Tuning Acquisition of App Users
Tuning Acquisition of App Users
Arsen Pereymer, Founder & CEO of Mobiquire.com talks on ASO, Games, Bursting downloads, planning for launch and why the App is just 30-40% of the full picture in App success.
We score all out interview anecdotes based on 3 axes of: acquisition, UX and retention.
Let us know on our Twitter account how you score it.
- User acquisition is an interative process. Attribute what sources are driving valuable users and then double down on that.
- Use a number of hacks in combination to burst your presence in the Appstore. Some of this is paid, reviews and download spikes.
David: Hello. This is David Jones and I’m here with Arsen Pereymer of Mobiquire or is it “Mo-BEE-quire?” I’ll let you tell us what it is. And he’s going to tell us a little bit about their acquisition capabilities. Hi, mate. How are you?
Arsen: Wonderful. Thank you very much for having me. I really appreciate it. So, yeah. Mobiquire is a mobile user acquisition platform to help mobile app developers acquire quality users for their apps. The keyword here is “quality.” We’re not about delivering just any user. We’re specifically delivering quality users which have a high retention rate for their apps.
David: Great. Okay. So tell us a little bit about that. Actually, I know you’re based out of New York. Is that correct?
Arsen: Yes, we are based out of New York. And we also have offices in San Francisco as well.
David: Okay. And is your customer base typically, are you working directly with customers that are doing apps or are you kind of working around the world? What’s the story in terms of your customer base?
Arsen: Okay. Tactically very simple. Primarily, our customers are top app publishers, many of which, you know, people have heard of. And we also specifically work with start-ups. We recently got funded and have developed or about to launch an app. And we want to acquire users for their app. Our main customer base is primarily US-based. However, we are expanding in other countries as well.
David: Okay, all right. So let’s talk about how you actually help out. So let’s say I am one of these early-stage funded start-ups, I’m possibly doing a utility. It might be around food delivery or message, or something like that. How will they work with you?
Arsen: Okay. It’s very simple. So first of all, we like to go over new things. We like to analyze the app, ensure the app has a great quality. Because like this, if we deliver users to the app, and the app is not high quality and the retention is down, then we feel we didn’t do the job right. So we want to analyze the app, ensure it has the right qualities which will retain users. We’ll analyze for example, ensure that push notifications are asked at the right time. For example, a lot of apps ask push notifications right when you launch the app.
Arsen: However, people know that most people will say “no” to them. So how can we go ahead and increase those chances of a “yes?” Maybe strategically pointing the developer where to ask for the push notifications properly. And also given the user, the request why we want to go and ask for the push notification. So we’ll go and analyze the app and so forth. Once we go and do that, we’ll start preparing a strategy and a campaign, to launch a campaign to go and try to drive traffic to the users. And then we’ll start using our algorithms and proprietary technology to do media buys, advertise the app, look in our own network of over 2,000 apps and then continue it from there.
David: Great, okay. All right. So I certainly agree with you in the push notification permission thing. We’ve seen kind of, like opt-in rates and we know around 40%-50%, if you throw it upfront. But you can get an uplift up in the 70s or whatever in terms of opting into push, if you actually make sure you’re tying it with the right kind of explanation.
Arsen: Exactly. And that concept pack could also work for asking for the user location as well. Same thing, if you’re going to ask immediately, most users will say no because they’re afraid to feel like, “Oh, now you’re going to be tracking me.” But if you tell the user why you want to go ahead and ask for their notification—for their location, then most of the time, they’ll say, “Okay, fine. That’s pretty cool.”
David: And does that extend across into other things like address books and stuff like that, like other capabilities on the phone?
Arsen: 100%. I think it’s very, very important to be very clear to the user why you want to ask for certain information, why you want access to certain things, because unfortunately the messages from the iOS and from Android, it’s a little bit negative and it’s scary-looking. So how you could make it very pleasant to the user, I think that increases the chances of a “yes.”
David: Yeah, I have another chance to have a look at Android in, but I know Android is going away from sort of you accepting the permissions up in the Play Store. But apparently, it’s going to prompt when you actually are using the application. Have you had a look at it to see whether the permission prompt is friendlier than on Apple?
Arsen: Yeah, I have looked at it. Unfortunately, it’s actually not friendlier than Apple. And it shows a lot of information right in the beginning, “Oh, we’re going to access your address book. We’re going to access your location.” And all that stuff, and it’s very scary-looking.
David: Right. Okay.
Arsen: So if you can make it “unscary”-looking, then you’re a winner.
David: Yeah. Absolutely. So how does this work? So I’m on this mobile app. I’m doing food delivery. Are you presenting with a platform or you’re primarily kind of like a consulting engagement?
Arsen: So we start up on the consultant engagement. Again, this is not necessary to go ahead and do. It’s something we recommend to go ahead and do. And a lot of apps say, “No, we don’t want to go ahead and do that. We just want to go directly to user acquisition.” In that case, we do that as well. That’s our main bread and butter. So our platform is actually very simple. You log into our platform. You state how many downloads you wish to receive and you start being specific. For example, you could start saying, “I’m interested in receiving 10,000 users in the New York area for between 18 and 34.” We can go that specific downloads, and then you select the start date of your campaign and the end date of your campaign. You could upload your own creatives, or we could create the creatives for you. The creatives which we support are interstitials, banners, videos, and even native ads. And that’s it, the campaign gets started and is on a fixed CPI, which means “cost per installs.” So it’s fixed for the lifetime of your campaign.
David: So usually, kind of, CPI is synonymous with “not quality.” Why do you feel is your approach is going to drive you quality users?
Arsen: Great question. So what we go ahead and do, initially, the quality is poor, okay? And the reason for that is it because it allows our algorithm and strategy to learn what are the best type of users for you. So we have very, very strong data analysts on our team and data scientists, who will look closely at the data and say, “Hmm, we brought X amount of users within a 24-hour period, okay? Now, let’s look at day three and how many of those users right now are sticking in the app.” And if we see that the retention is very, very low, that means we have to change our strategy internally to get different type of users. So that’s the key. So initially, our cost internally is much higher, okay? But we try to lower that down. That it doesn’t matter because for the end result is the users/actual customers pay in a fixed CPI. So our goal is to try to pinpoint exactly the best quality user, and it will determine as quality as our CPI, is somebody who will be retained in-app. So for example, somebody who downloads the app on day one, and on day three, they’re still using the app. So typically, what we try to aim for is between 30% and 40% retention rate on day three of the app.
David: 30% and 40%.
David: Right, okay. So that’s pretty good. So let’s see how would that work if somebody comes to you and says, “Actually, I want to measure the quality of customer by what percentage or the number of people who are actually going to make a transaction in the application.” Can you adjust your strategy for quality based on an actual revenue number?
Arsen: Great question. So what we call that—or actually the industry calls that—is a CPA, a “cost per action,” So we could go and measure by exactly that transaction basis. So for example, like in your example, food delivery. They may say, “Hmm, we’re only interested in the retention being that somebody actually makes a purchase or actually orders the food.” Clearly, we’re not going to be able to accomplish that right away from the first few days because we need to really learn a lot about the user. Well, once we go ahead and do that, that could be our goal and metric to go by. For that, the prices are a little bit different because it allows us, we need to first learn enough about the type of user to bring to the table. And typically, for those type of engagements, we require a set amount of time to run the campaign. So typically, sometimes you may want to run the campaign through one or two weeks. For those type of campaign, we require at least a month to able to run.
Arsen: And that’s primarily the focus behind that is so we could learn the best type of user to bring.
David: And so in that situation, you then kind of look at the performance, the CPA performance, over the course of 30 days as opposed to a three-day sliding window? That’s how that works?
Arsen: That is correct, yes.
David: When we say, “when you look at it,” this is basically your engineers or your analysts looking at that, and then you provide summary reports back to the user or does the user have a platform where they see this stuff or tweak? Do they tweak anything themselves?
Arsen: So, we have our own system internally, how we provide reports to the user. However, we’re integrating with a lot of third parties, such as adjust the IO app flyer, control them and so forth. Where those are third parties where sort of users will trust them more because it’s a third party reviewing the data.
Arsen: So, or at the same time, they can look at our performance reports as well.
David: Right. Gotcha. Okay, so in this podcast, we like to kind of talk about one win and one fail. Just firstly, did you guys come out of app development yourselves? Is that your background?
Arsen: Well, that’s my personal background. I had been actually an app developer since 2008. I’ve actually developed an excess of 400 apps, a majority of them have been games. I had game companies and game portfolios which I’ve sold, and app companies which have sold. So I know the business extremely well. And how it basically started was I put together a great, extremely smart team, a few PhDs, data scientists on my team, where we started looking at things and said, “We’ve learned so much about the app business, and now it’s time for us to help other app developers because we went through the hurdles, we went through the failures personally, and also as companies. Now, let’s see how we could use all those information which we’ve learned and pass it over to these young developers who have great ideas, but they need the help with execution and user acquisition.
David: Yeah, I was talking to a guy on Friday. Their company has been going since 2009 and they’ve had across the portfolio of games over 250 million downloads. And so what happens as an organization, they’ve developed an incredibly high amount of competence internally, that they just keep churning back into their app. But your approach is that you’ve kind of taken that competence and sort of, like now, try and basically create a business model around taking that to other developers as well.
Arsen: Exactly, correct.
David: So tell us about going back over all those games that you developed, tell us about what was kind of like a one win that sort of was a real surprise for you? You know, you were doing something, you’re doing the same old thing, you’re grinding away, but then one of the things or one of the tactics just kind of quit.
Arsen: Okay. So the example that I’m going to go and bring up is what actually happened the last year. I believe it was February of 2014. So we had one game, probably you have heard of it called “Fly Birdie.” It was one of our companies, when all of a sudden, we were continuing through what we were doing, all of a sudden it started growing into charts. And at one period of time, it actually went to the number one game—sorry, number one app—in the US and like 30+ other countries.
Arsen: And over there, we were doing our standard stuff of app store optimization, improving things with the retention, stuff like that, things which we’ve been doing all with our other games and apps. But somehow, in some miraculous way, that game made all the way to the top, and actually achieved over 8 million downloads in a very short amount of time. We’re talking about four or five weeks.
David: So was there a magic trick that just happened to work for you?
Arsen: The magic trick, the best way to say what the magic trick was at the time, during the Flappy Bird phase, so I guess we rolled that wave a bit. And I guess that was the magic trick over there.
David: There was a lot of guys trying to ride that wave. [Laughs] So did you get it right just in terms of keywords or the numbering of the app or what?
Arsen: A combination of a lot of different things: a combination of keywords, combination of going viral on social media. And it’s interesting because these two factors is stuff which everyone does anyhow. And just, I guess, we were very lucky. And like you said, that was just the win which happened.
David: Yeah. Do you find that app developers kind of—is it such, kind of like a tactical thing nowadays? So we just had Labor Day here. Do you feel as though there would have been apps that have gone up with very Labor Day-specific things or they adjust their keywords just specifically for the holiday and so on?
Arsen: Yeah. So I haven’t a lot to say on that. In short, that concept of being very season-specific, and I quote “season” because it could be Labor Day, it could be Thanksgiving, it could be Halloween and so forth. So those things will work, yes. But the problem is when you have thousands of other app developers doing almost exactly the same thing, the competition is so fierce that it doesn’t work as it used to. So if we were having this conversation two years ago, I would say, “Yeah, it works great.” Make your app very theme-centric. Make the app icon-themed centric, theme meaning “season-wise.” Right now, it’s so hard that it hardly works well,
David: I guess by extension, you’re saying what worked for you in February 2014, even if there wasn’t other Flappy Birds wave right now, the competition of riding that particular wave would be so great that it’s tougher. That kind of tactic is kind of almost had its day. You still got to do it, I guess, but it’s not as good as what it was it was in February 2014?
Arsen: 100%. But even more so, tactics change all the time. So it used to work, and I’m serious about it. What used to work a month ago probably does not work now anymore. Now, it doesn’t mean it doesn’t work at all. It still works but it’s a combination of lots of other things. I think this brings me into a topic of ASO, app store optimization. I have so much to say on this which I know we don’t have the time. But in summary on ASO, ASO can work. But if you’re doing ASO, just the ASO tactic, it’s not going to work. So it has to be a combination of a lot of other things as well, other tactics, whether it’d be a burst campaign, whether it’d be a major advertising campaign, whether it’d be a social media influence or campaign. All things combined, and ASO will work. And I think the very logical way to understand this is let’s say you pick the keyword, and because you are doing other things like burst campaign, you are going to go ahead, improve the ranking for that keyword. So if you’re not going to do anything else and you’re just going to pick the right keyword, for you to get ranked for that keyword is going to be very, very difficult because nobody knows Apple’s algorithm. But part of the algorithms which is sort of figured out is that the download ratios and the review ratios matter a lot, not only on the actual app, but on the actual keyword itself. So that’s basically the example. If we’re just going to do ASO on its own, it’s not going to work. But if you’re going to do ASO in conjunction with a lot of other strategies and techniques, then it will work.
David: Right. So where does that begin an end with you guys? Like which parts do you touch and which parts don’t you touch for a customer?
Arsen: So we primarily focus our bread and butter is on the user acquisition. And for us, user acquisition is a combination of paid user acquisition and social media influence acquisition. And on the latter part, basically we go ahead and we have a lot of great relationships with a lot of social media influencers, a lot of great YouTubers, a lot of people on Twitter who have millions upon millions of followers and subscribers on YouTube. And we have lots of deals worked out to help promote certain apps, if they’re high-quality, and if they fit into that social media influencer tactic. So we use a combination of both of these methods to help promote the app. And then by learning from there, we will be able to go ahead and drive the right users. So just to give you a very quick example, so let’s say you have an app, like you mentioned, a food delivery service, we would use our powered one social media influencer who happens to be heavy in the world of cooking. So clearly, their users would be more interested in your app. So the chances of success there and getting the quality user will go higher. But we will not approach an influencer, for example, who deals with handyman stuff, because chances are those users, even though you will get them, would not be of high-quality users. So that’s just a very simple example of what I mean by optimizing always the campaign and trying to pick the right users to bring to you.
David: Right. Got it. Like I said, tell us, you know, we’ve had the one win. Tell us about the one fail. I guess, if you’d had 400 apps out there, you’ve seen a few fails in your time. Tell us a little bit about that experience.
Arsen: Yeah, of course, there was a lot of fails. But I don’t consider them fails. I really more consider them learning experiences.
David: [Laughs] Naturally, you got to take something away.
Arsen: Yeah, of course, right? So I think one fail using those same words would be where the example sort of I just gave right now, the ASO stuff. And this is how we learned that we started doing a lot of ASO and only focusing on that on certain games initially. And that was failing. Yeah, we were going ahead getting ranked higher but the downloads and the ROI wasn’t there. So that’s how we took that experience in just doing ASO. And we learned that, “Hey, if you just do ASO on its own, it doesn’t work.” And then unfortunately, we failed a lot with that until we realized, “Hmm, something is wrong. Let’s start doing another tactic and then seeing what happens from there.”
David: It’s kind of an interesting thing, you know. From time to time, I talk to a developer who’s like a single-man shop, and they’re doing their app. They completely understand the domain that they’re trying to work in. And they totally nail the user experience. But it just seems more and more, it’s actually really a full production-type capability. You’ve got to have such a broad set of skills to really sort of execute so quickly and stay ahead of the pack because the meter or the following apps would just come along so quickly and club you if they have a better execution which is seen around marketing and stuff like that. Do you think that’s the case?
Arsen: I very much do. It’s becoming so much challenging that it’s not just about the development and the product anymore. It’s not. I personally feel that the product and the app is only probably 30% of the whole bigger picture. There’s a whole entire launch strategy which has to come into place, a marketing strategy. It’s very sad that you see a lot of great apps with amazing ideas and even amazing execution on the app design and development, just failing because they miss the other pieces which need to happen. So yes, I agree very much. It has to become more of a full production type of thing. And that’s why it’s very unfortunate. A lot of indie developers sort of fail before they even start just because they may not understand that they need a full-fledged production as you mentioned.
David: Yeah, it’s kind of like the old Kevin Costner film, “If you build it, they will come,” is kind of hopeless optimism, really. You’ve actually got to make sure that everything’s marketed. And I’m really interested in this concept of like the launch strategy. So there are specialists in the Bay Area who just focus on just that one thing. Basically, they come in for the project. The launch is going to happen. And they actually make sure all of that stuff happens in concert as opposed to kind of like sitting around the table, spitballing a few ideas and just trying. There’s kind of like a tendency now to go out and find somebody who actually knows all the dials and all the knobs to tweak to make it really, to try and optimize it the best way possible.
Arsen: Exactly. And you can have the launch strategy. I’ll give you a quick example. We’re just speaking with a developer last week actually. And they developed an app. They’ve been working on it for the past six months. It’s actually the whole team, and so, “Yeah, okay. It’s going to get improved by Apple this week, this upcoming week. And we’re going to launch it two-three days later.” I’m talking to them saying, “Wow, that’s not the right way to do it, right?” You don’t go ahead. You know, it’s great idea. It’s going to get approved. But you know, launch it two days later. You have to do the right stuff. You have to give the right PR. And it takes a few weeks to go to do, even when it’s approved. So yeah, and unfortunately a lot of app developers feel that once they get that green light from Apple. That means they can launch it the next day. But that doesn’t work.
David: Yeah. So tell us about in your experience with games. Probably what you’re doing, you probably had a very familiar footprint in regards to the reliability of their app and so on. I think a lot of that app developers who are kind of like, you know, you put the heart and soul into something for six months, and they’ve really tweaked it, but they’re still not sure whether it’s going to crash when they actually release it into the world. Do you have any kind of perspective on self-launch versus big bang chart launches for somebody who’s basically doing their first app?
Arsen: Well, first of all, they should test a lot. And I’m not referring to test, only installing it on their phone and maybe their friend’s phone. I’m talking about use what Apple gives you, which is the test flight program to get a lot of your friends to test on different devices. And I’m not only referring to testing, making sure it doesn’t crash. Clearly, if you crash it then we have a different problem there. I’m more referring to testing the user experience. Because what may work for you and your friend who’s developing the app, being nice and dandy on the user experience, it may not work for the general public. So you definitely want to test a lot on the user experience piece. So yeah, in short, just continue to test it out very, very well. Next on the soft launch, one of the strategies which I like to go ahead and do, and this is very common, is if your launch country is the US, don’t launch it in the US first. Launch it in the smaller country first and see what the quality of the downloads will happen there and your reviews which will happen there. And then launch it in the US. So that’s typically what I would like to go into. I do smaller test launch markets.
David: Great. Okay. All right. Is that kind of wrapped up in your service or is that just something you’ve done in yourself in the past?
Arsen: It’s something we’ve done on our own. We also provide that service as part of a launch strategy.
David: Yeah, okay. All right. Cool. Well, I think this has been really valuable. One last question and it might be off the wall, I don’t know. But I believe iOS 9 is going out today. Do you think that you should basically stop supporting versions like iOS 6 and so on now? I think the penetration is less than 5% or thereabouts. What should developers actually focus their energy on? Should they focus on high-value experiences for users who are on 8 and 9? Or do you think they should be backward compatible as much as possible?
Arsen: Well, I’m for whatever Apple says. And Apple says current minus one. So if current is going to be iOS 9 tomorrow or in the few days whenever they release it, I would say iOS 9 definitely and iOS 8. But it all depends on the app and the community which they want. For example, if the demographic and the community of the app is a different country which they may still be on iOS 7, then you’re going to be forced to do that as well. But it’s sometimes very challenging because some of the features which you want to go and use is specific to iOS 8. And now, you have to do double. And sometimes, like a simple example is storyboarding. The storyboarding processes in iOS 8 is kind of different than it is in iOS 7. And you sort of have to write things twice. So it becomes challenging there. So the rule of thumb, I always like to go and say what Apple says is current minus one. But it all depends on your community and if your demographic-
David: Yeah. Like in Indonesia, for example. Gingerbread on Android is still really large there.
Arsen: Yeah, exactly.
David: It makes it pretty tough. All right. Arsen, thank you very so much for your time, or I should say, thank you so much for your time. [Laughs] I haven’t had a coffee yet, so that’s my excuse. [Laughs] Yeah, I really appreciate your input and look forward in talking with you again soon.
Arsen: Thank you so much. Have a wonderful evening there.
David: Okay, mate. Take care.
[sh_tag id=”dj second tag test”]