LiveOps - F2P Sales Best Practices 2020

the first talk here that i'm going to be speaking to is about free to play sales and this framework that i felt is pretty useful when it comes to thinking about about sales which i call tto targeting timing and offers now i want to first start by talking about why are sales so important and these are actually real comments from folks in the industry who i know that have talked about various applications of sales and so first in terms of bundle sales there was one dev studio that mentioned wow i couldn't believe it implementing bundle sales increased revenue by 30 another dev studio talked about personalized targeting uh this was a studio actually based out of korea that had talked about how they implemented a machine learning against timing and size of iip but they mentioned our implementation of our

per user machine learning to adapt timing and size of iap offers increased ltv by 30 and then there's a games publisher based in europe that did talk to me about how they have implemented segmentation with sales and basically saying we've mainly improved revenue performance with our partners through sales segmentation and targeting so in terms of like the overall kind of environment that we're operating in today i wanted to kind of point out that you know sort of back in 2015 if you if you think about kind of like the you know the basis of competition or how you win in 2015 it was really about having good product capable ua and kind of decent live ops but today i would argue that we're kind of increasingly in a more complicated and competitive industry today and so you need a lot more things you need great product and you need depth with the product you need advanced sales tactics which i'm going to talk about today you need a deep technical infrastructure you need some kind of great ua and pm teams working together

and you need also live content and features road map and plan that is executed well and so my claim is that from what i've seen in terms of alpha in terms of you know kind of driving big gains i would claim that optimizing sales is in my opinion the biggest driver of alpha that i know of at least currently today and that kind of talks to the point that increasingly the other change that we're seeing from 2015 to today is that when we think about the game in free-to-play mobile the game is kind of the tip of the iceberg but underneath the game underlying it is increasingly a lot of infrastructure data infrastructure tools infrastructure and so capabilities around liveops automation crm testing facilities like a b testing or multi-arm bandit but there's a lot of capabilities that underlie the game that are going to be important in order to be successful so now i wanted to talk specifically to this point around sales

and this framework that i call tto targeting timing and offers so what do i mean by this pretty basic targeting is about who do you target with different sales so really a focus around segmentation timing is around when to initiate sales and offers it used to be 2015 more about calendar and events it's increasing become more behavioral and predictive and then the offers themselves what to offer and that has to do with things related to psychology to how you do pricing and discounts and how you do creatives and so i wanted to talk about that but kind of taking a step back and looking at the broader picture in terms of what's happening what we've also seen from 2015 to today is kind of the shift and the shift is from broad targeting to narrow or personalize targeting from calendar or event based time timing to behavioral and machine learning based timing and from general broad offers to contact specific and very personalized individualized offers so there's like this general shift from kind of broad

and context-free to increasingly narrow and context-specific so now i want to kind of dive into each of these areas in terms of targeting timing and offers and i say dive in but really given this is going to be like a 20 25 minute presentation it's still going to be a little bit high level in terms of targeting what do we mean and basically the the notion of targeting is really around segments right what groups that we target and when you think about a segment the segments should actually be very similar in terms of action in other words they should react similarly to an offer they should be similar in terms of value whether it's from an ltv perspective or profitability avenues however you've defined that value and also in terms of needs they should have the same needs or behave in similar ways i also generally recommend that your segments be tractable if this is machine learning or computer based you can have more but for human beings i would recommend having like five to six segments or fewer and then generally your segment should

also be messy meaning mutually exclusive and comprehensively exhaustive basically meaning that if you have a player and you have different segments that a player should generally just be sitting in one specific segment and that you've covered all of your players across all the segments that you've defined and also taking a higher level view in terms of segmentation there's kind of i would say three high-level ways of doing segmentation one is demographic meaning profiles based on a type of population age gender location things like that behavioral which is segmentation based on user behavior which is what i'm going to be talking about and there's also psychographic which is segmentation based on psychological profile and again the basis of what i'll be talking about next few slides is more focused around this behavioral segmentation and so when i when i think about the types of behavioral segmentation that i've had experienced in or have seen you know friends and industries run i

basically think of these six primary methods and so the first that i've highlighted is uh what i'm calling rfm cross something and rfm is basically a methodology that was actually developed back in the old kind of e-commerce and catalog days but is is a way of segmenting your user based upon these three metrics r f and m r meaning recency f frequency and m monetary value and i've highlighted this because i gave a version of this presentation about a year ago but since that time i have actually seen data of a studio that's implemented not rfm but what i'm calling rfm cross something where they've taken rfm and they've taken another type of segmentation combine them and driven really really incredible results so basically like 40 to 100 percent increase in our boo there's also k-means clustering there's also vip regulars which is kind of a methodology that was popularized by this guy by the name of jeff howe who was the former cto of kabam and

when he talked about what they did on contest of champions there is also sequential pattern mining which is basically trying to determine you know who your best customers or your best players in the game are and then you write out the sequence you kind of track the sequence of events for those types of players and you try to find patterns and my my guess is we'll find out when andre talks about user journeys but he might talk about sequential pattern mining machine learning which has been increasingly being used this is an area that i kind of rely on data scientists for but when they talk about whether it's you know random forest or genetic algorithms or things like that these are these are approaches that are being used increasingly and seem to be having a lot of good performance and finally is something that i call progression pair buckets which is basically looking at your pear bucket groups by progression so like a d90 peer bucket group or you know whatever progression you want to use whether it's days or levels or sessions

and then seeing how your peer buckets kind of split out based upon that progression and so just to give you a little bit of a better understanding in terms of one of these approaches we can talk about rfm but with this approach you basically want to split out your user base in terms of quintiles meaning you know 20 chunks against these three metrics recency which could be a day since last session or however you want to define it it could be days since last meaningful session or or whatever that recency is for you and your game frequency and the money spent and so as you can kind of see in the slides when you split your your player base into quintiles against those three metrics then you can kind of create like a 3d graph um like this cube and then basically what you want to do is you want to try and plot your these these users and then try to find clusters right and then based on those clusters that should describe a cluster of similar types of players

and similar types of behavior you can also apply k-means against that to try and find specific clusters using this approach and i do think there are also some third-party off-the-shelf tools that actually run k-means for you i don't think they do it against rfm but i believe like amplitude has a feature i think they call it like compass or something like that where they also run some of these approaches for you off the shelf all right so that is targeting i want to now talk about timing and so timing is really about when should you send offers to your players and basically though what i recommend is thinking about the types of timing that you could have for offers in your game and then based upon each of these types of offers think about how you would change or experiment with the type of timing that you have and so in terms of these categories of timing first in the upper left hand corner we basically want to think about general or

fixed timing meaning you start a game and there might be an lto a limited time offer a starter pack you might start a vip promo like x days after a user has begun you can have a weekend or monthly sell but these are kind of like offers and sales based upon fixed timing the other type of timing that's very traditional is basically special days so calendar-based timing there's holidays there may be certain game achievements like hey we just hit 1 million users or clash royale i have their you know this is our 10-year anniversary and can have specials and offers and things like that based upon that in the upper right hand corner there are also timing based upon game events or trigger based timing so these are usually auto triggered so for example if a user gets tagged as a as a player that is potentially going to turn so you've got some churn now so running in the background and the player gets tagged then you can send them a re-engagement sale for example it could also be progression based so a very

famous example of this is clash royale you get to the next tier arena tier and then you're immediately offered a sale to try and help you become more competitive in that tier there's also pre or post in-game events pre or post offers and sales based upon game events so like you know legendary has these like complicated events but you know any any game with like a big event but you can have an offer before the event to help players try to win in that event and finally is the type of timing that seems to be driving the most that seems to be the most effective which is basically running your machine learning based algorithms and i mentioned early on for example you can have machine learning try to determine the best time based upon average purchase time of other players in their segment it could be based on difficulty analysis or you could just feed your machine learning algorithm 50 signals 100 signals and just have it figure out when's the best time to offer a specific player an offer or sale of some kind and

just to kind of highlight now this is a slide i took from another presentation on gotcha but i just want to highlight thinking behind why it's good to offer to give a player a sale at a certain period of time and why the timing may be important so for example in japanese gotcha games they have what's known as a player stress model and so as you purchase more gotcha in the game the designers of the game intentionally try to drive player emotions higher or lower and there's this notion of of basically as players start to lose that they get more frustrated and then there's like this optimal monetization space and so like the ideal time to monetize a player is this point where they're starting to get frustrated by difficulty by losses or things like that but before they get so frustrated that they leave so in between their user monetization point and their user churn point so this is just an example to conceptually describe why it is why timing actually does make a

difference in terms of when you might be able to send an offer and being more successful and finally just some other considerations to think about with timing but you should also factor in things like time of day time of week the in-app timing in the ux flow right so are you going to pop up like 10 offers at the start are you going to try to integrate an offer within a game flow availability meaning is the offer going to be time limited is a player just going to have like an hour to accept or a day or things like that and then frequency meaning like the number of offers that you're giving the player as well so that is timing and now i wanted to jump in and talk about offers so the offers themselves and so the offers is really about what are you offering and i break it down into four components of the offer the first is in terms of the items right so what items should be offered is it a single item is it a bundle of items the pricing in terms of how big of an iep is it and so you know some players fifty dollars for

a purchase may be you know maybe fine but for other players fifty dollars may be too much and so for the different segments what is the pricing point that makes the most sense a discount how much of a discount do you need to offer to convert players at certain points in time and then the creatives which is you know how you present the offer to the player and so not having the time to go in depth in all this what i want to do is just kind of give you guys some of the key considerations to be thinking about when you think about each of these kind of components and so for the items themselves some of the things that you should be thinking about is first in terms of the progression and objectives right and so you want to offer items that will allow players to progress more or achieve objectives in their games you also might want to offer items based upon events so you have an event coming up offering items to help them in that event also items around excitement fortnite is very famous for this like being able to do that marvel integration or the ninja integration

so that you're offering things to the player that they're really just going to be excited about and then the last thing that the last category i had called the psychology of spend this is something that my co-founder at the startup studio that i have talks about a lot where he thinks about why is a player in the game what are they trying to achieve and then designing items around those psychological motivations for being in the game and then the other thing just to consider in terms of the impact of the items would be also thinking about analysis in terms of as you offer these items what would be the impact on economy and progression i think this is something that in our last digital conference we talked about with respect to for example clash royale and then having offers and things in that game that are pushing players to the uh to the game frontier and then also what is the impact going to be in terms of my content pipeline and so for a lot of for example cosmetics battle pass systems there could potentially be a big

impact on the development team and what the cost is going to be to deliver a battle pass for example that is considered valuable by the players the next thing is pricing here i have a picture of jeff witt here because in terms of like price opposition and price anchoring it's this is stuff that i i've completely learned from jeff and i think that if any of you guys follow his blog you should probably learn more about that stuff from jeff but basically the the thinking here is when you think about pricing you should think about not and i'm not advocating for this but if you do want to off-escape the price through gotcha or other systems how you do your price anchoring to make users feel like they're getting a deal how you do discovery like understanding what that price point is and so for example companies like scopely are famous for having like step up step down models but i think discovery increasingly will become important as we have for example uh changes in privacy in ios 14 and so as you're targeting players and bringing them in you can't count on

like whales anymore based upon your targeting so i do think that discovery and figuring out right right price points for segments within your existing player base become more valuable another point is just understanding value versus willingness to pay and what i mean by that is that there are certainly some games where you might want to just offer something to pay even if it doesn't have a lot of value and games like that are games like you know mobas or shooters games where players are finding a lot of entertainment enjoyment and if you offer them something they'll pay for it just because they're finding value from the game itself rather than from the items and then the last thing that i want to talk about is understanding the items that you're offering from the perspective of is it a luxury versus substitution good so i think this becomes more important if we depending upon the economic times and if we go into like a recessionary or depressionary environment then research has shown that if you are offering items that allow players to play more

and it's more of a replacement for entertainment hours or value then that's something that's going to continue to sell but if it's more of a luxury good which is a lot of spending by wales and that actually starts to decline in terms of purchase and then finally in terms of like the impact of pricing i would actually recommend that you look at peer buckets and this is from actually a different presentation but i just wanted to kind of conceptually show you this notion of like okay as you change pricing if you see for example a cluster of paste spend then think about what that's going to do from the perspective of price times your transactions right so does that make sense so if you've got if there is a high cluster of players that spend a hundred dollars in your game and you've got and you're mainly monetizing off a battle pass and you want to price it at ten dollars then that means you're gonna have to sell 10 battle passes but think about for example what the impact in terms of the number of transactions and things like

that will have on your game okay and then finally discounts so some of the things to consider with respect to discounts is really about first the predictability of the discount so you don't want your discounts to be predictable you want them to be variable and random so that players don't just they aren't trained to just wait for the discount to come you also want to have a limited test discount strategy so whatever strategy you employ for your discount like i wouldn't roll that out to your entire user base right like test it with specific segments make sure you have a hold out or control group and then as you discover discount strategies that work then roll that out more and then i do think that the use of machine learning and personalization seems to be very very effective with respect to determining discounts that actually convert very well and the analysis here is really about trying to understand what the ltv impact is based upon your discount making sure that you know you're not having a negative

tv impact that you're not just bringing revenue forward and finally i just want to talk about creatives creatives make a difference this could be an entire talk in and of itself but the point i want to make with creatives is that it you know just based on my experience it seems like a lot of studios don't actually treat in-game shop creatives like they do ad creators right so there's all this focus on kind of you know top of funnel in terms of acquiring users but again if we have things like privacy changes in ios 14 i would recommend you put more focus on how do you actually convert players in your game and this isn't an offer but you can see for example like in call of duty mobile the change in terms of this creative for by premium pass from season six to to season seven and currently i mean you can tell that you just based upon that change in creatives you can definitely imagine that that's actually converting a lot better and so in summary and i know i went

through this very quickly but we do have limited time here in summary i just want to say that basically two things one that i believe that using this tto sales framework can drive a lot of alpha a lot of additional performance in live ops but also want to caution that some investment is required and the first thing is kind of some of you heard us speak about at the in the networking phase before we started the conference first is data infrastructure right you need to have an analytics solution of some kind and i do think that if you're that current off-the-shelf tools aren't good enough to get the visibility you'll need to properly optimize so this will require some investment in a data structure that enables custom views but at the same time for example uh and i'm not saying that this is the solution you should use but i have heard of a number of studios out there using something as simple as like a firebase plus google bigquery plus periscope which is cheap and relatively no low

maintenance and low upkeep and secondly is the tools and infrastructure right and so in order to execute on some of the capabilities and things that i've been talking about you're going to need a tools and infrastructure capable of supporting personalization segmentation ability to manage hundreds of offers easily the ability to dynamically change prices discounts things like that but i would say for a game at scale and by not hiring like you know you know i don't think you need a full data science team i mean we can hand it over to andre to give his thoughts on this but i believe you can actually be pretty cost efficient by having one or two folks implement at least some of these things in a way that can help you drive performance especially for games at scale so anyway that is my presentation please let me know what you think and happy to talk to anyone offline about some of these practices and capabilities offline in the first

generation of mobile gamers came off consoles from titles like super mario warcraft and league of legends now as we enter the era of mobile 2.0 with three and a half billion devices global gaming has mainstream easy to play wherever you are and something for everyone the categories have proliferated the best gaming companies in the world need deeper insights a framework to capture the market opportunity and now introducing game iq a revolutionary analytics product that delivers a taxonomy of genres sub-genres and comprehensive feature analysis matched with app annie's legendary market data you have the insights to level up your performance

designed exclusively for the rock stars of mobile we have one question here from edgy santos how do you set the initial price of an item let's say you created a skin for a specific hero how would you start pricing it what are the things you need to consider how to know if it is expensive or cheap i would say that the initial price of an item i would say that in practice the way that's generally set is basically that that's kind of like your pm or your producer just being like well it feels like it should be x and then then that's kind of how you start and then you would start a b testing that or if you've got better facilities in terms of like a machine learning based type of approach then you can do that as well but i would say that what i generally see would just be starting with a specific price and then testing different prices uh from there from troy to quote myself from chat i

wonder if varying up sales discount will erode a player's confidence and give them an additional question to enter is now the best time to buy should i wait until tomorrow the weekend etc yeah i do think that the frequency and amount of timing does have an impact and i would say that the best way to actually to execute that would be through using some type of machine learning based approach right so for example in clash royale i've never bought anything unless i had a 5x discount on it and i've just been trained to wait but at the same time i think that trying to manually figure out how much of a discount when to offer a discount the frequency of the discount to offer to different segments in your player base for human beings is is very very difficult so i would try and rely on algorithms for stuff like that do you have any specific examples of rfm

usage so i do actually well let me see so so i can share with you some links in terms of rfm but i think to actually implement it it you know uh you you kind of have to figure out how to implement it yourself there isn't like an off-the-shelf tool that does rfm as far as i know but i will say that i have seen the data from again a company that has implemented an rfm segmentation strategy they actually crossed it with another type of segmentation as well it has performed extremely well for them on converting non-peers using personalization for offers what would you say are the key elements we should be focusing on to begin with pricing content and value timing i would focus on value and there's actually a really great talk by the lead on clash clash of clans eno i forgot his last name but if you go to

the supercell youtube channel he gives a talk where they have actually done exactly that which is basically they've increased conversion so basically what they've done is they've taken their player base and increased uh the number of players you actually spend in the game so increasing nonpayers two pairs but they've actually decreased our poo uh our poopoo so i would say that uh that would be um in the interest of time that would be the the best reference for you to kind of look at but generally speaking i would say that the best way is to increase that value but i would say that the trick that i would recommend based upon the um the the talk that i've given today is really to try and understand what that is for your primary segments right and so you know value for one segment may be different from value to another i think we'll try and uh hit some of these questions that were asked before and maybe we can just open it up to anybody who wants to answer this but one question are there any tips to blend

vip model with regular iap model um i think that the best integration of sort of vip or premium systems with i would say maybe not regular ip model but with free models are kind of like the integrations that we're seeing a lot in terms of battle pass but i don't know is there any other thoughts in terms of vip and blending that with regular iep models yeah no i i i would i would love a clarification on sort of what a vip model is because i i do think to your point joe like battle passes are sort of the the the most recent most successful thing that people have done and uh you know they they pretty well blend sort of reward progression with spend i'd be surprised if there's a better implementation out there right now all right from ben sutherland do you have any comments or best practices on setting different prices for the same

items for players in different countries oh i i can't answer this i don't know if any of you guys have a good response for this one i i can jump on it okay so i think there are many models out there you don't need how many companies that are doing it but also there are many models that you can you can test outside and i think it's kind of normal to price it according to the country you can see that also the the big uh the big services you know you look at the for example netflix or spotify there even they have pricing according to your country so i don't think that this is something that uh i don't know kind of cheating or something like that it's need to happen there are many many out there you can look very easily on the internet and start learning it and what are the you know you have the um many very known very very known uh like conversion between countries but

after that it's a long process to find the right balance that that you have in each country and it's also no end of the day we are talking about the paybacks and so on the cac in low in more cheap countries are also cheaper so you need to find the balance which will give you the the best payback with the right pricing which will give you the from one hand the revenue but from the other end you will have a conversion rate and you will have enough paid use yeah so maybe the one other way to think about this is kind of more of uh from the perspective of price discrimination and so like you know different prices but you can deliver it through iap which is your highest you can deliver it through sales you can deliver it through battle pass you can deliver it in different methods so that again it's the same items but you're able to to kind of fairly or perceive this fairly charged different prices for the same when we're talking about even especially in small product and so on the better is to use

like in my opinion like one magic like a conversion and then to just change everything according to it it will be much more easy and it will be also scalable so you will not need all the time you need this kpi to each country but once you have it it's automatically change everything in your game according to it next question from mikhail kakov okay we've got someone famous here at two different companies i've worked at long term over a year a b test shows that sales only changed purchase timing but didn't influence overall ltv of the player just like you said in your clash rel example buy only on sales have you seen similar results my response to this is that if we are talking about 2015 type sales 100 like in fact sometimes when you run sales ltv actually reduces but i would say that with the with the newer kind of approaches that that i talked about

these are the types of approaches where you're actually uh increasing the arpu increasing monetization by your players and i think one way to think about it is i don't know if people in the audience are familiar with kind of this this notion in marketing uh that was recently posted about called adjacent user theory so what i would try and do is i would try to think and actually andre had a really good mapping in terms of like how you could think about um i would actually try to use andre's framework um but think about pair buckets by progression and then think about from two kind of two different sides like how can you make how can you have an adjacent payer theory in other words that you've got a certain type of payer and you've got you've got a certain segment of players that are adjacent how do you move them from one bucket to the other you can do that through uh changing you know kind of you could adapt your game design you can adapt

your user experience but on the offers and sales side you could also try to move them in that direction through through sales and offers as well so that's how i would think about it and just having seen the data from folks that have run different types of approaches using segmentation using offers i've seen really really really strong results that actually increase ltv so we're not we're not back in 2015 in my opinion i don't know if anyone else has any other thoughts on that i think there is there's sort of two i mean there are many more than two ways to to talk about sales but like one is sort of to elaborate a little bit on the 2015 style of sales right deep currency discounts are going to teach people to wait for deep currency discounts and if all your sales are are knocking 30 off of the the price of your imp package yeah people are gonna stop buying your your base offers that shouldn't surprise anyone people people will always see a good deal especially when it's their money in place the way that you can run sales that i

think are going to be more creative to ltv is by sort of to paul's examples from from league of legends right you you give them these limited time offers on things that they can't get otherwise whether that's bundle sales whether that's you know these these limited time objects that creates value where you don't necessarily have it at a baseline right you can't get the christmas skins in overwatch by playing anything other than the events and buying the boxes during that event in overwatch right the same is true of many other games at many other points in time yeah i think that's a great point in terms of like if you're offering stuff are you expanding the universe of products that you're selling or are you keeping it the same this is kind of like the um you'll often see people complain about hey you're offering this item in battle pass that you already offered before don't do that and so so the more unique items that you offer on an lto basis i i totally agree you're kind of expanding the universe of

the economy all right next question i would love to hear your opinion about our pricing strategy we usually give the new item bundled in an offer and we let the community decide the price basically we have the ability to make the item transferable and our users start selling trading between each other and we decide the price based on that i don't know what game that is but i i have seen kind of like um a variation on this where it's like you design your own sales bundle so you like throw stuff in and then you you buy what you've thrown in so i've seen that but i i don't know if any of you guys have any opinions on this one that's a real economy i mean that that that's that's ultima 1990 altima okay uh i mean the closest thing i've seen so far right is like poker games right you can make your own bundle you can choose the size of the bundle right and obviously

in a true poker game i guess you win chips and and so there is that exchange right there of the currency as far as letting people decide i mean if they're deciding what the values are i'm not seeing the incentive on optimization yeah i mean i would just decide zero like exactly that's what i'm thinking something i'm trying to think like is there a game a game feature that encourages to go up but i i doubt it yeah i mean you saw stuff like with artifact right i mean not that artifact is like a case study in success per se but like it at least had something of a secondary economy where the price was both developer influence and community influence curious you know if they're letting go control of the prices of these items hopefully there's other levers like how many items do you need and stuff that they could control to compensate for not being to control sort of the price they're getting for the items yeah i think it's a little bit dangerous because you don't know what the impact to the economy is going to be

but i would say a couple of things one you might want to like look at some of these games and i know licking grit has some reports that talk about some of these games that do your like build your own sales bundle so that i think that's something interesting and then also there are i know a few companies use kind of machine learning like i believe it's like random forest or something like that to determine what items get thrown into a bundle so you might want to look into some of the research behind that as well okay next question offer packs that have a very huge discount like 300 500 are these usually focused on new payer conversion i would say for me i don't feel like human beings are the best to figure out the frequency and the amount of discounts personally but what i would try and do is really more around trying to figure out what your segments are and based upon you know your segments

and progression of the segments uh i would try a lot of different discount strategies and based upon what seems to work then roll that out to the broader segment any other thoughts on that i i agree and i just can add that for new user i think it's less by the way less dramatically to show them that in the beginning you have like 500 percent and there are many better techniques i think to let's say to get the first payment and to convert them and yes as you said also segmentation and by the segment to show exactly what is the best uh the best deal and to know how we'll use the react because it's also not about all the time what is the discount but how you present it because you can have the like three uh similar exactly similar between by value of offers but the different theme and different like way of uh showing it to user and you will get a completely different uh

different result so i think it's also a point there that's very important okay yeah you know this whole you know percent value add is is you know joseph mentioned uh price obfuscation uh in this presentation and that sort of like value-added percentages is one of the the ways to do it right because a a player is probably not going to know unless it's a very simple game what 300 what 500 means and if you go more complicated game like an rpg you know like for example afk arena when you hit a level milestone they'll hit you with a 50 or 100 sale that says 732 with uh 20 you know faction specific gotchas lots of gems lots of essences and other stuff no one knows what the value of that is because um a lot of that is not sold directly in the store it's it's so limited time but like um so you can get away with a large uh percentage sale values in these bundles um

but don't expect like if you're trying to quibble between 300 500 it's hard for players to know kind of what that means so like um i i can't tell you what percent you should aim for be but i don't think players could really understand that either higher is probably better but it's it's um it's it's it's kind of weird and i i don't think we we understand you know what the best strategy is for those value add percentages hey folks glad you could catch this presentation so good to have you join us this was actually part of a larger digital conference around best practices in live ops so click the link to basically get the agenda for all of the other talks that were part of this conference and links so that you can watch the other talks as part of that conference also if you haven't checked it out yet check out appiany's game iq product game iq is a great way to get market data on the mobile gaming market

and app annie was the sponsor of this conference and finally if you want to be notified of future talks in virtual future virtual conference events please make sure to click subscribe to the game makers youtube channel and click the bell notification so you can be notified of future events coming up we've got a lot of cool stuff coming so stay tuned and great to have you with us catch you later