We do videos of game replays that people can share with friends. Seeing the game, actually seeing what the game is, it inspires some emotion. “Hey, I’m having a great time with this. It’s a fantastic game.” What we’re trying to do with Everyplay is to replicate that moment. We record the gameplay as it happens into a video container. You don’t have to think about it. It just happens. Then they can share it. Like if they had a great fight in Clash of Clans or a great race in a racing game, they can share that in their networks. We see that when a share happens on Facebook, 12 percent of the people who see that replay, on average, immediately go and download the game. Twelve percent is a really good when you think about conversion rate.
Seufert: My approach to user acquisition probably comes from a different angle than the way most people do it, most studios. I’m only one guy on the team, and there’s no way we can keep on top of daily reports from every number, to see how many users we get or what kind of quality is there. What I try to do is focus on what I call the downpay market. If I’m trying to optimize user acquisition costs, all I’m doing is reducing expenses. There’s a limit to how much I can reduce those expenses.
But if I focus on increasing the value of those users, practically I have a limit as well. But the idea is that I can increase the value of a user much more easily, with much less work, through analytics when I have a small staff. What we try to do is focus on the proper limitation of freemium. Making sure that freemium economics work well. We focus on attention, and we focus on engagement. If we do that, we have models which can predict how much users will be worth. Then user acquisition is very simple because all you have to do is bid less than that and make money on that spread.
You can take the opposite approach if you have a large team. It’s very rewarding if you can study every single day — this quality of users came from wherever. Then you can adjust your spending to those channels. My belief is that since all those channels basically tap into the same pool of users, it doesn’t provide as much value.
Bielau: What do you think about outsourcing user acquisition as a competence? Using a service provider to manage those campaigns for you. You mentioned one problem being the lack of resources. Do you think there is room for a service provider like Fiksu to manage campaigns for you and charge 10 or 20 percent? Is that something that, as a free-to-play game provider, you should even consider, or is that giving away one of the key things you should be doing yourself?
Seufert: Yeah. Another one of the key things you should be doing, which is very large, is managing the channels. What you should be doing is managing the freemium implementation of your game, to maximize the lifetime value. I can easily foresee a small developer not wanting to invest so much money and having a strong analytic system already. But if you didn’t have that, I can see it making a lot of sense to outsource that to another company to run it.
Takahashi: Pulling back a little bit, this trend of big companies moving into the market and buying as much of the market as they can is a pattern that’s going to continue. Big brands are coming into mobile. Should that be scary to all the developers out there because they have infinite budgets for user acquisition and don’t even care if they’re losing money on a per-user basis?
Bielau: Are you referring to our Japanese overlords? [Laughter]
Takahashi: Some of them. But even someone like Disney. Maybe they get so much benefit out of the value of their brands — which is a form of non-paid user acquisition — that they can spend far more than they do on mobile user acquisition. How do you compete against that?
Laakkonen: Well, I think that’s an overplayed hand. One of the biggest publishers in the world is Activision, and they just started. Looking at Disney, mobile has been really interesting for them because it’s bigger than you would expect. I think the big players are actually leaving a lot of money on the table. In the end, spending more than you make is not a strategy unless you’re playing for an exit. If you’re in a situation where you can lift a company and sell it rapidly, that’s a strategy.
But we haven’t yet seen what happened on Facebook. Zynga paid a great amount of money to buy themselves domination of the marketplace and create this cross-promotion network that ensured their leadership for these two-and-a-half years. We have not seen that happen on mobile. Unless someone creates a similar constructive cross-promotion network that dwarfs any others, I don’t see companies regularly spending more than they can make. They’ll just count their money down. Those users will move onto other apps. They’re throwing their money away.
Bielau: Maybe in addition to that, we’ll see big players who have their own app store or their own mobile classic network. People like Best Buy or Wal-Mart are undertaking some projects and learning how to monetize from mobile developers and mobile advertising. On top of that, all the device manufacturers, the hardware companies, are beginning to offer distribution services and preinstallations. Eventually that’ll be available on platforms or in markets that we haven’t seen being fully monetized or even offered yet. Outspending is one way to do it, but in the long term, I think the indies will find their distribution channels in order to make acquisition affordable.
Takahashi: One thing that holds some of the big guys back, too, is that there are aggressive marketing tactics that they can’t engage in. They have to protect their brand, and they don’t want to be caught in that trap where they’re seen as exploiting consumers. When it comes to aggressive marketing, where is the line for acquiring new users?
Laakkonen: The line is where the customer says, “What the hell is this?” I think Apple has the right idea on this. They look at an app and say, “Is this good for the consumer?” If it is, it should fly. If it’s not, it shouldn’t. It’s like porn. I know it when I see it. Some of them that we saw start running high-risk marketing like incentivized downloads or bot farms and that kind of stuff are already killed. I think we’re moving beyond that. So there’s always some loopholes. My approach to anybody thinking about those would be, “Life’s too short to play with that crap. You don’t want to do that.”
Christoffersen: I would add to that I don’t think any of these gimmicks have been proven to work. I don’t think they’ve been proven to provide a return on investment. Even incentivized downloads, the returns are so small. The monetization rate is so small. If you spend money that way, you might end up with more users for less, but those users don’t stick around, and they don’t spend any money. At the end of the day, you have to bite the bullet and pay the price for a quality user. It makes more sense to focus on the game being good and implementing freemium properly.
Takahashi: I think some what we’ve done so far is scare everybody about user acquisition costs, but at least we’ve got their attention. What can be done for the little guy? Gilad, this may be a good time for you to weigh in with your explanation of revenue sharing.
Rotem: First of all, you need user acquisition. You have to do user acquisition, as I see it. You can’t count on getting featured in the App Store or the Marketplace. What I’m advocating is to stop paying cost per install — or paying an ad company for getting a user to install an app — and start working with whatever network, whatever channel you choose to work with. I feel that the industry is going to a more performance-based model, where we form a partnership between the traffic provider and the developer. There’s revenue sharing and cost per action [where the user does an action that proves the user is more loyal than one who simply installs it] — paying for actual paying users — or revenue sharing, dividing the revenue between the two.
Takahashi: How would you guys react to that?
Christofferson: There are so many different channels out there. There’s companies like Chartboost, offering the opportunity to do direct deals from one developer to another developer, exchanging clicks, to do cross-promotion. Also, supply is growing every day. There are so many new devices being activated every single day. This is to everyone’s advantage. Also, my advice would be to take advantage of seasonal dips. We see before Christmas — this happens every year — that prices keep going up. Everyone wants to get their game out there before the Christmas sales. You see a lot of developers trying to reach a higher ranking before the App Store freezes, or just gets prolonged, anyway.
Christofferson: That’s true. We decided to pull back on the period leading up towards Christmas, and rather focus on the period after Christmas, when there were so many new devices in the market. Because there was so much new supply, demand actually went down, as the brand advertising started to run out. Most game developers don’t want to spend very much anyway. On top of that, you have all these people with new devices who are looking for games and who have time to play. We took advantage of all those things.
Rotem: What I’m saying is that you don’t need to be worried about costs rising or lowering that much. This day is less expensive, that day is more expensive. When you share revenue, you aren’t worried about the quality of the traffic, you’re actually worried about the lifetime value. You have to know what your acquisition costs are so you can know that the value is higher than your acquisition costs. When you’re forming a partnership with a platform supplier, a revenue share model aligns your interests. You’re paying only when you generate a new real player.
Takahashi: So how much revenue are they sharing?
Rotem: As much as you want. As I said, this is a partnership. The way it works, when we decide to get to a revenue sharing deal, we both sign an NDA, we go out for drinks, we shake hands. Then the developer shares the sensitive data that they never want to share. We tend to mix lifetime value and conversion rates. What we need is to estimate lifetime value when we actually see revenue from the users. Generally speaking, roughly, we’re talking a 50-50 revenue share after platform fees. Facebook and Apple get their 30 percent. After that we take 35 percent of what the user spends on the game.
Laakkonen: We have revenue sharing deals on Facebook where we’re using our traffic from our cross-promotion network with publishers who want their game to grow. We took all the risk. We could have lost all the traffic. We ask for 50 to 65 percent revenue share back to us, after platform costs, only for users we brought. We took on the risk. You can still find your users out there and you would own 100 percent of them. This is actually a specific form of publishing. This is distribution. In publishing deals, people would take as much or more, and they may not even fund it. When you look at the spectrum of what you can do in user acquisition, it’s getting smarter.
It’s time to maybe think about outsourcing. If you decide on your core competence, you can go mobile or Facebook. You still pay. They’re taking a revenue share model system, a distribution partnership. Then you can go all the way to a publishing deal, in which somebody else is publishing your game, like Activision, and they acquire ownership of your [intellectual property].
But before you jump into user acquisition, you should probably start on the quality of the game. If you can’t put in any money into quality, nobody’s going to give you users. Nobody’s going to publish a game like that. You have to start by getting your own costs in order.
Takahashi: Billy, tell us about App-o-Day and what’s been changing in app discovery lately.
Shipp: We’ve all the seen the growth of app discovery services like Free App-o-Day over the last few years. Some new services entered the market. The reason that’s happening, and the reason these services are attracting funding, is because fundamentally they work. Dean talked about discovery happening in the App Stores. Jussi talked about discovery happening via word of mouth. Both of those are great mechanisms, but you can’t count on an App Store feature.
You need something to kickstart word of mouth. Running a promotion is a great way to do that. A sale is the classic model in the marketplace to drive demand. If you have too many fish at the market, then you need to sell more fish. Then you lower the price to get more customers for a period of time. That’s exactly what these platforms do. I think we’re seeing more users gravitating towards these platforms as a way to find great deals on great apps. We’re finding more developers using them as a channel.
Takahashi: What kinds of things are going to come to the rescue here? Is consumer app search going to be an interesting feature? Something like a Google search for mobile.
Laakkonen: I’m not really a believer in that. Outside of the App Store, app search won’t matter. For the average consumer, when you ask them, “How do you find games?” it’s either “A friend tells me” or “I check the app store.” There’s not going to be much room left for specialized app search even when Apple’s App Store search competitors give Apple a few tips on how to fix it. There’s not enough consumer demand for an application that lets you search for other applications. It’s a niche. I’m not really high on it.
Bielau: I see the App Store’s evolution going in a direction where the whole social forum comes into play, and this will be dominated by Google and Apple at first. I see that as soon as those two bigger players are opening up and providing voice and SMS [text message] services themselves, as the telcos do right now, they have on the one hand the information about my app usage — what games I play, what apps I own — and on the other hand they’ll have that interaction layer between people. When I call Jussi, when I’m sending an SMS or a message or chatting with him, I’ll get a recommendation based on my preferences and based on my user behavior in the app ecosystem.
These kinds of contacts, I would say, will be happening soon, promoted by one of those two. That could be by audio, when we talk, or prior to the talk. Maybe the call is for free after an app recommendation. Or it could be in the context of messaging. The system makes a recommendation that would fit my preferences, as well as having an indicator of what my friend Jussi has done with his applications before. This is just the whole approach. Currently, the app stores are more or less closed systems. It’s kind of a broadcasting approach. I definitely see something happening in that other layer.
Disclosure: The event organizers paid my way to Hamburg, where I moderated this panel. Our coverage remains objective.