Games as a Service is Dead

[Music] jon's much taller than me and so I think that's better can everyone hear me great we're gonna go fast cuz there's probably gonna be a lot of Q&A here if anything doesn't make sense save it for Q&A and please fill out the reviews afterwards it means a great deal turn your phone's off if I hear it I'm gonna single you out and she John said games as a service isn't dead and he's wrong for the vast majority for the vast majority of view of the audience oh it's definitely dead like 90 percent 95 percent of you it's totally dead so we'll talk about how to adapt to that reality before I start I'm like the CEO of iron horse games or publisher and by we I mean me my infant son and my darling devoted wife we've generated 1.2 million installs across seven titles

since December most of our titles with one exception are in open beta and I'm gonna be talking about that a lot today and we have one title that's launched you can find it right now in Google Play it's being featured idle sort 2 it's awesome we started publishing games in December so we've done a lot of organic traffic in a short amount of time and I'm gonna try and do my best to explain how we've done that well the title got your attention that's why you're all here games as a service is dead if you're trying to do games as a service once through life you can still apply the methodology with some tweaks that I'm gonna give you you just have to do it before you globally launch this presentation is largely about how to maximize installs and open beta and what metrics to focus on once you get those installs into your game first for those who don't know games as a services the basically the operation of the title once it's out with with the goal of improving your retention and monetization metrics usually there's some weaknesses around those that folks trying to shore up once they're out the ideal situation that you're trying to

achieve with games as a service is you're making enough money from the title to support the team that's working on it and also to afford marketing it's probably a handful of you guys in here in that position but the vast majority of game developers are not concepts was first introduced in the heady days of MySpace and Facebook and also Pogo which I'm not going to talk about that much because that was actually before my time these are social web based platforms that were defined by virality and if you weren't there I was awesome like you could just hoist a game up on Facebook and boom 14 to 15 million active users in the first month for nothing no feature no marketing spent you were just getting traffic like crazy doing nothing the best games on the platform the ones that were at the top in terms of da you were about 20 to 30 million daily active users 1 to 5 on MySpace they're stronger attention driven by casual titles so everyone remembers those notifications that your crops are withering those had a two-fold purpose the first obviously

was to get you to tell your friends to come help and on whether your crops and drove k-factor on the games but the second one which is really important people forget all the time as it as a reminder to come back and play the game if you have any games on Facebook instant games you should be abusing the notifications that is the whole purpose as those notifications to get people back in the game anyway that's a separate topic altogether so why do the games thrive there well teams were operating games with massive da yous you're talking 1 million 5 million daily active users for large games small changes in your metrics could mean big changes in your daily revenue numbers marketing again was a factor but it was a small one why is it different on mobile so raise your hand if you just hosted a game up on iOS or Android and got 12 to 13 million installs and then everyone in this room is going to attack you that's right it doesn't happen and there's a much heavier reliance and user

acquisition outside of the gates if you don't get platform support and/or you don't have a massive marketing budget kind of dead in the water the crazy thing about it and the reason I wrote this talk in the first place is that games as a service as a concept is the same so we've changed platforms entirely the presence of massive views users coming into your game for free doesn't exist but we still operate games the same way we did back on Facebook and MySpace with the same assumptions that are built in this has to be adjusted for mobile before we do that we're going to talk about what mobile conditions look like now last July we eye the game called idle Empires on iOS and Google Play we're gonna look at pre and post install numbers here around the launch this first bump is us going out on open beta so we got some pretty good users there second bump is us getting featured the third bump is sadness it's just a plot just a plateau nothing Apple looks the same but with less users we

got one big bump from features don't get me wrong getting featured is awesome on both platforms without it I don't think many of our businesses would exist but it is a zero-sum game so this is what our money looked like on Google Play and Apple as well so basically if you get installs you're making money if you don't get insults you're not making money as simple as that so what does that mean for games as a service well it means you get most of your installs and most of your money early and I mean real early first couple weeks first month of the game being launched what's not being shown on those charts is the fact that you're early cohorts are your most valuable so folks think that oh there's just an endless amount of users who are interested in my game out there and I just need to find them it's not true it's not true at all like there is a finite number of users that are into your game and the further that you get away from them the less valuable they get they're just not into you re features and now a lot of folks in here are like oh I'm gonna get a refuse so

screw you Mike and much larger than yeah well joke's on you the seven to ten percent of the initial installs you saw from your first feature nothing it's really a small amount of installs you need to be in a position where you're playing games as a service before you get live to capture the maximum of value for the maximum value from users that you're going to get so there's a silver lining we're actually in the best time ever in terms of two and access to potential users in open beta so I've been doing publishing I don't know six six years something like that and I've been in free-to-play for a bit longer than that and having access to these users is kind of it's a it's it's a great opportunity right now and it's better than it's ever been Google Play open beta is a great source and kind of the only source for these users that exists on mobile today if you

take a look at any of my games that are out right now and Google Play they're all in open beta and there's a reason for that it's because I know how to get installs there I'm gonna tell you how to get installs there and what to do with them to prepare for your inevitable launch and all the money that you're gonna make or some small subset of you will so why open beta folks who don't know open beta is a state that you can put your game into on Google Play when you're in open beta you can designate the game to be available that anyone anywhere that has the link to download it what a lot of people don't know is you can still rank in keywords when you're in open beta which is awesome you can still rank in for instance an idol soar to we ranked I think 1 or 2 an incremental RPG and we've done so for almost the entire duration that we were in open beta you still get listed and store listings so this is something that a lot of folks don't know either you can still be listed in games that are recommended for you games you might like simulation games the benefits are you don't have ratings so your game doesn't get a

rating which is actually good because you're gonna be changing the game quite a bit in the early days and you don't want to be saddled with a low low rating as a result of those changes users being upset with you for all those changes and finally you don't have a chart position which sounds bad but wait there's something on Google Play called new free games if you don't know what it is it's a special chart that exists on Google and it doesn't exist on iOS you were there for 30 days once you release globally in a specific gia so if I were to release today in the US I'd be on this try for 30 days your position on the chart is determined by the velocity and volume of installs that you get you do not I repeat you do not want to be on this chart if your game is not to maximize the value that you're going to get from the installs on this chart so the chart on the left is what it looks like when your game is found found quote-unquote by Google so one to two weeks after you open up into open beta

Google will find your game and start putting into category lists and those the list I was talking about before and I'll detail that further so you have two weeks two weeks to prepare for these installs the question is what do you do prep for the algorithm to the best of your ability by improving page conversion rates optimizing your ASM and so I'm gonna go through these in detail but essentially you're looking at optimizing the store art for your icon you're localizing your store page descriptions screenshots a be testing screenshots and working on your preview and there's a bunch of kind of tricks and catches here and I've done my best to detail them all as we go forward so for those who don't know a store conversion rate is how effectively your page on the left on my left you can see idle sort - that's our sword it converts users at a rate of 15% which is okay I've seen numbers that are much higher on more casual games you can look at 40 50 60 % for your conversion rate and on much more hardcore games which with a

limited audience that can be as low as 5 to 6% so I contest - the first and most important thing that you can do on your game and you need to test big variations here so for instance the skull that's showing on the icons if you tested those against each other you're really not going to get too much statistical significance between the two they're almost identical it's big variations that are representative of what your game is about if you think about if you're misrepresenting your games through the icon people clicking on the icon are gonna hit your landing page and the jig's up they're gonna find out what kind of game you've got so you're just gonna suffer in terms of conversion rate so don't lie tell them what the game is about and put forth some interesting marketing art for them to look at we run a B tests on our icons and yielded a 24% increase in our installs as a result this is no joke it's real installs it's a relatively minor change that you can do it's from an art perspective you need to rotate these every two to four weeks if

you're aggressive you should be doing it every one to two weeks the way you should think about it it's been a lot of year probably too young for this but there are department stores department stores have display windows like Macy's or Bloomingdale's when you walk by them they put attractive interesting setups in their store window to get you to come in and even if they create the most amazing alluring store window it's going to get old after time it is them marketing to you to come into the store your store is the same thing it is a marketing asset even if you're not spending a dollar and user acquisition marketing as far as your store is concerned and last but not least retest your old icons old icons don't remain old kite icons for long just because they didn't perform in the past doesn't mean that they won't perform in the future so rotate those in so a feature graphic this is another important one so we've run tests and seen ten percent increase in our installs if you think about it that's the first thing you see when you hit the page is the feature graphic which is the background to your preview video and the icon those two

need to work well together but it is also a big marquee image and that you see at the top of your page should rotate this every two to four weeks out of preview so we saw a two percentage point lift in our conversion rate just by adding a preview video these aren't that expensive it's roughly three to four K if you don't have an artist on staff that can do it for you you can always talk to a marketing company I'm sure there's a ton of marketing folks in here right now kind of sneaking down in your chairs but they will make one for you literally they all have these SDKs you're all getting emails about it constantly to integrate their SDKs ask them to make a marketing video for you nine times out of 10 it's going to be better than anything you can do because they are a marketing company and this is a marketing creative you can rotate these scenes around for a quick refresh as well so basically what you can do is take the last scene and put it first and your first scene and put it last and that is a new video at least as far as short attention spans platformer concerns there is a catch so

this is one of the weird peculiarities of the Google Playstore so on Google Play if you don't have a preview video there are sections of the store where your game will not appear period you'll never get there you have to have a preview video to be in these sections of the store so whenever you're running an a/b testing it looks like it's kind of like and we're running this test it looks like it's about even between having a video not having a video why would I do it you do it because you get this invisible bump in your impressions that's not reflected in the a/b test and that's real localized your store descriptions and screenshots again this one's important it's cheap and it's easy to do you don't have to localize your entire game it's not a hugely popular sentiment you're gonna get some customer service complaints that you're not going to be able to read but the gist of them is like you tricked me but they downloaded the game and they're probably enjoying it so 6/1 way a half dozen the other you can see a one to two percentage point lift in your CV our

conversion rate per geo for description localization same thing for your screenshots it's really easy to do super simple I used one sky I think they're good they're certainly fast and cheap I only speak English so I guess that they're doing a great job in terms of translation I haven't got any complaints oh one last bit look at the countries where you're getting the most installs and really only do it there it's not worth doing it all over the place you only want to do it in countries where you're getting a large amount of impressions and instance maybe tester screenshots so we saw a 5% install gain here this is really similar to icons don't misrepresent your game test big variation you should rotate these every couple of months and retest your old screenshots again like icons they don't remain old for long this is a really interesting part of the Google Play Store so icon are the screenshots by far are the things that matter the least in terms of your creative and the reason is if anyone has an Android phone and you

open it up and you click through to a page like you see the feature graphic you see the preview you see the short description the title you see what categories are in there see the icon you do not see screenshots those are all below the fold so they're less important in terms of the ability to increase insulin conversions this is a fun new one that we found out two weeks ago so even if your game is in portrait you should add landscape screenshots sounds controversial but again this is one of the tricks of the Google Playstore very similar to what you see with preview videos there are sections of the store where you will only be featured if you have landscape screenshots even if your game is in in portrait mode we even with with that in mind we saw seven percent increase in our installs we got more impressions from the Google Play Store it's super easy to do you just package one of your portrait screenshots with some landscape art Bob's your uncle you're done so no time conversion rating

this is the sad part these go down constantly people are bored of you and your game immediately like almost immediately you are no longer cool just like that and your job is to to the best of your ability keep these numbers up you want to get as many installs in as possible during your open beta periods so that you can improve your metrics and actually make some money during open beta as well the only way to do that is to keep keep these numbers up metadata matters so I've had a couple conversations with folks who say store description title don't matter you're wrong so on the Left I funny I'm talking to you it's right here in the front row and on the left is the developer console page that lists all the information about the game sets title short description full description on that in conjunction with your bundle ID and the update text you use form the word bank that is used for your keywords so when

people search on the store and your game comes up it's because they searched a term that's heavily indexed for your game all those terms come from this page here so testing your short description is really important 5% left of installs with words super easy but like everything else on the Google Playstore there's a catch so if you're testing this you not only want to pick the one that is driving the most installs for you you want to pick the one that also has in key words that you want to index it so you have to dance this carry this careful dance where you're making sure that you're picking the winner but you're also picking a winner that helps you index strongly and the keywords that you care about so bundle ID there's a lot of indications that bundle ID is one of the most important factors in their keyword in metadata like the the items in your actual bundle ID filter through into searchable terms about your game

what a lot of people don't know is once you upload your apk for the first time that's your bundle ID can't change it after that one and done so you need to be thoughtful about what you're naming your bundle ID if you look in URLs on the Google Play Store you can see people are doing some really cheeky stuff so you'll see like minecraft tycoon Pokemon fight whatever be careful around Terms of Service with Google Play I'm not going to go through the Terms of Service in detail suffice it to say you should read it and know it before you start messing around with your keywords because if you're found in violation your games gonna get taken down at a very least your update it's gonna get rejected from the Google Play Store so here's the fun part when you rank in keywords before the Google Play algorithm kicks in you can see an increase in 50,000 installs over a 30-day period period it's awesome so getting this right finding keywords that are not only representative of your game but also index high in terms of search traffic is super key and you want to

make sure that those are worked in multiple times to all of your meta data without violating Terms of Service on Google Play that's really important and for the love of God just just get specific like if you're doing game and free and you mentioned those a hundred times unless you have like the most generic Orang game on the planet it doesn't matter you can be first ranked on all those nobody's going to download your game because it bears no relationship to what they were searching for in the first place when they were searching for gaming free I can almost guarantee they're not looking for Hardcore card battle game that's not what they want so the Google algorithm kicks in the moments at hand let's get some installs so on the left is what the Google Play Store page looks like and there are all these lists that get populated on the store and they're populated by your download behavior and what you're doing when your game is added to those lists one to two weeks after you release in open beta you get a buttload of installs that's a industry

and scientific term we're talking thousands more installs a day we went from a couple hundred to seven to ten thousand at our peak we saw a huge jump in our keyword ranks so your keyword ranks are driven by not only volume but velocity so it's really important all that metadata stuff we just talked about that needs to be set up before the algorithm kicks in for your game because it creates this virtuous feedback loop right so you get all these insults from google play that pushes you up on key word ranks the keyword ranks and installs drive more installs from Google Play and you get a crazy mad offenseless it's really great when it happens it's disappointing when it doesn't you get much faster a/b test completion so you can a be test in a day or less and it doesn't last forever you got 30 days so this seems to be the magic number on Google Play for whatever reason you have 30 days from the moment this happens and then the next 30 days you're going to be 50 to 75 percent less impressions so you're getting insults

where you focus on average revenue per paying user is I'm gonna make the argument the only thing that you should really have your eyeball on the reason why you can change it quickly so average revenue per paying user is fairly straightforward to change you need to make a first and foremost you need to surface offers that are meaningful and valuable to users if you can't be bothered to surface the awesome thing that you put in the store don't bother putting in the store nobody is just gonna stumble into your store and find it put it in an icon that pulses on the main screen so people know that there's cool stuff to buy let them refuse your offer as opposed to just making sure that they ignore it why not go store it's much easier to change their attention I'm a strong believer that your retention is a reflection on the genre he chose the art style that the game is in and your game design if you're trying to change any of those three things at this stage of the game it's too late make a new game your time

is much better spent making a new game I've seen teams go down the rabbit hole trying to fix retention and they never come out the other side if they do they have nothing to show for it just wasted time and money conversion rates largely kept I know that some folks in here and think that they've got it's it's perfect monetization strategy you're totally gonna like plough through have a six or seven percent conversion rate on a daily basis it's not gonna happen like your early cohorts maybe four to six percent and it's great when it happens but those numbers always come down one to three percent is the range that you're gonna see for converting users and the spenders average revenue average revenue per paying user is the only thing you should be thinking about at this time so what do you do now you need to prepare like you've got one shot and making money because you really do I mean this is it you need to set yourself up so that you're getting as many installs and fun as possible and open beta you need to work on your store landing page that increase those

installs before they actually kick in and you need to raise your art poopoo that is the new games as a service and it's the only one that matters for the vast majority of people in this room and the vast majority of games you're not going to have an art poo that's gonna allow you to spend unlimited money on marketing or continue to operate or afford operate your game it's just not going to happen and that's not has nothing to do with your skills as game developers those are things that could be outside of your control like the genre you've chosen like an endless runner with rare exceptions and I'm sure I'll hear about those rare exceptions during the QA but the exception to the rule is not the vast majority of developers an endless runner is never for the vast majority of people going to be in a position where you can continue to operate as games as a service finally and this is a really important one if think about yourself from the Google Play side so Google's trying to think well which games are gonna feature this week and they look at Steve's game and they look at pun e's game Steve's game steve has extensively used

the google play toolset it's a be tested like crazy he's really optimized to the store conversion rate punny didn't do any of it funny just kind of threw a dead body under the Google Playstore I was like oh that's let's see what happens Steve wins Steve wins that every time because he's shown effort he's shown that he cares about using the store in a way that is meaningful to Google and using the tools that they pull that's it [Applause] does anyone have any questions yeah oh yeah I won't yell at you um I may be short with you if it's a bad question because there are there are such thing as bad question I hope you don't have one we will know soon what's up first of all I thanks for the talk thanks for coming yeah kudos for the quick bait title that's awesome I think

of half of the audience here came to prove you wrong so what's up I mean it's what are you really saying is that developers should stop trying mobile or no and I'm saying that there's a there's a finite window to make changes in your game before it gets live right like you if you're in a fortunate position to get a feature or even if you're not you're still gonna get it I spike of installs in those first 30 days when you're out on google play or ALS right we've all seen it your release and your first 30 days are actually pretty good your game needs to be in a position before that happens to extract the maximum value right like the operations the this sort of operating your game like a service or business all that's got to happen in test market for you to extract any value at all right if you don't do those things and you wait you're never gonna be in a position to afford to do the once really yeah that's wrong I just wanted to add one thing that's right if you're like of course one of the top

grossing games it's not that at all you can keep doing profitable your way yeah but I've seen also some small developers finding some niche well while they cannot scale indefinitely they can do profitable way that's right way so I would advise looking on ages as well I think those things are dependent on your capitalization for marketing if you have the funds to market or if you can generate the funds from the game you should you should definitely do it it requires a pretty specialized skill set and there's as you said there's a cap to how long they can do it for and how many installs they can acquire me it's possible and there are people in the room who add those types of games right they've been able to effectively operate them over a period of time but they're the exception of the rule as opposed to the vast majority of people aren't in that spot Thanks sure yeah brother good question mother hey what's up hey thanks for the frank talk and actually giving us numbers helpful yeah I'm curious if you have any tricks or tips for iOS No get lucky no so iOS is tricky

right you shouldn't be doing test markets there but you you have to do user acquisition really to get any scale of users in there and it can be expensive I've recently tried a pre-order so if you get featured for a pre-order I've seen devs quote like 50 to 70 thousand pre-registrations free game which is pretty rad the challenge with a pre-order is you can't do a test market so I hope you don't have any bugs because you're just gonna launch blindly on iOS with no testing with real users I Oh s is kind of a different animal Google built out this this tool set and it's pretty great iOS is still a great platform they have really valuable users and a lot of them they just don't have the same sophistication of of testing tools cool thanks for your question Oh first yeah all right oh sure I just wanted a tiny bit of clarification so you said when you launch an open beta

there was about two weeks before things like index with Google and then you get like a download increase so that's outside of whatever download increase you would get when you go out of open beta yep to launch that's right okay thanks I was wondering if you were seeing any significant deviations from one market to another like were there any that particularly stood out for you and also with I don't understand can you ask for example if you choose to say launch us only and but then like you go say launch Japan and or you launch pick another market right and as you're looking at this data are you seeing any differences in the drop-off rates and the retention rates but yeah Americans are the best I mean if anyone's looked at the revenue numbers Americans are paying the bus they usually retain in terms of total revenue

or in terms of both in terms of mean it depends if you've looked like if you if you have made your game culturally relevant in Korea and Japan and China there's an argument that you can make that your game is gonna really thrive there and there's tons of examples of that but by and large most of the games that I work on really resonate with American audiences thirty to forty percent of the installs again in open beta are from Americans 50% of the revenue I generator from Americans and that's that's pretty consistent with what I see on iOS on iOS the exception is China which is really come roaring into the picture in the past couple years both for premium games they're a great market for premium games but for the love of God don't price them like you would for an American like there's a currency there it buys different things in China its price it accordingly but yeah China China in the u.s. make up the bulk of revenue I see on iOS because the reason I was curious because the

games that we've we've been building games in Japan for almost twenty years now and what we see on the deltas are significantly higher on the Japan side than the US and the app set I've worked with in the US side right attended flop well let me know if you want to operate and make changes on my game in Japan and let's start with you that's what we're doing but just give me your card afterwards Thanks awesome hey thanks for the talk really really good impressive Thanks all I have two questions actually uh-huh one of them is gonna be part advertising sorry I have to root for my homeboys so what do you actually know what actually do for a be testing is it native Google Play tools or you go like third-party developer native Google Play tools have you ever tried the guys from Laos and I do the cheapest thing possible because I am small and there is great product John and Belarus is called split metrics all of you guys speed

matters yeah yeah really really go guys absolutely no but I some I'm not from these guys oh yeah I am from Belarus that's why I'm getting for my phone voice all right cool I'm sure it's trying but second one is closer I come eSports industry so perhaps you know that the games as a service if you add a competitive now feature no eSports is the worst oh my god you know exactly what I'm talking about sorry we gotta wrap this up a holy eSports is the worst did your level rate a little bit yeah I mean it's it's matchmaking based on skill and all of the monetization that allows you to buy things that make you stronger in RPGs can apply because it base it breaks skill based matchmaking so instead you have to say you have to sell items that are like aesthetically pleasing and unless you have a massive user base which you have to acquire because it's just not that popular in genre you're never gonna do it alright sorry these

sports are the worst I'll talk to you outside [Applause]