ABOUT AURORA: LinkedIn  | MEDAL TV


| Careers at MEDAL TV




ABOUT ROCKETSHIP HQ: Website | LinkedIn  | Twitter | YouTube

KEY HIGHLIGHTS

⚒ Why Aurora and her team built 1 prototype per week for 10 weeks.

🩺 How Aurora and her team gathered user feedback to decide what to build.

⏳ SDK fatigue and how to overcome it

☔️ The common mistakes that apps make while considering viral flows.

🛹KPIs that impact a referral flow.

📱The power of dark social.

KEY QUOTES

Make the sharing experience enjoyable

For Fun Run, it was more around all the funny moments that happen when you’re playing with friends, like moments where you’re pulling a friend back with a magnet right before you’re crossing the finish line. In the Adult Coloring Book for instance, sharing just a screenshot of the final moment is not as unique as sharing the time lapse of you actually constructing that drawing.

Don’t expect your users to go through multiple steps of promoting your product

The user needs to say – I want to record something and then they say- I want to stop whatever is happening and now I want to share this potentially multiple megabit file through a social network. And there are so many steps, as well as places along the way where the user would drop. We thought- our game is so cool, there needs to be a better way. And so we knew we needed to differentiate from that. What came out of that was that we focused on only sharing highlights and making it as easy as possible to share those highlights. I read a study a while ago that people’s attention span on mobile is like eight seconds. And so we kept the recordings to 15 seconds, and our default is five.

Show them what to do

A story I like to tell when I get a question like that is back in 2014-2015, when I was still with Dirtybit we had this big game studio that was building an ad network at the time, approach us and tell us about this concept of rewarded video ads. At the time, we only had rewarded video ads hidden in a challenge scene, just like a lot of challenges you had to complete, 20 coins for watching a video. And they told us- if you move this rewarded video to within the game loop, you’re actually going to monetize way better. We were still reluctant because we were so scared of the impact on retention. They showed us numbers, but we were still reluctant. They eventually flew to our office and showed us exactly where to prompt the user, how to prompt the user, and all of that, and we implemented the changes we were convinced to try. And after that we did a case study with Chartboost, which was now like 50-50, split on revenue from in-app purchases versus ads. We came to a 15-85 split. It was really huge.

Clarity in communication to users is vital

Whenever we found bigger developers that we wanted to work with, we made similar dissections to explain exactly how to communicate it. It’s all about how do you communicate to the user that this is first and foremost, a value add for them? And how do you make it as easy as possible for them to follow through. And so we’ve also done a lot of content marketing on that.

Overcoming SDK fatigue

On the SDK fatigue, we tried to speak to developers, we tried to structure things in a way so that it would be easy for a developer to go through our docs and be like, “Huh, I wonder how much time this will take me”. And then like an hour later, it would be integrated, it would be a prototype that they could show their team and say, “Hey, what do you think? Should we integrate it deeper and really communicate everything in a good way to our users?” So we spent a lot of time on our documentation and making that as simple as possible. 

The misconception of virality

Virality won’t fix a broken game. Unfortunately, a lot of the developers we’re working with come to us, and say “this will be great, people want to share the content, because my game is so great.” And then the game isn’t necessarily as great as they hoped. I try to be candid, where I can. Obviously, I shouldn’t kill anyone’s dream. But I try to help and try to address the core game first.

The need for clear steps to a referral flow

Why this is beneficial for them, how they’re going to do this, in the case of a referral, getting the reward, because the thing about the referrals is that it’s not just about the sharing, someone needs to be on the recipient side, and follow through on their end. And so if you as a user think that “Well, my friend said they’d click the link, I should get the reward,” then that isn’t usually what would need to be done. Your friend needs to actually click the link, open the app and trigger the SDK call on the other end, for the rewards to come. And so there’s frustration among users if they don’t understand that, which is hard to educate.

What is Feature Engagement?

We talk about feature engagement, which is when someone is playing a game, and they see the share capabilities, and they hit share. That’s feature engagement. Second is share ratio, which is you’re hitting that share button, or you’re actually hitting a destination or tapping into destinations to share to. And the third is install ratio, which is when someone is actually sharing and a friend starts to play.

Why your Install Ratio will never be 100%

The last one, which is usually what people are most focused on, is install ratio. That is someone selecting a share destination and a friend actually going into play. And a few reasons for this also not being 100% is that there’s so much that can go wrong in the communications with our friends. There are also share destinations that don’t allow you to have a link. So you can’t actually correctly attributes this install. And there might also be that the reward isn’t enticing for the sender to properly follow through on having your friend join.

Private messaging apps vs. public forums

Another thing is how dominant private messaging apps are for this is still surprising. Because it’s something that I keep having to educate developers on. I think everyone still would very much like to go viral in a public space like Twitter. But now it’s more close knit communities. When I say private messaging, it’s not necessary just one-to-one, a group of friends on WhatsApp or even like a Discord community, can be a shared destination.

FULL TRANSCRIPT BELOW

Shamanth: I’m very excited to welcome Aurora Klaeboe Berg to the Mobile User Acquisition Show. Aurora, welcome to the show.

Aurora: Thank you Shamanth for having me. It’s been in the works for a while. So I’m excited.

Shamanth: You came very highly recommended. And I followed a lot of your writings and the deconstructions that you will speak about in this interview. I’ve been very impressed. So it’s a pleasure and honor to have you Aurora.

Aurora: Awesome, thank you so much.

Shamanth: Let’s start with something you said- in the summer of 2015, we spent 10 weeks prototyping a new thing every week. Tell me more about this. What were these prototypes? And what inspired all of those?

Aurora: That’s a great question. That’s obviously about the origins of Megacool. In 2015, Nick who founded Dirtybit, the creator of Fun Run,  and myself decided to move to the States. We were present in Norway. We wanted to really embrace that very ambitious American startup culture, which we felt was lacking back in Norway. The way we wanted to go about this was to prototype 10 different concepts over summer, spend one week on each concept and then decide what would eventually become Megacool. And the process for this was inspired by Wooga, who at the time had this prototyping funnel, where they put in 40 prototypes. And then 10 came out to be tested in the market. And maybe five went on to soft launch, and one made it out on the other end. And we wanted to apply this concept to how we went about deciding what our next venture would be. These 10 different prototypes had all sorts of shapes and forms. Some just became like mockups that you could move around through Keynote presentations, but it was really to make sure that whatever we ended up building was actually something people wanted. So that was the baseline. But then, ironically enough, what ended up becoming Megacool was actually the only game related idea. So we wanted to assess other things as well. So we tested a photo sorting app for your phone, tested the CMS tailored to services like hairdressers and then also worked a little bit on a privacy and motivation focused solution for reading articles online.

Shamanth: That’s crazy. And I know you said you almost wanted to move away from games, you had only one gaming related concept. And I am impressed by the sheer variety and heterogeneity in the concepts that you did describe and to do one per week, and that’s a very aggressive pace. So what happens next? So you take 10 weeks, presumably coming up with 10 prototypes? What happens next?

Aurora: One of those weeks, we were attending a hackathon. And we built out this concept.. And it’s really about building a feature rich and easy to use weightless tool for game launches. Based on the feedback we were getting from that it was mostly around- this is something that a lot of people have, we didn’t necessarily see that we could add as much feature rich content on top of it to really build a sustainable business at the time. 

Since then, there’s been a lot of tools that have done really well. None of them have gone on to be big businesses. Through this, we were like, well, we’re constantly asked for advice on organic growth, and we would help a lot of companies do that. So how can we do this more at scale? And how can we really figure out what is one thing that we’re helping people with in terms of organic growth that we can build the solution around? 

So we started dissecting and publishing dissections for popular games at the time, like Crossy Road, Agar.io, Candy Crush, and we pointed out all the things that did well to create growth and all the opportunities they left out. And I know that Miniclip actually implemented some of the changes we proposed after the Agar.io posts. So that was pretty cool. But what we did about these dissections were that, anywhere we gave suggestions, we linked to a beta signup page and had people sign up to beta, not necessarily saying much about what the beta would be, and then ended up interviewing the people that signed up and trying to understand exactly what they needed and where we could build our tools. The dissection part was really inspired by useronboard.com. They didn’t focus on gaming at all. So we saw that as an opportunity to differentiate. I think right now I want to call out growth.design as probably the best dissection out there.

Shamanth: Yeah, Samuel Hulick from User Onboard was on this show. He is as hilarious a person as he is on the dissection site. But back to your story. You were like, hey, there’s gonna be a beta. There was no product because you didn’t know what it would be. In some sense, you’re building what the customer wants. And the way you describe it, it makes me realize how hard it is to not be attached to your own ideas and just stay open to what the customers will say. And it also sounds like the initial prototype of onboard you guys said, this gives us a starting point that we’re not going to build that but we’re actually going to listen to what users actually what. Yeah, that’s very impressive, like I said, to just distance yourself emotionally from the stuff and actually listen to your users. And when you’re listening to your users, what are some of the things that folks said and how did these insights ladder up to the product you eventually built?

Aurora: It was two-sided, I really want to give a shout out to Candywriter, the folks over there Kevin and Nadir. And if you don’t know Candywriter, they’ve actually been working on putting out new genre defining games. Early on was What’s The Pic, and a lot of, what’s the games, then it was Letter Soup, which we’ve seen tons of clones from since. They also did an Adult Coloring Book, which I don’t actually know if they were the first to do the coloring books. But they were one of the most memorable and engaged people that we spoke to. They told us that they were constantly looking for growth mechanics, where they were the first one to export them. Until a platform would turn it down. For instance, they were among the first to really take advantage of screenshot sharing to Facebook. Which was one of the main growth drivers for What’s The Pic back in the day. They were really eager to work with us and figure out how they could implement the next big growth feature. And so we built an alpha version with them, which was good because they were iOS only for the games that we worked on. And that really helped us move faster, only focusing on one platform. So we did that with the Adult Coloring Book and Letter Soup. That was super exciting to see: the growth strategy.

Shamanth: Yeah, just to jump in when you see the alpha version. Can you talk for folks who don’t know? Can you describe the product and what it did?

Aurora: So going back to how we ended up with what it became, both Candywriter, as well as our experience with Fun Run was really speaking to- how do we make it super easy for users to capture and share content from their games. Screenshots had been a way to do that. But for most games, it didn’t really capture what the game was really about. And for Fun Run, it was more around all the funny moments that happen when you’re playing with friends like moments where you’re pulling a friend back with a magnet right before you’re crossing the finish line. In the Adult Coloring Book for instance, sharing just a screenshot of the final moment is not as unique as sharing the time lapse of you actually constructing that drawing. And another thing that we were talking with them about, which really resonated, was to tie this all with referral programs because referral programs have been huge for non gaming services like Dropbox and Uber. But we hadn’t seen them enter the gaming space yet. And so we wanted to build that out with them. So the alpha version was a light GIF recording solution that was tied to a referral program.

Shamanth: By that you mean a user could record a GIF, share it with friends and they would be incentivized to do so. If the friend accepted they would get in-game currency.

Aurora:And other supporting things for this was that Everyplay was pretty big at the time. So we spoke to a lot of studios that were using a replay. What they kept telling us was- we know there’s something there but we don’t think that the way Everyplay works. The user needs to say – I want to record something and then they say- I want to stop whatever is happening and now I want to share this potentially multiple megabit file through a social network. And there are so many steps, as well as places along the way where the user would drop. We thought- our game is so cool, there needs to be a better way. And so we knew we needed to differentiate from that. What came out of that was that we focused on only sharing highlights and making it as easy as possible to share those highlights. I read a study a while ago that people’s attention span on mobile is like eight seconds. And so we kept the recordings to 15 seconds, and our default is five.

Shamanth: Yeah, and I know you’ve mentioned earlier, but I think what made the difference was that you treated GIFs as a standard of output rather than a video file. That sounds like Everyplay said export this entire video and you thought a small snippet is good enough.

Aurora: Yeah, at the time, we also looked at what format delivers the best across the sharing destinations. And an animated GIF was actually playing better than a video. It would usually auto loop while in a video you need to play. But I want to add that, since then, we do now support both video and GIF. The quality is still in the GIF format, because we pass it on through to the destination depending on which it’s optimized for if it’s GIF or video. So it can do both.

Shamanth: The way you described the workings of the product, Megacool, sounds like that developers had to integrate the SDK in order for the user to be able to record. And what isn’t as obvious is that developers have to integrate it in a way that users actually understand what’s going to happen. And the users have to use it correctly, rather than have it be a feature in the game that a user never notices. So there are multiple layers here, like you have to educate the developers, they have to educate the users in a proper product focused way. So how did you think through this, what would appear to be a multiple layer problem?

Aurora: I think first and foremost, we were approaching this from a perspective of being game developers ourselves, right? Like what would make us do this? And what would make us understand what to do? A story I like to tell when I get a question like that is back in 2014-2015, when I was still with Dirtybit we had this big game studio that was building an ad network at the time, approach us and tell us about this concept of rewarded video ads. At the time, we only had rewarded video ads hidden in a challenge scene, just like a lot of challenges you had to complete, 20 coins for watching a video. And they told us- if you move this rewarded video to within the game loop, you’re actually going to monetize way better. We were still reluctant because we were so scared of the impact on retention. They showed us numbers, but we were still reluctant. They eventually flew to our office and showed us exactly where to prompt the user, how to prompt the user, and all of that, and we implemented the changes we were convinced to try. And after that we did a case study with Chartboost, which was now like 50-50, split on revenue from in-app purchases versus ads. We came to a 15-85 split. It was really huge. Yeah. And this we’re talking about 2015. And then, through my journey with Megacool, I don’t think it was until 2017-2018 until we started to really see rewarded video ads being the norm in everything. 

Having this experience, I knew that it wasn’t just enough that we had credibility from being game developers ourselves, but we needed to educate our users the same way. And so we actually took the dissection piece forward.  Whenever we found bigger developers that we wanted to work with, we made similar dissections to explain exactly how to communicate it. It’s all about how do you communicate to the user that this is first and foremost, a value add for them? And how do you make it as easy as possible for them to follow through. And so we’ve also done a lot of content marketing on that. But it’s really like a chicken and egg situation, right? If you’re looking at a studio that has implemented any growth feature, and they haven’t done it well, they are doing well for other reasons. It’s easy to just replicate what they’re doing without necessarily getting the benefits. And so it took a while before we got an implementation in place that did really well that other people could follow. I think working with the Zeptolab team on their game CATS was really one of the catalysts for that.

Shamanth: Yeah, it sounds like you were effectively consultants. You’re treating this almost like an enterprise-sales implementation. You have to do it for them because if you left it for them to figure out, it wouldn’t necessarily work out the way you intended it ought to. Since it wasn’t a one-size-fits-all implementation, it sounded like the product had to be customized for different games in different game loops. 

Aurora: I think this is a good opportunity to also mention why we wanted to do this and really help the game studios. With Fun Run, we didn’t have funds to spend on user acquisition. Our LTV was really low, we always had to think very creatively. And we knew that there were so many games out there that didn’t have the resources, or didn’t have the time to build these things by themselves. And so we looked at what we were doing as a way to help the Indies as well as driving down the user acquisition costs for the larger players. That was a big part of why we put so much effort into doing this.

Shamanth: And I think this can be especially valuable for indie developers that don’t have access to the crazy funds that sometimes scaling UA can require. We’re talking about SDKs, a term that I’ve tended to use is SDK fatigue, just because again, in my years in working with studios there’s been a certain resistance to integrating yet another SDK as a lot of developers put it. Did you face that friction and how did you address that?

Aurora: Definitely experienced that and had to address it multiple times. It wasn’t just the SDK fatigue though, it was also “Well, we’ll just build it ourselves.” And I think we’re still waiting for that to happen. But really on the SDK fatigue, we tried to speak to developers, we tried to structure things in a way so that it would be easy for a developer to go through our docs and be like, “Huh, I wonder how much time this will take me”. And then like an hour later, it would be integrated, it would be a prototype that they could show their team and say, “Hey, what do you think? Should we integrate it deeper and really communicate everything in a good way to our users?” So we spent a lot of time on our documentation and making that as simple as possible. Obviously, if you don’t speak to the decision makers, then when we try to buy stuff, the decision makers by being very smart about that. And also like the content marketing, the credibility, not just from us. We’re very much not an ad network, we wanted to make that distinction clear. But then, I think the most important thing is that we’ve been persistent over time. Because you see, all these companies start, they integrate their SDK, and then they could go bonkers. And then you’re like: Well, what do I do now? And so us staying persistent at the conferences everywhere has built sustained credibility.

Shamanth: To switch gears a bit, you’ve seen viral referral mechanics for so many games from close up. And obviously, you’ve dissected a lot of them as well. So what have you seen are some of the common mistakes that developers make while setting up referral based flows or viral flows or invite based flows? 

Aurora: Virality won’t fix a broken game. Unfortunately, a lot of the developers we’re working with come to us, and say “this will be great, people want to share the content, because my game is so great.” And then the game isn’t necessarily as great as they hoped. I try to be candid, where I can. Obviously, I shouldn’t kill anyone’s dream. But I try to help and try to address the core game first. Then the other thing that we’re seeing is that people are just slapping on growth features. So a very common thing is to just add a “Share to Twitter” or “Share to Facebook” button in your game. I think that’s still persistent. But in 2021, you’re not necessarily seeing games go viral on Twitter anymore. I mean, there are exceptions. But predominantly, all of the sharing happens through private messaging destinations. So putting those share buttons in there for Facebook and Twitter is most likely not something that your users are going to use and are just taking up the inventory space. And then not actually communicating to the user what is going on. Why this is beneficial for them, how they’re going to do this, in the case of a referral, getting the reward, because the thing about the referrals is that it’s not just about the sharing, someone needs to be on the recipient side, and follow through on their end. And so if you as a user think that “Well, my friend said they’d click the link, I should get the reward,” then that isn’t usually what would need to be done. Your friend needs to actually click the link, open the app and trigger the SDK call on the other end, for the rewards to come. And so there’s frustration among users if they don’t understand that, which is hard to educate.

Shamanth: Yeah, and a lot of these distinctions are seemingly subtle, but so critical. I know in some of the blog posts that you had that you talked about how important the clarity of the communication is. Somebody who’s receiving the message, if they don’t know what they’ve gotten, there’s just no point in a referral flow. I’ve seen some of that, in my experience as well. When you’re looking at a referral flow, what are some of the KPIs that you would recommend looking at? And also, if you’re looking at certain KPIs, what might be some of the important ways to impact these KPIs as well?

Aurora: So from a high level standpoint, we’re defining it for three metrics. So we talk about feature engagement, which is when someone is playing a game, and they see the share capabilities, and they hit share. That’s feature engagement. Second is share ratio, which is you’re hitting that share button, or you’re actually hitting a destination or tapping into destinations to share to. And the third is install ratio, which is when someone is actually sharing and a friend starts to play. If I were to start with feature engagement, that is the funnel, right? That’s where it’s really important to make it clear to the user what’s happening, and what they have to do. So things to consider there are actually having incentivized referrals, that’s a great way. Because as a user, I’m more likely to share something to a new friend who’s not engaged. If I knew that, that’s how I would get an incentive. If there’s no incentive, I’m more likely to re-engage an existing player, because they know what’s going to be relevant for them. 

And ways that you can also increase the future engagement is to actually add sharing capabilities to several places of your game, you can do it and the store, in the community or friends part of the game, and you can do it post game. So have a preview of the gameplay that was just recorded and share that. That’s just a few examples. And talking metric wise, based on our data, anything over 15%, we consider good. So 15% of your player base are engaging with sharing features on any given day. And then when you look at the share ratio, it’s about how you’re hitting share and actually selecting a share destination. The reason that this is not the same step is that we’re tied to the native share flow. And you don’t want to clutter the view with a lot of shared destinations too soon. So you want to bundle them up in one share experience, and people can drop off here, because they’re just exploring the future. They’re not intending to share, maybe they know the destination they would prefer to share to. This again, comes down to communication. And this is also where we really recommend having a preview of the content you’re actually sharing. So you know what’s going to happen. And in terms of metrics here, we recommend anything over 25%, which is considered good. And the reason it’s not a 100 is that you have so many users that are constantly going to be clicking around anyways, with no intention of something happening. So this is more based on some realistic metrics. And the last one, which is usually what people are most focused on, is install ratio. That is someone selecting a share destination and a friend actually going into play. And a few reasons for this also not being 100% is that there’s so much that can go wrong in the communications with our friends. There are also share destinations that don’t allow you to have a link. So you can’t actually correctly attributes this install. And there might also be that the reward isn’t enticing for the sender to properly follow through on having your friend join. And here we’re looking at recommending anything over 10% as a good share ratio.

Shamanth: Yeah, that’s such a seemingly simple flow, but you just described exactly how complex and multi-layered it is. And what’s also impressive, especially for somebody looking at it from the outside, is how each of those single steps can be optimized and improved. And as you just pointed out, and I think that’s something that a lot of outsiders don’t quite appreciate, because it just seems like a relatively simple flow. When you looked at a lot of these flows, what are some other things that have surprised you?

Aurora: I mean, The first one is really how far we’re yet to go. I think the analogy of the rewarded video ads. How long it took for that to become mainstream was there. I think it’s still really hard to do this in a way that is easy to communicate and make sure that the players understand it and want to participate. So that’s one thing. Another thing is how dominant private messaging apps are for this is still surprising. Because it’s something that I keep having to educate developers on. I think everyone still would very much like to go viral in a public space like Twitter. But now it’s more close knit communities. When I say private messaging, it’s not necessary just one-to-one, a group of friends on WhatsApp or even like a Discord community, can be a shared destination. Another thing that we didn’t really touch on, about how hard some platforms or dominant players like Facebook make external sharing, which also breaks the attribution flow a lot of times, so it makes it hard to measure the success even though there was success there. One example I’d say is that WhatsApp for instance, on Android they allow us to share both the length and message which we preset, just to ease the conversion rate and the GIF or video, whichever we want. While on WhatsApp, iOS, you can only share either the text and the length or the GIF. So yeah, it’s hard. 

Shamanth: Yeah, I didn’t know that. In some of your blog posts, you’ve shown how dominant private messaging platforms are, and that nobody talks about this, when they talk about going viral. I imagine you guys had that visibility into that data, because you have the SDK, and you see the number of messages that are getting sent, but otherwise, it’s hard for any outsider to conceive of exactly how impactful this can be. And when you also think about it, at least a lot of casual games I have seen have tended to have very strong key factors. But some of us are intentional about it by using clones of the sort that you described that a lot of them even without those are, when I have looked at those on themes, I’ve always hypothesized that these are private messaging platforms, but we never really know. It sounds like you do and you have the data and perhaps link to some of the space.

Aurora: It’s interesting. It’s coined as the dark social, because you don’t have visibility into it. 

Shamanth: This has been a fascinating conversation. We’ve covered a lot of ground. And I’ve certainly learned so much. Virality is a little bit of a buzzword. You’ve obviously talked about your own personal journey, which I find very brave and instructive. And coming up with a product and being completely open to what it could be. But also, you’ve gone deep on what has been around for a long time in terms of virality and referrals, and built and conveyed a very nuanced understanding of that. Very impressive. I’ve certainly taken a lot of notes or or, and I’m excited to put this out into the world. But before we let you carry on. Could you tell folks how they can find out more about you and everything you do?

Aurora: Definitely reach out to me on LinkedIn or Twitter. Ideally, email is most responsive. My email is aurora@medal.tv. And I should probably also share that, since we started, we got acquired by Medal.tv, which is working within the same genre, but is a consumer focused platform primarily for PC gaming. We’ve now also launched an Android app, which lets players run Medal while they’re playing in any mobile game. And with the tap of a button you have the x seconds that you want to record from that gameplay, which is really good for communities, as well as developers who want to make bug testing easier while the game is still being developed. Medal is growing rapidly. So we’re looking for product managers specifically for our social push. So if you’re interested in this space, I’d love to talk to you. And we also have a subsidiary called Top.gg which is the largest Discord server. And we also have exciting roles there.

Shamanth: Thank you so much for being a guest in the Mobile User Acquisition Show. It’s been a pleasure and honor to have you.

Aurora: Thank you so much. This has been great. Thank you 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, Google Podcasts 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!