The Secret to UA in APAC

Peggy: Hi, I'm Peggy Anne Salz here at Mobile Growth Summit. We have been taking opportunities to talk to some of our speakers, some of our sponsors. And in this case, we have Robert Garfinkle. He's Senior UA Manager at Nexon M and Robert, first of all, great to have you.

Robert: Thank you for having me.

Peggy: And Nexon M, I mean, I know what it is because I'm following the industry but some people might not. So tell us a little bit, first of all, about the company.

Robert: So, Nexon M was founded in 1994, in Korea. Nexon M is the publishing arm in the West Coast of the United States. We're, of course, headquartered in Korea still. Nexon M is really well known for Maple Story. And we originated the free-to-play model on web and PC in about 1995.

Peggy: You did that? You originated it?

Robert: Well, I mean, not me.

Peggy: Not you but okay.

Robert: But our company was the deployed what we would call now in-app purchases in a free-to-play online experience. But that's the history of the corporation. What we've been doing for the last number of years with Nexon M is that we're trying to bring Western expertise in publishing and in gaming to the Eastern development might that is the Korean gaming developments.

Peggy: So building those bridges in a way?

Robert: Yeah. I mean it's been done. Well, it's been attempted quite a lot of times rather famously with GRI. There's a lot of Bandai Namco. There's a lot of other very large Asian publishers that have made serious inroads in testing or in bringing their brands westward with really notable... Nintendo comes to mind. We're not quite on the same maybe level as a Nintendo in terms of brand awareness in the West, but we have a very strong presence in our home markets.

Peggy: And you also work a lot. I mean, you're traveling in both directions, right? Spending a lot of time in APAC?

Robert: A lot of... Well, one of the great things about being in the Bay Area is that you're surrounded by the most advanced technology in the ad tech world, which is, of course, what user acquisition is. It's all advertising technology. So, because we see the newest stuff coming in and we are able to test and we are able to learn things at a more rapid rate than that some of our contemporaries in Asia, there's a lot of trips back and forth to pass knowledge to bring what expertise we can to the more global offices.

Peggy: So, what is that expertise? I mean, what is it? Because I would imagine that, well, yeah, okay, there's a lot to learn from us. There's a lot to learn from Korea and across APAC. But what is that exchange really focused on? What is it that we have to offer that's so amazing, that is what needs to be integrated into, so to speak, the formula for success?

Robert: Well, the formula for success is maybe a broader topic and a large...

Peggy: Okay. You also did a workshop on it in a way, so we'll get there.

Robert: Yes. So I think what we share and what we communicate across the ocean as it were with is in building more advanced modeling, bringing creative knowledge and expertise in terms of the kinds of ads you build, not the actual output themselves because, of course, there's different aesthetics between the West and the East...

Peggy: The localization.

Robert: But in terms of the way that you approach these things where, you know, there's a heavy focus on cinematic assets, cinematic ads in Asia, whereas we are much more gameplay focused. And so we're bringing more of the gameplay elements into the video as well, maintaining like a lot of strong, you know, converting assets. And what that means ultimately is we're able to leverage more advanced media-buying methodologies, more advanced ways of bringing this technology in house in markets where historically they've been much more agency-driven, much more hands-off. And it's just kind of you sign the check at the end of the month to the agency who is running the campaigns. We're trying to bring the expertise in house for running Facebook, running Google, running all the ad networks.

Peggy: And also the platforms are a lot easier to understand. I'm hearing that a lot in the industry you know, moving things in house, it just makes sense, not just even from like cost-effectiveness, but you just want to have control over the data. You want to be able to move quickly. That's what you have to do. You know, you have to change these creatives up very quickly. You gave a workshop as well here at Mobile Growth Summit. What was that about?

Robert: So, the workshop's topic was understanding good CPI. And CPI, of course, cost per install, is a major KPI that is paid attention to by everyone. And I walked into the room and explained that CPIs are kind of artificial. They're the cost per install involves your creative conversion, which we call Installs per Mille, just to keep everything consistent. So IPM and it's cost per mille for a thousand impressions, which is what the network gets paid out on. So the media has fixed-ish cost. It fluctuates a little bit, but it's mostly kind of the same. And the creative conversion varies quite a lot. So that we focused a lot on how creative conversion impacts the CPIs that you're seeing.

And we built a little chart of basically how core or how casual a title is along an access of IPMs in order to show or illustrate basically potential CPI. So even in a circumstance where you are buying flat CPM, which isn't advisable, but if you were to just spend $20 for every thousand impressions, hypothetically, there's a very different real cost per install that occurs when you're getting 1 person out of those 1,000 to install the game versus if you're getting 20 people of those 1,000 to install the game. That actually results in a very large swing. We're talking about a CPI of $20 at the one end versus a CPI of $1 at the other end. And that is what I think the fundamental shift in our thinking when it comes to how CPIs are discussed is what I wanted to promote with this workshop.

Peggy: And also, the CPI as a metric, is also very different. You know, we talk about CPI. I remember back in the day when it was just all about, I just want to buy something cheap in the Philippines and get my data and do my thing for my soft launch and then move it on. I mean, it was all about almost this thing about just pay attention to the cost, but you're also saying you have to consider the conversion in the cost.

Robert: It's more than even that. One of the major points in this whole conversation is that there's no really such thing as a good CPI, except in so far as the CPI is lower than your LTV, your lifetime value. So if you have LTV of 24 cents, even if you can drive a 25 cent CPI, that is a bad CPI.

Peggy: Got it.

Robert: If you have a lifetime value of $50, any number lower than $50 for a CPI is a good number. And that is a fundamental understanding that I think maybe gets missed sometimes. It's obviously part of the grander formula when we're talking about games and the business around them because you do have to bring users in and you do have to pay oftentimes in certain markets for these things. But there's a right price for every user. And it may take some time. It may take a lot of data to really come to that. But we do media buys in every region around the world, except for the core APAC markets, which, of course, are managed by our Korean and Japanese and Chinese counterparts. And we do not buy in Iran because we're not allowed to.

Peggy: Okay. I mean, you did a workshop that's all about sort of like knowledge sharing. Was there something, maybe someone asked a question or maybe there was some exchange where it was like, "You know what? This is very interesting," even for you as giving the workshop.

Robert: Sure. So, the second half of it was very focused on these case studies, these kind of prompts focused on getting people to think about how you manage up expectations because oftentimes the pressure to reduce CPIs or to focus on CPI is coming from someone who...

Peggy: Wants to see it.

Robert: Exactly.

Peggy: Balance the budget.

Robert: They want to have an easy number in a checkbox, right? And that makes perfect sense from a certain perspective, if you're looking at the finance or if you're looking at this or the other thing, but there's a much more dynamic approach. So there were three case studies, which were I prompted and then we had a break for two minutes and people kind of shop talked amongst themselves and then we all kind of discussed it.

Peggy: It's like on-the-job training. You had to like figure out what you're going to do in that situation. Cool.

Robert: And a lot of people share some decent experiences that they had, some frustrations that they had, in negotiating upwards with executives on this or that level of understanding. The room's general grasp of the material was very high. And a few people focused very heavily on kind of like the way that you test things or they focus more heavily on creative conversion. But I was very satisfied with the amount of audience engagement that I got.

Peggy: Anything you learned at Mobile Growth Summit for yourself? I mean, you know, I mean, you are advanced, obviously, Robert. But, you know, it may be that something stick, something you're gonna check back to the office perhaps.

Robert: I mean, I've met with quite a few partners that I think are rather promising. I'm interested in seeing the growth of this kind of bespoke or interesting platforms like Missed Play. I'm very interested to see how Reddit rolls out their install product, which they recently released. And I'm very I had a good meeting with Adjust to talk about their most recent addition to the fraud suite, their click verification and depression verification, which I think has a really good opportunity to bring better technology towards the fraud prevention side of the business, which is, of course, a very large concern for any publisher.

Peggy: So this will be the third time I think you were at a Mobile Growth Summit event, right? The thhird time?

Robert: Yes. So, I...

Peggy: What keeps you coming back? Not just because we ask you to speak obviously, but...

Robert: Yeah, that helps, but I keep coming back because I find the content rather compelling and I feel like as a participant in this economy owe a certain amount of my expertise to improving the discourse. I do believe that while we all might be in competition with one another for an audience, those audiences vary and they differ, and if you make a good product, you've made a good product. Ultimately, the level of the user acquisition philosophy that is impressions are verifiable and performance can be tracked and you can build a business off of advertising in a certain sense. This philosophy deserves promotion, deserves to be brought to the fore.

Peggy: That you can really monetize with advertising, which is true.

Robert: You can.

Peggy: I mean, and it's also thinking about, you know, data at a user level. I know that so you can segment. I mean, are you that far in your thinking, you know, how to now approach the next big step to be more cost-effective or just be more efficient in your spending? Is it about segmentation? What's the next step now, once you get your head around the right KPIs?

Robert: I think so segmentation is very important, especially when it comes to the social channels where you have lookalike campaigns, right? So you have these algorithms from the major social networks that allow you to find people who look functionally or demographically similar to an audience that you designate. And those have been around for a few years and I think that they are some of the most impactful methods. But I find that what we're moving more towards as an industry is an automation, automation of bidding technology, machine learning within the way that you are engaging with the media that is not purchased in such a direct audience fashion. So you have a network with, let's say, 10,000 publishers. That's just a number pulled out of nowhere. Of those publishers, they're the long-tail formula famously written down and noted by an Amazon employee, a book, a great one.

Peggy: Chris Anderson. Yes

Robert: Yes.

Peggy: The book, "The Bible." Yes.

Robert: Yes. That states that within that 10,000 publishers who are serving ads, only about 10% or less will have 95% or more of the overall volume. So you need to find a solution for well, actually, it's more like 60% of the overall volume, but it is the only ones that can be really measured in a lot of ways because it's the only groups that are giving you enough installs for you to accurately gauge the performance. Sorry.

Peggy: No. Sorry.

Robert: But just to continue my thought...

Peggy: I was worse. I thought it was someone I needed to interview. Okay.

Male: No, I just need a battery.

Peggy: No problem. No problem. I thought you were the guy who was waiting. Okay. Okay. We're good.

Robert: Sorry, that's is distracting me.

Peggy: We're good.

Robert: So what I was saying was about the long-tail, right? So there's these two approaches. You can build a system that automates the front half, right? You can find a way to like identify key performance indicators that guide a bid, either automated or non, for those, right. You can adjust those on a day-to-day basis and find like your perfect value for each of those. And if that competes on the waterfall and the CPM is high enough, then you will continue to receive impressions from those. The problem is the remaining really large chunk of publishers...

Peggy: You're ignoring it.

Robert: The long tail.

Peggy: Yeah.

Robert: These are oftentimes a very large portion of the overall installs. Like 40 to 60% are in this one to three publisher like bracket and finding methodologies to address that segmentation.

Peggy: Yeah, because otherwise, you're not getting the most out of it. You can't just broad stroke...

Robert: Exactly.

Peggy: Approach. I get it.

Robert: You want to be absorbing a certain portion of them. You want to be bringing those into the campaigns because that's your growth potential. Also, there will be ones that become a higher percentile like overtime or through featuring or through [inaudible 00:15:23].

Peggy: So that would be what you have to do through AI. That would be a way to get literally more bang for the buck out there.

Robert: And so there's a number of different algorithms that are becoming more, well-used, more mature, including I appreciate DSP, which is a self-service algorithm that allows you to determine how much of the long-tail you kind of want to let in on a regular basis. There's a number of others. I think that this year we're going to see this technology really coming to the fore and becoming much more impactful towards user acquisition.

Peggy: It makes sense because we want to get more out of what we have. You can't ignore that. I mean, you gave a workshop and you share freely, which is what I love about talking with you, Robert. So if you could end on, you know, a growth hack for 2019, maybe something we're not watching. It doesn't have to be so simple. It could just be like a trend or something to keep track of. I mean, we talked about AI. We talked about looking again at the long-tail, thinking about segmentation differently, rethinking CPI, which was very smart in your workshop but leave us with one more.

Robert: So, I don't much like the term growth hack.

Peggy: Okay. No one does really does. [inaudible 00:16:34]. I know, but okay.

Robert: But what I would say is that what we're seeing in emerging markets is an extremely fascinating and extremely large opportunity. So, you know, we are all familiar with the lessons learned by the growth of the Chinese app market, which is that early movers who got into the market early enough and found a place remained as top-grossing apps to this very day, although, over time, native players become larger and larger and in the particular case of China because of the government controls, it's rather pronounced.

However, a similar trend is occurring in a number of different markets in so far as the user literacy rate for gaming is increasing over time as more and more gaming apps are downloaded, user retention increases, user monetization, retention increases. And to the point where markets like Russia, Brazil, Turkey, India even can be extremely meaningful, not only from a user perspective but also from an ad monetization or IAP perspective. They do require heavier levels of localization. They do require a lot of consideration, especially in terms of where you're buying your media, what you're optimizing for, and how you're checking your assumptions for things like LTV because there are just different profiles for different markets in a broad sense because of that literacy aspect.

Peggy: Of course, once you've cracked it, then you can sort of move it out across the markets because the platforms are better this year. You know, at least that's very good.

Robert: Telecom is reaching a new series of heights in a lot of these countries and we're really bringing stuff to the fore. Sorry, just too distracting. We're really able to see these trends moving in a positive direction for these markets. I think we're going to see a lot more China-like markets in the next 10 years in mobile gaming.

Peggy: Well, that's great because that means this is it. Go global with your app in 2019. Robert, thanks so much for taking the time...

Robert: Absolutely.

Peggy: ...with Mobile Growth Summit and thank you for tuning in.