Maxing out your Facebook user acquisition for games

[Music] [Music] - mika-nee Kapoor sister I see a lot of familiar faces here so my background I used to I used to be building a company called amplifier I was running there bTW : we sold the unity they're running the global beauty for a while we after wish I went to a beach 7uc in the my talking tom around here so I was their CEO for a while I'm a year and a half ago I moved back to back to back to Finland started working on different things one of my clients is smartly doing Facebook marketing platform basically works with you know you've probably heard a company called eBay and stuff like that

what smartly has done in the past year or so it's been very actively building the gaming vertical business meaning you know providing these services for games companies as well and I was happy to have cybela here you know happy use of this party platform well a couple of words smartly basically today I'm on it Manus is about 1 billion dollars on their technology for Facebook and Instagram now games companies they're quite a lot you a Ubisoft of course the nor there's quite a few others well we're not today here talk about smartly really I mean we wanted to really dive into Facebook marketing how can how are some of the game companies using Facebook to kind of maxing out their user acquisition how what do they do so I wanted to get cycle I here who actually I met back at the day is adaptive fire when you add Miniclip you move to space we always get kept in thoughts and and and you know a few months back you kind of six months ago

you kind of moved to you be right yeah yeah so so we're gonna talk a little bit about her experience on launching games how did Facebook help what were the problems with Facebook and that's why we have Morgan here as well only through the problems only for the problems and he's the guy who's gonna fix them right but this is what we're gonna do to begin is a little bit about sort salt park the last the recent title you launch and also what do you learn from it so but having said that maybe a couple of words what I what are you doing now at Ubisoft well I joined you saw six months ago with a mission you know to strengthen our you a performance and expanded to something like well-oiled machine that we can apply to rest of the games so South Park is one of the launches that we've worked on recently and I guess there was somewhat testing ground but also big opportunity for us to see how far we can push with Facebook but

also with all the other networks that we watch yeah yeah and of course Facebook is one of the biggest channels that we advertise on yeah okay Morgan what's your role on Facebook so I'm in with Facebook for about three years now and I'm working on the gaming team where I work with developers across Europe to help them basically succeed on Facebook through our platform but that's also the family of apps of Facebook so Facebook Instagram Messenger for example yes than games etc yeah yeah I mean so again my experience like back of the days joined a building amplifier one of the first like the rewarded video players really like we were not the first to market what we're gonna you know one the market at the end of the day we sold the unit and and many things have changed because it's not too far as a few years back but at the time you couldn't actually do anything on Facebook for you a by the way so time has changed change quickly so now now Facebook is one of the most

important platforms now can you share anything like how like in terms of just like how important Facebook is give me some numbers I know you don't want to but like indicate well it is actually very important depending on the game for some of the games less important than for the others for southpark specifically it was quite big pickle Sparkle fan and following fan base and following on Facebook is over 44 million users oh that's it likes whatever and then taking the game genre into consideration there is so much more it's a card battle game slash RPG and so on so we would potentially target all of those too and naturally using you know all of that I was told to use machine learning work it seems to be a buzzword in Italy I heard it just right now so obviously using app event optimization and look-alikes and all that jazz yeah I

don't want to give numbers just because then they would just think too much of themselves no I'm kidding I would say depending on the game some of them could be 50% already I'd spend some of them could be less I think yeah I think it's kind of fair enough because I mean my experience like what I see nowadays when people start when they when they saw flaunting even if there's an experienced team or team starting people started with Facebook so that's like the platform people that what I thought I've seen I mean what's your so I mean you're just touch upon something that I think is very interesting is that Facebook's mission changed recently right so we're more focused on building communities and making those communities come closer together and community it's such a big component and games and you've seen it with South Park right so you have a big community on our platform and you were able to succeed thanks to the community as well and being able to leverage that for your you a so I do think importance about developing in your game and doing you a as well it's being conscious about

the community you build around that game - yeah I'm just one of them yeah yeah you guys also use seamstick run like anything you do there in terms of like getting users yes actually we do use Instagram as one of the placements Weld only we don't use Instagram alone just because for something like myth core games it doesn't really work that well maybe for casual games something I don't know ads based games it would be better but for these particular genre it's one of the placements we use it doesn't work well in every single country out there it's particular to some of the countries have you have you leveraged placement optimization so placement optimization just a kind of share on this is the capability of using the inventory across our family of apps to be able to place your ads within the and you mentioned that Instagram is going to vary according to countries etc but the advantage of using placement optimization is performance so usually when you activate multiple inventory or

platforms your performances will remain the same or get better because at every particularly time of the day a user might be more expensive on Facebook or Instagram so it's better to have more inventory but I can understand why sometimes on Instagram like you might want to have a separate strategy when it comes for example to Instagram stories which hit 300 million in in November recently daily monthly active users I mean it's great platform to be able to tell a story in addition to what you're doing already on Facebook yeah I think in addition to that we shouldn't forget about the demographic of Instagram because Instagram users younger than to commit with Facebook let's say so all of that is a minor's my nieces and nephews they're all on Instagram and yeah we take a different approach on Facebook and Instagram as well like there is

overlap in the audiences but Instagram also provides incremental audiences versus Facebook we consider them as two different communities people go on Facebook to follow their friends get recommendations great place to get your games discovered for example but then Instagram is a platform where people go to get inspired they follow a lot of strangers look at visuals it's a lot more visual storytelling than you would have on Facebook for example yeah I would say also just to share with the audience the creators it's a lot more important on Instagram than Facebook for both of it is actually important but what we've noticed the ads that work better on Instagram placement as a placement optimization is when you build creative - for the platform yeah yeah well you play with dimensions but also playing with the visuals as well and a creative team yeah and I just want to jump on that because I do think that creative today is probably the art of war is born from creative

where you really need to be able to be flexible adapted and according to results you're getting and performances you're driving creative is really the only component today that is not really like automated targeting can get it automated with worldwide targeting then you have a prevent optimization which allows you to reach valuable users on our platform through how you establish what a user is or valuable users so all of that is automated but creative isn't so it's really the part where it really makes or break performances sometimes don't you agree on the fact that sometimes your creative can just make your campaigns like perform much better than before I would say yes creative could make it or break it and targeting as well so what you're gonna do to help that that's my question I go I go to crap a coffee right I think this is kind of like going ice because while you were

talking about creators and so on I thought about that recently Facebook released is creative something to where you can look at it and then you can see which creative perform better haven't seen it yet I have not seen Oh shame yeah I'm probably in one of those buckets that Facebook where I don't get to see yeah you can actually go you don't need that smartly we can just go to your Facebook Ads manager and there's a tool where you can see creative performance you can see it on your mobile as well and mobile app adverts and things like that so because I poked around and then kind of nosey see how it works and things your hectic right I can't help it but then say okay that's cool that you build these kind of tools and it's accessible to everyone it's kind of easy to use but then with the creative optimization let's say because sometimes some of the creatives work and some of them don't but it's really hard to understand why it would be cool let's

say if Facebook took an initiative and build some sort of suggestions within within this if it's a reporting tool and explain why those creatives work either let's say this video is longer than the other and that's why and that's why these kind of videos were better on Instagram or Facebook nice be the messenger and things like that and I think it's very interesting like working with all these different game developers across Europe they the main challenge is usually the data they can get in regards to that has performances and I know our product teams we're constantly funneling this kind of feedback and this is typically feedback that we'd share with them so that they develop tools that down the road would hopefully give you the the type of data that you're looking for cool I think I have some suggestions we can talk about there's a long list of because I wish we had and maybe you guys have the power to do that so we can use it we really love

developing the tools we have and feedback is always the thing is that you're gonna already went through my 48 questions but I'll figure out something else now I'm just going back to so I think it's kind of it's kind of like going back to kind of recap or wrap it up in terms of Facebook and Instagram yeah I mean you need to be I think so my my experience on Instagram you need to be you need to think more like in terms of like let's tour is this great platform but you really need to be creative with how you kind of leverage that Popo it's not like you're just doing it right yeah so so there's a lot of stuff you need to think about but lately it's gonna let's go back to let's call the short part right you reason launched it like soft launch was last year can you share anything like what's your what was your experience like what did you do particularly on Facebook I want to use it for like you know soft flaunting would you'd love to think yes yeah I know I'm really curious to know

like the the steps you took so like from soft launch to like global launch during the soft launch actually it's been quite quick since I joined the company like six months ago in some room so it was already halfway through the soft launch so we had a very short period of time almost like four months or something to just trump it up and get our stuff together and say okay we're ready to global lunch we launched the beginning of November and once that roller coaster because that don't think we were ready everywhere but we just had to go and grab everything we have and see how far I can push it really everywhere in terms of the you know came you know elements or like you know wanted to say SH you know like you know I think it's morally way friend full months to you know rock it out put some tech behind it data warehouse then usage of networks then we added a lot of networks at the same time and then trying to prepare what while we

are trying to prepare for global launch and then some new team members joining halfway through oh how much would you prefer to have like how many months um I know there's no like there's no there's no recipe but there's like best practice yes there's no idea recipe but if I had a choice with let's say with all the tools plugged in and to have it somewhat ideal each situation I would rather have just for this testing and preparing for global launch in the best way possible probably like six months yeah operations just to have you know a few different plans I think in the end it was still a blast we have done you know we deliver success actually better than I expected maybe I wasn't optimistic enough when you say success in terms of KPI is what if you can share the KP as you were looking at what kind of metrics were you following constantly when you were doing yourself launch for exact well soft

launch we were looking at basically during software internally you a front what we needed to test we need to understand who are our target audience we need to understand where the not just South Park fans but also beyond that where do they hang out what else they like what do they do what other games they play and then see also what do they you know do inside the game do they like the events do they like purchasing particular packs do they prefer particular avatar particular game character as well and how important South Park brand to them as well because we've discovered that for some of them self our brand is not that important because they played so many other similar games that the majority a big driver for them was the game genre as opposed to South Park and then for South Park fans it was more of South Park IP more of a driver basically I mean you kind of like there's a lot of interest based targeting you were thinking about like in the beginning and they moved on to like up events right

did you also look at like multi country look-alikes or overall yeah yes so during South lunch we couldn't do multi country look-alikes like worldwide and all that so we did only soft-launched countries that we could and then of course during the global launch we just went all the way out ahlian and so did you transition from doing the default app installs to app event optimization which allowed you to maybe find these users for purchase or maybe value optimization actually yes and that's thanks to our challenger Ricardo who is later in the crown yes that was the strategy whose advice by our very valued thank our manager who's advised all these things we work in partnership to make sure that we share a lot of data but also at the same time get some sort of advice and feedback and at the beginning because we know we all know in the gaming industry when you launch the app in the new territory for the first time you have this honeymoon period when everyone is high on data and you think

you can move the mountains but not really two weeks later the reality hits and it's when the real you know muscle you a muscle is used so during the first two weeks of course we needed to make sure we get the volume and make sure that we get the day use and user base so people can use PvP mode and all these things and have fun and engage during the first two weeks of course we go for installs only targeting everything that's possible out they are not just look-alikes but also likes an interest or different game genres and so on and so forth and then smoothly straight after that as soon as we get sufficient amount of data to you know let's say approve and validate the theory that this particular event is good enough then we go with and optimization and as soon as we have sufficient amount of data to start using value implementation that's when we start doing that as well and were you able to actually measure the LTV of these cohorts before and after moving to value optimization or app event optimization for example and can you tell us a bit more about that did you

see an evolution there we saw an evolution there is a combination of two things it's what's network like Facebook can do in the combination with game performance itself so let's say I don't know not every game is highly magnetized and the same for everyone right so I would say South Park is you know IP based games and it's fine and it's with the mythical genre but it's it wouldn't say I wouldn't say it's the same as clash of clans right so exactly so that means there is a ceiling that we can hit it is better with value based optimization and with Ayoo as well but it wouldn't hit it wouldn't like you know make it ten times better just because yes algorithm so on Facebook network push it that far but at the same time we need to realize the reality how far we can push and also be realistic with that I mean we can compete with a few gaming companies out there or few

games out there but there is a there is a ceiling as well but of course during the game development as well because it's only been since beginning of November there are more features back there coming in the product development studio is working on it there are more in-game events that are coming there is also some very engagement stuff coming up so with time that would have evolved and LTV would evolve and that's when we will tweak and change and adopt so can you like in terms of you know moving to at the event opportunities I mean what specific events for you were you're tracking like you know purchase and those things are like obvious but yeah so purchasing things are obvious so for well we have what we built internally with a team at ubisoft that we have a few boss and soul events that we track with our tracking providers but also with in-game analytics and we mirror that the same for Facebook SDK because we mirrored that separately and we have five core

events that we track on all of the games despite the gameplay and everything for example first purchase to identify spenders the revenue etc etc so all these things and then after that depends on gameplay so let's say in south park we look at the data during the soft launch together with bi team and figure out which post install events can be let's say classified as best predictors of LTV so which ones are the quality ones so we can apply for our day-to-day optimization so for South Park specifically let's say tutorial complete you think you don't complete in most of the games is IKEA is somewhat fluffy even to track however when you play the game in South Park it actually takes thirty five minutes yeah so you must like the game to stick around for thirty five minutes in one session right so in that case it was strong enough but some of the other games tutorial is less than five minutes so maybe it's not it's too early and it just depends on on the game

in game let's say you are an yeah and then for South Park specifically the PvP rank so when you enter PvP rank and you try and you try the PvP mode and you try to play a few times and you achieve certain rank that's also very important indicates that okay disgusting so basically take those events from VI and then go and ask developers to implement its post install events inside are MMP tracker and Facebook SDK as well to make sure all of these events are available whenever we need to use your team and and you and your team spend a lot of time it game developers in Europe do that like help them to kind of you know attractive right defense because you know where to go but like there's a lot of companies are kind of you know because purchase I mentioned because that's the first obvious one right but there's so many typically before that depend on game genre and there's so many like you know events the tutorial come with Facebook Connect whatever so there's like different things do you need to spend a lot of time kinda helping people to gonna understand what I mean we don't know the businesses as

well as these businesses know themselves obviously but we're there to give them guidance and understanding so like when you were working with Ricardo probably helped you understand like better like some best practices or some some things that other businesses do that are successful we got ours back there saying thank you at all he's not like but like digging into data takes a lot of time so so some of us actually do spend a lot of time and helping and fixing or figuring out how to make it work better for you but most of the time we're just here for guidance because you'll be able to know what is a valuable user for you what is that events that you need to leverage in the app event optimization for example you just said that your tutorial is long enough to show that there's value if you do it in one session where as tutorials that last maybe two three four minutes is not enough or not enough meat for you to say this is a valuable user yeah it's - its rate too early in the process of

like you're not you're not really on board if you will that's a user typically like well depends not game sessions might be super short or anyway but that so there's you know again there's no recipe there's there are best practices you know the tool okay exactly and I mean if you guys somebody like Facebook because you have so much data available you might be able to see some similarities per game genre I would say there are best practices or purge on row yeah but no two games is the same and we I mean as you probably know like one best practice that you've experienced once on a game and they're gonna try and replicate it might not work so one of the best practices we have at Facebook is always testing and learning and iterating from that you've probably had a situation where you try to replicate something that worked before and yeah find similarities per game genre because yes there are differences if you go down you know drill down very far but there are you know the core things that kind

of apply to most if you applied the game sure absolutely like some things that are replicable for example are usually the the structures so like the country structure the approach of scaling right so you're gonna start off with on a soft launch on a small market or a tier 2 market test out see how that works and then use those learnings for your global launch and then scale in to markets with worldwide targeting and tearing systems if you want but then when it comes to like targeting specifically it can really change you can have a match 3 game say I want to target people that love candy crush for example and it might work for one game but it might not work for the other title you have because it's not that similar yeah yeah something else you also you were featured right in many places itself were quite a lot in many countries I don't know how many but I mean naturally thanks to South Park IP and how much work Madden traded on their

silly TV show so like would you say that that has any impact on your Facebook strategy or you a strategy that you're gonna get featured everybody wants to get featured right but yeah Karen even though you know like you can't trust like you would get so but would that change like any of the things you did for for the copper ones as a strategy I would say with that goes back to getting featured as one thing but also the first two weeks the honeymoon period and of course when when you just go to new market and luckily if you do get feature that's even better that means our strategy would be to use as much as as many channels as we can and to buy as much as possible in a short period of time so it doesn't mean that we have to bid high because we didn't the South Park to feature it was

amazing it lasted only two weeks but yeah ok I will take it anyway so featuring definitely there is no there is no proof a computer number on it and there is no tracking to featuring so we can't say but if we look at baseline and look the time when there is featuring happens in particular countries you really see big spikes I mean it must come from feature naturally during that time when there's a lot more marketing buzz there it takes more than one impression for people to download your game and try and to go through that 35 minutes tutorial right so naturally if people get comfortable and they see a few impressions here there and there and they go like oh yeah I seen that absolutely let me try that out and if friends play even better you know so it does affect and we don't put it into the you a like hard strategy as if we get target if we get featured we're gonna do X and if we don't get featured we do it's like if we don't get featured in

some some countries but we still you know a South Park IP is strong we will just need to push hard like maybe the other way around like if you like when defeat and then the honeymoon is over right when the featuring ends like did you have accuracy bleah like bit higher or like of course its strategy of like how many like like impressions you actually want to drive and you know whether you are already doing like far away with that depends etc but like what will happen to you like yes so the life gets harder when the featuring is over for sure and then the honeymoon is over and then it just depends on a business metrics so it depends what the game target is game goal so if a game goal is to break even within 4-6 months or - I don't know heat ry+ after I don't know three months six months or whatever and by how many percent it's really business agreement

between let's say you a finance product and the I so bi looks at what we have right now then finance looks at what we need to gain but we need to get from this game as on like P&L and all that side product looks at the pipeline and can do and anyway goes okay is it possible or not and to that point I like how has a prevent optimization changed that strategy for your UA because I feel like in the lifecycle of the game like it kind of changed the game so it love to kind of get a sense of how you've leveraged it after feature for example without these organic how did you manage to keep it going we keep going with a parent organization also in combination with install ads and vo as well value-based optimization I don't think it's changed our strategy completely it does affect the quality of

the user so we get so they say the LTV of a o users much better than install ads but at the same time we need volume we need people we need more people to play with each other we need them to enter PvP mode we need them to engage and have fun and enjoy the game before they start paying maybe and things like that so it's always a marketing mix I'd say I wouldn't say that Ayoo would be 50% of the ad spent probably more than 50 would be still in solids so AO would be the smaller part of it and the vo as well so there's a balance of like you gotta need to feel the funnel the installs and then kind of start off I mean you can optimize if you don't have you know instant right so it's kind of like yeah it's an iteration so but I mean we have like few minutes we have like 5 minutes and I still want to give a chance to ask questions for these guys so but a couple of things before that I mean I'm gonna pick back

on the Instagram stories and again it's something that I've kind of I kind of find that very exciting the whole stories and then the crowd of it but at the same time I haven't seen too many successful gain campaigns there or you know filled so I think do you have any experience of like give me a games company or campaign that have been very much you know getting great results on these stories I'm just gonna give you the one I see on my phone I always prefer I'm seeing the played Emma Cadden right now with clash clash and I mean it made me download the game so yeah obviously what we see with Instagram stories that I think is very valuable is that like 1/3 of the people that are swiping through stories actually engage with the brand so I think it's a very interesting metric to look at and actually to test it's still new right and surround stories for businesses is been owned like a year max but we're seeing a lot of success but what we're also seeing is

that the performance metrics vary according to countries so like I've personally seen success with Instagram stories in one market versus another because that market specifically interacted more with Instagram stories yeah I do think it's important to use it in the marketing mix I think that using placement optimisation allows you to have feed into Instagram as well so that's why I'm like all about placement optimization but I do think Instagram story has that engagement rate that is very interesting for answer so would you say Instagram stories are better for branding or performance you a Olson I mean I'm saying the placement optimization is great for performance I think that branding is a whole different world as well like a brand like Ubisoft definitely will use branding differently than a mobile in like a mobile studio which is much more driven on the performance metrics when you're thinking

about branding in general more platforms is important because today we live in an omni-channel world so if you're only on Facebook you're limited to Facebook inventory so branding is a whole different topic but when we're talking about performance I do think that adding in Instagram to placement optimization or is crucial if you want to drive better performance because of the extra inventory last but not least and and a question that Morgan you don't need to answer you don't want to so just recently you guys kind of announced you gonna change the news feed so it's gonna be more about the people who engage with more about your friends less from third parties means less ads as well so how does this impact the sigh column and she's you know running or marketing and you get you get 60 seconds but I know you can talk about an hour but and you don't need transfer by the way I mean you know it's a busy stuff that I get a lot you know from people

like what's gonna change I'm just very excited about this simply because it's Facebook has always been about the user experience first we're always trying to develop our tools so that people actually enjoy the time they spend there and interact on the platform as it pertains to your business specifically I have no idea it's a recent change but we have teams that are really conscious about how businesses and individuals that interacts together I'm sure will have a better sense of how it works down the road but specifically on the impact of your business I don't know sorry it'll be good to find out because in the past we knew that there is one impression per user per day unless it's published story from the fan page then you get two impressions per user per day I'm more interested in these kind of metrics because then if I need to adjust my ad spend on Facebook according to that let's say expectations of how much we can spend profitably on Facebook that would be very interesting I mean right

now in the ads portion on UA like there's no impact at all it's gonna be more on the organic side um so on that front I were I mean right now I have no no idea but I really do think that it's beneficial for Facebook as a platform for people to enjoy time spent there and ultimately it's gonna be beneficial for businesses because these people are gonna be engaged and be able to find meaningful content at the right time and that's what you want as a game company is to be in the newsfeed of someone who wants to play your game yeah exactly I think that's a healthy change right something we need but at the end of it there's this is obviously one of the concerns that people another icons remember like a question like you know is anything gonna change what's gonna change you know do I need to double my budget or you know wanna go out there for that yeah but that's it I thanks guys we have a little bit I guess for a couple of questions any questions you don't need to ask anything for me but these great guys

Monaco you always have your questions sorry no no pressure yeah I bring the mic to your home that helped thanks mark from North so when you mentioned the portion of budgets going to mobile app installs you said more than 50% going to mobile app installs was this in the face of soft launch and how does that share compared to when you're doing user acquisition for another game so now that you have also apparent optimization and the value optimization yes so specifically install ads during a soft launch and then the combination of both even right now even after the global launch is over it's a combination of both because South Park in particular needs a user base you need the PvP mode and in into all of these then depending on some other games for example some other live games like

we have horas even hungry sharks and all of them there are a little bit more casual and different gender I would say more female style for those it's still the same combination it really depends on all TV because app event optimization is value optimization is more expensive when you measure just CPI so cost per day one retained or something like that it's more expensive so you would need to kind of justify the cost and then in the end you still need the volume for some of the games I mean otherwise why do you have the business right so we will need to have a mix anyway I would say on average 40 50 % install based and so based traffic and then the rest would be a mix of the rest - any more questions I mean if you if we have a game one day that has LTV or 50

bucks and I just go all in to a value based that's it and just bid 50 bucks yeah it's a tip of the day build a game in 50 bucks right yeah but then we have to remember all the players right you can't grind them too hard you want them to come back to your game the way through that sure that cool well if no further questions Thanks thank you very much thank you let me got Morgan in psychotic [Music] [Music]