Good Design Makes Happy Customers: A Microtransaction Design Primer

[Music] we are gonna get going if you all promise to silence all of your noise making devices I know you're gonna do it you're gonna be really good okay my name is Kristen Cox I'm a game director at ArenaNet I've been making games for a little over 14 years and for about the last 10 years I've been primarily focused on microtransaction design and microtransactions strategy or leading teams that are responsible for microtransaction design or microtransactions strategy most recently for Guild Wars 2 but my career has mostly been large-scale mmo's so a lot of the examples I'm gonna give here are gonna be from large-scale mmo's but a lot of the stuff I'm talking about should be fairly Universal hopefully so today we're going to talk about microtransaction design I'm actually going to do you one better we're going

to talk about good microtransaction design you're all gonna get your money's worth which is gonna be nice so you know it's important to ask what do I mean by good because that could mean a lot of things especially when we talk about design in this case I'm talking about designs for systems and items that please players that inspire players to engage with them repeatedly and then increase the quality of your game as a whole so that's what we mean by good design and in fact that's a pretty broad definition for good design that could be applied to basically any type of design and that's because microtransaction design is game design it's a specialized part of it but it's not really all that segregated but it is a specialized part of game design and it comes with a couple of special considerations some are really simple perception shifts and some are really highly unintuitive so the bulk of this talk is going to be me walking through some of those considerations the special considerations that I feel are the most important when it comes to doing good

microtransaction design and it's gonna be organized around these big ideas loosely so let's dive right in because there's a lot of stuff there's like 95 slides okay understanding value the most fundamental so one of the most basic questions that we ask and design is why why does the system exist why does this player want to go into this cave all the way up to why are we even making this game good design comes from having good answers for why answers that focus on players needs and what they want rather than the designers needs when design decisions are motivated by solving developer problems like we need this game to be longer so we don't get dinged in the reviews we inevitably end up with weak design and weak player engagement this is why when we ask the question why are we making this microtransaction it's really really important that the answer never be

because we need to make money okay but yes you're saying we do need to make money actually that's important to us of course it is game developers have lots of needs money is actually usually kind of a fundamental baseline need for you or your company or any anyone else is trying to sell something that makes sense but we're talking about design and we're talking about you acting as a designer in this scenario and so if you want to do great design you're going to have to set aside your needs and you're going to have to set aside your company's needs and you're going to have to focus on your players needs I'm not saying this as some kind of like anti capitalism let's make art not money like I'm that's not what this is about your micro transaction items are going to be more successful if you focus on player benefit more than your need for cash because the first rule of selling anything the first rule of Commerce is that people only buy things that have value to them basically we want one too

far players don't care about your needs I just went right by it because you all know that right players are applying a pretty brutal value assessment to everything that you release and your need for cash or cash flow how much revenue you're making it just doesn't factor into that assessment they really just don't care a great example of this is at a high level is the decline and failure of the monthly game subscription business model there we go so why do monthly game subscriptions exist well if they do exist anymore they sort of don't I thought about this a little bit came up with a list this is not like it completely exhaustive list but I think it's a pretty good list of the top reasons that monthly subscription fees existed so that list obviously has pretty high developer value and pretty low customer value it's really strongly skewed towards the developer slash publisher as audiences for online games got bigger and they got more

sophisticated their ability to see funding continued development as a community effort diminished right that the games got bigger the companies got bigger players expected companies take care of themselves it wasn't about hey we're all getting together and gonna you know be scrappy and sort of support our favorite devs it actually became a business and so at that point subscription fees became even less attractive to players of course I'm not suggesting that this is the only reason that the industry made the shift from subscription fees to free-to-play when we're talking about you know big PC e mmoes but it's more complex than that but from a design perspective we're just thinking about design subscription fees come from a place of really weak customer value and that contributed to them underperforming pretty significantly in the face of newer models so many of you probably noticed we talked about the value of monthly subscription fees unless left something out that might have seemed important to you and that was access to the game and

I did this because I want to talk a little bit briefly about inherent versus derived value so items were systems with inherent value right they provide some benefit to the purchaser that was not created by the seller an example of this is Cosmetics bright cats players want to express themselves they want to look the way they want to look and the Hat provides that all in its own your game probably didn't invent the idea of hats so you would get that for free to some extent so items were systems with derived value are ones removed from their bed fit monthly subscription fee is a really simple example the value is the game not the subscription the only reason the subscription has any value is because the seller decided to gate the game behind it this creates a situation where the opportunity cost evaluation that the purchaser is making is removed from the actual value and that puts a higher amount of pressure on the thing that you're purchasing to be that much better the value needs to be clearer it needs

to be higher I'm not saying derive value systems are inherently bad of course they're not you know currency is a derived value system they do a lot of really great things for us but you need to be aware that you don't really get that for free once you remove the evaluation the player is making from the actual purchase that they're making you have to go a step further you have to have more value for them than you normally would the example of the monthly subscription fees a few games did really really well with this model you know in hindsight a very small number but some of them did really well because of the value of those games became such an overwhelming proposition for the players that they were willing to basically do whatever it took if you had access to the game and a lot of that came down to you know cultural zeitgeist and social pressures and things like that and also the quality of the game which is really high now I assume that every game that released under this model kind of assumed that yeah our game is so great that people will definitely jump through the subscription hoop to get to it what's kind of ironic about that is they're

probably right just not about the subscription hoop but they're probably right about the value of their game because when we actually did make the transition in MMORPGs from monthly subscription fees to free to play a lot of games actually operated under both models and pretty much all MMORPGs that even had a little bit of success had a much higher average revenue per paying user a month than $15 like double triple that players via were actually willing to pay a healthy amount of money a $15 a month $30 a month to play those games because they valued those games or they valued the things in them they just didn't really value this description fees in general players actually make two really distinctive value evaluations one happens before the purchase and what happens afterwards before the players rely on you pretty much exclusively to communicate the value of the microtransaction item or system that they're going to buy after the purchase players evaluate how satisfied they are by that purchase and a big part of that

evaluation is going to depend on the gap between what you promised them and what they feel they were delivered and you really need to get this right because microtransactions absolutely 100% depend on the second evaluation the first one's important kind of you have a funnel you have to work on but the second one is really key they are micro transactions we have to have repeat purchasing so you really have to get that one right and you can really hurt yourself a lot if you don't control the first one there's a couple of ways this usually happens one is implied value by marketing right so attempting a manufacturer of that usually vague kind of value from marketing like oh this item is sort of associated with this good thing and if I buy this item that I'm going to get more of that good thing this is a really common tactic to sort of juice up your first evaluation of course you know a lot of times when this kind of ideas taken to an extreme it's pretty absurdly

it's pretty crazy to be like man if I just eat this Snickers bar then topless women will be everywhere obviously that doesn't make any sense but we do it all the time and this really erodes players trust you know it happens in micro transactions as well sometimes not as Extreme as those examples but it can be really simple things like oh I'm gonna have feature a pro player using this item and just you know suggest that if a normal player used it they would play as good as the pro player or it can also be really manipulative I mean you could say we've doubled the drop rate on this item in a random box but you know that the double drop rate is still so incredibly low that the average player is never going to get it you know a lot of these things I'm talking about are actually pretty classic marketing techniques but that's not really a good argument for using them players don't care that other industries use them they just care how they feel when they're evaluating their purchase after they've made it the

second way this happens happens a lot in my career and that's obfuscating value the most egregious form of this is intentionally making items or systems hard to understand because if they were clearly understood the player wouldn't be seeing them as valuable as you're selling them for this can actually happen unintentionally two games are really complicated and if you have a game with a really complex progression system or a lot of complex systems in it this can just happen but whether you intended to do it or not obvious getting value is pretty bad for player trust because weirdly enough people don't like to feel tricked so let's talk about random boxes we're talking about obfuscating value these are going to come up remember good design comes from satisfying player needs and random boxes do fill some very specific player needs but they're also incredibly easy to design in such a way that their value is really murky and that can lead to confusion and eventually resentment and once the player opens a random box and they feel ripped off they are much less likely to

buy another one in the long run so how can we avoid this how do we design a better random box so let's go back to our wise why do players buy random boxes it's not because they love to gamble I I mean it's it's part of the answer it's it might feed into your design but this is not an acceptable answer to this question if you really want to dig into what motivates people sure people do like to gamble but they still don't buy things that don't have real value for them and adding a random element doesn't prevent people from evaluating their satisfaction after they've consumed your item in fact players are really sensitive to value outputs so these are two random boxes from Guild Wars 2 the numbers aren't real don't get too excited but they are two random boxes from Guild Wars 2 they're both cost $1 to purchase random box a outputs about a

dollar 50 on average when a person opens it so the thing they get out of it is worth up usually about a dollar 50 random box B has their output at about a dollar 75 Suzy a really clear thing happened here random box a came into the market and people were like oh this is interesting I've seen you talk about this and it looks kind of cool I'm making my first evaluation I'm gonna purchase it then pretty quickly players obviously made an evaluation that this box was not worth it they were not happy with it and sales went way down conversely apparently for this type of thing a dollar 75 was the sweet spot because when we released the second box not only did it sell really well right away and we already in that first week saw a lot of repeat purchasing but then we got a word-of-mouth effect where we had even new purchasers in the second week and they became repeat purchasers as well and it did significantly better the players didn't have access to the drop tables the players don't don't hear the dollar 75 amount they're evaluating this based on just their satisfaction after they open the box and it can be

tricky to figure out exactly what's going to be satisfying for players but you need to be focused on trying to hit it so players are really sensitive value right but where does that value come from right what how what do they want random boxes for I said they satisfy a couple of specific demands so what are they well they fall into two categories the first is jackpots so in these random boxes mostly what the player opens when the player opens the box they get sort of small incremental consumable rewards but then there's a really small number maybe even as low as one jackpot item that's really really awesome that they're going for these boxes provide players a chance to get something that by necessity must be rare if your game actually supports rarity markets or they allow players to get something expensive for cheap if they're lucky that's the draw that's the value for players buying this kind of box balancing the price of an individual box versus the value of

the jackpot is sort of the key for these boxes the jackpot is the thing the players are evaluating you on in these kind of boxes the non jackpot items they kind of just fall away players don't really see them as having value you may try to argue like no they're totally worth you know 25 cents or something apiece but it doesn't actually affect the way people evaluate these they just look at the jackpot item so if a player looks at your you know jackpot random box that cost a dollar piece and they decide the jackpot item is worth fifty dollars to them and then they don't get the jackpot item after spending fifty dollars they're gonna feel ripped off and they're not gonna want to hear about you know rarity rates across a you know the entire population and they're not going to want to hear about but on average you would have gotten it at that rate they only really care about their own experience and so trying to hone in on figuring out how players are going to value that jackpot is sort of key to pricing these so that you get that great second week repeat purchasing the second

type is grab bags these are random boxes where all of the items in the random box are pretty similar in value they actually usually come out at a fairly even rate grab bags allow players to get something for a cheaper price when they don't have a strong preference for exactly what they want to get so if players are like I just want customization options I just want to have more things in my wardrobe but I'm not really willing to pay a premium for a specific thing then they can open a crap bag and just get something random and they're pretty happy players judge the value of these based on the entire content of the box right is this set of things that I might get good do I want them would I be happy if any of them came out of the box a good way to sort of price these and we'll get later in the talk I'll talk a lot more deeply about pricing but a good way to sort of find your value for these is to think about what is the lowest value item that comes out of the box and is it still a

good deal if I get that item when I open this box of course random boxes do you get used to satisfy other needs and they might have satisfied me that's super specific to your game but what's important is that you have a clear audience that sees clear value in your random box and that you remember that random boxes cannot be used to satisfy every need right that's more of a developer need that's not a player need so there are some players needs they do satisfy really well and they're great for that but we can't sort of force players to use them to satisfy other needs that they'd rather not use them for because that makes them angry we've seen over the past year a couple of cases were pretty poorly designed random boxes have caused a lot of player backlash and really high-profile games and that backlash affects every game even games that use random box as well or you think my I designed my random boxes perfectly and I'm always thinking about my customer value those those very high-profile

backlashes affect you too random boxes are really good at some stuff they make some players really really happy some players would really like them but if we abuse them they're going to become synonymous with cheap money grabs and we're gonna lose them they really could just go away completely lose them out of our toolbox it really wouldn't be that hard we've definitely seen a similar kind of thing happen with other tactics and other options for game design in the past so it's not out of their own possibility but avoiding that really comes down said to understanding what your players want and what they value and player motivations that's a good segue so we're not thinking about money right now we're just thinking about great design we're thinking about value but hey how do you know what your players value right why do people want things we could zoom all the way back out and just talk about why people do things at all and lucky for us there's been a ton of research into why people do things human motivation instance we're talking about video games

we don't have to worry about basic motivations like food and shelter because most people don't use video games to satisfy those needs thank God and we can just focus on models the deal with war high-level human motivations and I actually like this it's actually sort of helpful for me to remember that when we sell things in video games we sell microtransactions we're not selling anything that anyone truly needs so when people decide to buy things that we make they're making a really free evaluation that's pure and they're deciding just based on how good is this item and how happy does it make me I kind of think that's great so there's a bunch of models for high level human motivation but we're gonna use self-determination theory because this is GDC and I have figured a lot of you are familiar with it I also like it right it's pretty great because it's it's all about these high-level motivations and it really cares about the kinds of things that drive people to put a lot of time and

effort into stuff that might not seem like it has any sort of obvious extrinsic value to them and you know its focus on that kind of intrinsic motivation makes it really really suited for doing design work around like rewards progression and monetization which is great for anybody that doesn't know just really really quick STD tells us that many high-level human behaviors are driven by intrinsic motivators and not extrinsic rewards and punishments further intrinsic motivators exists to satisfy mostly three high-band human needs the need for autonomy the need for mastery and the need for relatedness so when we look at micro transactions we can break a lot of them down into these three categories so here's a list of some examples of the kinds of things that satisfy the mastery need the autonomy need the releases the relatedness need so that's actually pretty simple right if your micro transaction just satisfies one of these

needs then you're doing great and it's gonna have value and people going to buy it cool but what needs are best suited to be met by micro transactions and should you try to meet all of them with your micro transactions or should you only try to meet some of them should you try to meet you know the same ones that your game is meeting with other with other means all of this stuff is somewhat dependent on the exact game that you're making but let's go through some sort of core guidelines so aligning your motivations with your core gameplay can give you a lot of great synergy for example collectible card games all by themselves offer a lot of autonomy satisfaction right the more cards you get the more options you have in battle and that makes you feel like you're more free and you have more choices you could sell card packs that also increases the autonomy satisfaction this is the classic model that Magic the Gathering uses both their core loop of collecting

cards and their business model result revolves around satisfying and on motivation um but complimenting your core gameplay needs satisfaction can also be successful when collectible card games moved more into the digital realm we see more of them try to also satisfy the relatedness need by offering like avatars and board custom borders and things like that and that usually gets filled entirely by microtransactions because the core of the game just doesn't really offer much of that it offers a little bit but not much you need to watch out for some cultural biases here there's some pitfalls if your core gameplay mechanic is aimed at satisfying the mastery need and a lot of them are because games tend to focus pretty heavily on the master you need and your intended audience is largely Western you may want to avoid selling items that satisfy the master you need as well because traditionally Western cultures are uncomfortable with the idea of money being exchanged to gain mastery

especially in context where mastery is seen as earned we really value effort and when you start to create a situation in which we say effort equals this signifier of mastery if you sell that as well you're gonna trigger some very deep-seated cultural resentment basically you're gonna make it feel unfair so an example a real example is you've got a fighting game and as players one matches their fighters leveled up and they had a number associated with them and they became more powerful the more matches they've won and then you sold a way to increase that number as well most Western audiences would think that was pretty unfair because you already set up a situation in which that number is a signifier of ability and it's a signifier of effort it's a signifier of wins and so you're muddy in the water you're making that signifier not mean as much you're making it difficult people to understand what mastery really means in your game you can maybe argue that that microtransaction that I just

described is actually a providing an autonomy satisfaction because your trading time for money and traditionally we think time for money trades our autonomy satisfactory satisfactory meters but if that's true you you need to make that clear the example I just gave as I said very muddy like stand what this number means anymore I don't really understand where my mastery level is at so if you want to trade your time for money go for it but make sure you make it really clear to the player that you're not replacing mastery there are examples of selling mastery satisfaction really well in a way that doesn't feel inherently unfair I'd give you won the MMORPG blatant soul which is made by ncsoft has a system of weapon upgrades that requires the player to acquire a whole lot of a specific resource in order to upgrade their weapon they can go out and get that resource in the game or they can buy it you just purchase it from their store but once they have all the resource the weapon does not actually get upgraded unless they beat a boss monster with the

weapon so yeah you can speed up the process of gathering your resource but the thing that you're sort of signifying in the players mind the thing that makes the weapon get upgraded is beating that boss monster and that fulfills a very satisfying intrinsic master need as well you also want to understand what needs your game's reward systems are going to be aiming to satisfy as attempting to satisfy the exact same needs with rewards and microtransactions can lead to the two systems fighting with each other we suffer from this problem pretty significantly on Guild Wars 2 so because in Guild Wars 2 there's no stat increases on gear at the end game rewards for high-level content are mostly cosmetic we also sell a lot of cosmetics for micro transactions this creates a situation where our most dedicated players are motivated to complete the high level content by the cosmetic rewards they are intrinsically not motivated to purchase microtransactions for cosmetic cosmetic microtransactions and vice versa they're fundamentally just sort of

fighting with each other we've alleviated this a little bit over time instead of carving off types of cosmetics for rewards only and types of cosmetics for micro transactions only but it's really not ideal and it makes things particularly difficult for the next section which is pricing because it is confusing for players when something is worth effort but it's also worth dollars so we've been talking about designing micro transactions and just thinking about value and this the abstract way and now we're going to take it into a very concrete world we're going to talk about putting real currency values on your microtransactions in this section you're gonna see a lot of these graphs this is a type of demand curve but it's probably not what you're used to seeing just from like a micro economics textbook there's no supply curve because we don't need it and we don't care about it and get into exactly why but it's not important to us right now so the blue line is gonna be

demand that's just how many people or how many units would you sell at the price which is down at the bottom I've added the white line which is revenue just make it a lot easier to see where those how those different prices affect your overall output for selling the item so just traditionally in economics you wouldn't really see revenue graph that way but we're gonna do it here just make it really easy for you guys to grok quickly where the you know highest revenue-generating price is on each graph cool microeconomics basically tells us that pricing is really really easy you just got to plot this demand curve against a bunch of possible prices and then cool we make our little white line we see where the peak is that's it you've got a price so simple unfortunately forecasting demand to this resolution is really really hard especially if you don't have really specific historical data and pricing can be unintuitive because you don't always

know actually what demand curve you are plotting as an example virtual currency bundles so virtual currency bundles in case you don't know are hard currency basically in a lot of games the player can purchase a currency that they can then use to buy microtransactions instead of having the player pull their wallet out every time they want to buy a hat or a booster or every little thing instead you sell them big bundles of your virtual currency that they can then spend on little micro transactions as they wish a lot of games use this system and virtual currency bundles seem like they should have the most traditional demand curve in the world they're just such a pure there's such a pure demand for them because they are themselves just an abstracted currency so the higher the price the less people are going to be willing to buy it the lower the price the more people are going to be willing to buy it this seems really obvious and I actually see most games just operate on this principle automatically to assuming this is absolutely true actually the virtual currency bundled demand curves look like

this so they're really different than that traditional demand curve they're highly inelastic actually which it seems like they shouldn't be that's very odd so let's dig into why and I'll talk a little bit about this graph specifically to illustrate this this graph is based on this set of data which is obfuscated but it comes from a game that I worked on in which we ran a little experiment on our virtual currency bundle offerings so we just looked at first-time purchasers we didn't sort of muddy the data with people coming in a second time but basically on week one we offered all of these different virtual currency bundle offers from $1.00 to $100 then on week two we took away the $1 offer and on week three we took away the $5 offer and I know this is a lot to digest quickly but what this graph is telling us is that about 70% of first-time

purchasers always purchase the cheapest bundle available but when the cheapest bundle available is set to a higher price doesn't significantly change their behavior you don't really gain any additional purchasers by having the cheaper bundle people are just gonna buy whichever one is the cheapest that they come across and that seems really weird and you might be like wow you just had some really weird game that had this crazy thing happened but no it seems like this is pretty much true across the board it was true for every game I've ever worked on for sure but also back in 2012 Emily Greer from congregate which is an web so used to be solely web based game portal they've branched out to a couple of other platforms they have lots of games on their platform she presented a bunch of data for 150 of their games which showed that yeah this is the this is the virtual currency demand curve for pretty much everything so that doesn't really make any sense well

the thing is that it's not the virtual currency demand curve it's the fifth step in the funnel demand curve we are not really getting to find out what motivates people who just want virtual currency because we had them go through all of these steps to get to the point where they chose which bundle they were going to buy some people have messed around with this a little bit on mobile and done some really cool things where they actually have you pick your bundle first and that actually helps alleviate this problem but the funnel still ends up churning out a lot of potential purchasers and it mostly turns out the people who are interested in only buying the cheapest option so by the time they actually get to the point where they're making the purchase you've already lost most the people that you could sell a $1 bundle - that's actually unfortunate I wish it wasn't true we should all be working on making this funnel better and it is getting better over time but it is true and if you're actually going to make decisions based on a demand curve you better know that you're using you're making a decision based on this demand

curve and not this demand curve and that can be a little bit tricky if you just look at the surface level so demand curves are tricky yes and if you're introducing an entirely new class of microtransactions into your game or you're launching your game for the first time you're gonna have to figure out how you are going to figure out how to price them because you don't have any historical data to build your demand curve off of so let's just walk through a fictional example a step-by-step on putting a price on a micro transaction item that's new so in this example we're gonna say that you have a racing game and that game is heavily focused on players creating custom paint jobs for their cars it's a big part of the game and you've designed a micro transaction that allows players to store multiple paint jobs per car and just switch between them at will so that's a pretty cool micro transaction that fits into your primary motivation awesome so first

to figure out how to plot this demand curve we just need to figure out what kind of demand curve it is because there's two basic types of demand curve there's elastic and there's an elastic so elastic demand curve are what you probably think of when you think of demand curves right this is the thing that's like guns and butter and you're like microeconomics class in college you're like yeah if more people want to buy it if it's cheaper and that's true for some things there are some things that have sort of a universal appeal an example in games for this is boosters in games with number base progression right would people like their progression to go faster yeah pretty much everyone would use a booster if they had it but whether they're going to buy it is based almost solely on whether they think the price you're charging for it is worth their benefit and the lower your price is the more people who are going to decide yeah that's totally worth my benefits it's a really basic demand curve but there's a lot of things that actually have an inelastic demand curve this is a curve in which changing the price doesn't

significantly affect the number of people who are going to buy it this happens a lot with things that have a niche demand a good example in video games is cosmetic items if your game offers me the ability to change my head into a pumpkin and I hate pumpkins and it doesn't matter how cheap that pumpkin head is I'm never going to want it but if the only thing I care about is pumpkins and I'm obsessed with pumpkins and I need to make my character the pumpkin Queen then I'm gonna be willing to pay quite a lot of money to change my head into a pumpkin because that's going to go a long way to satisfying my very specific need so your pumpkin item is actually gonna have a pretty inelastic demand curve the size of your market is going to be smaller but the people in it are not going to stop buying your item until it gets relatively expensive and people who don't inherently want your item aren't going to enter the market until it's really really cheap so that's an inelastic demand curve our example item this paint job storage item is actually somewhere in the middle it's gonna tend to be a bit inelastic because

you are cutting your market off a little in that like if someone is playing this game of yours and they don't want more than one paint job for a car they're never going to want your item so it's not universally appealing so it's going to be inherently a little bit inelastic but since we said that the core motivation one of the core motivations in your game is is customization and personal expression you're gonna have a lot of people who are interested in this I it's gonna be at least somewhat elastic we talked a little bit just now about size of market and that is your next big consideration so how many people could even ever want this item we start by saying like if this item was free how many people would want it and if any of you have ever designed rewards in a game you know that free does not mean everybody wants it there's still a demand curve for stuff that's free so but it's pretty simple so we just use this simple kind of chart how big is the market how many people are gonna want it and we also look at how that relates to elastic and inelastic because if you

have an item that almost nobody wants but it's really elastic that means are going to be selling very few of them for a very low price you might not want to do that you might be okay with it it depends but it's usually that quadrants usually not so fantastic for return on investment our paint storage item is got a smaller market because as I said it's only people who want more than one paint job we've cut everyone else out of that market and it's sort of here in the middle of elastic and inelastic so now that we know that we can get the shape of our demand curve okay you know we kind of know the size of the market so we can plot that and we also kind of know the shape of the elasticity so we can plot that we still don't really know what our possible prices are though and how exactly they're going to affect a demand and that one is tricky figuring that out it usually involves a lot of research and you know maybe some forecasting if you have that available to you you're going to want to look at a couple of things particularly what is your average revenue per paying user for

whatever time period you usually look at some games look at average revenue per paying user a day average revenue per paying user a week average revenue per paying user a month you want to use that as a baseline because if you're like in general we expect people to spend about $10 a week on our game and you introduce an item that costs $15 you are asking players to pay more than you expect them on average to pay in a week for one item and you guys have an idea of how that's going to affect your overall market so you can look at some projections for that you can also look at some comparable titles you know look what other people are doing in the space there's enough free-to-play stuff that you're probably not making something that can't fall into some category you want to do this anyway because your players play other games and if you you know try to say hats in our game cost $100 a piece you better have a really good reason for that because that is not the baseline people are used to paying so using those two things just looking at what your projections are for the game what the

the market is we can at least get a good idea it's not going to be perfect it's not gonna be exactly we can plot some possible prices and actually do this revenue graph and there we've got it we've now plotted what our our revenue is and we've got our peak our revenue optimized price you've done it now you've figured out your price or maybe not because I have one more question should you charge the revenue optimized price this is not a question that I think gets asked enough in monetization design this chart you're looking at I'm going to dive into you a little bit here it's a little bit more of a complex economics concept but bear with me so micro transactions all exist as monopolies they all operate as monopolies if I don't like the price of the boosters for your game I cannot go to another game and buy the boosters there that's not available to me so

we're all operating in monopolies monopolies push commodities to be more inelastic if we're actually being totally honest all micro transactions are inelastic we would never consider anything we sell to truly be elastic in an economic sense and I like a pure economic sense I could have a theory because we operate in a monopoly so everything is inherently somewhat inelastic without competition purchasers are more likely to continue buying something even when the price goes up but that doesn't mean that they are happy to pay that price the concept of socially optimal price comes from what's known as a necessary monopoly which games are not but I think it bears understanding necessary monopolies are monopolies that exist because competition is too inefficient for the market to bear but the thing they're producing is really important and we need it so a really good example of this is the pharmaceuticals industry so we actually don't have enough

high-end biologic biology researchers and medical researchers to fund 12 or 15 different pharmaceutical companies researching the same drug it doesn't exist it's too inefficient and the market can't support it but we need life-saving drugs and so we have a necessary monopoly and we allow that to happen this chart is actually from policy policy making around pharmaceuticals and basically what we've got is if you produce a life-saving drug and you plotted the optimally revenue optimized price for that drug it is going to be very high because it is life-saving and if people need it to live they're going to be willing to pay whatever they can which is gonna be high now a lot of people won't be able to afford it and so they'll die and that is going to incur a social cost and so when we see these socially optimal price that's the price where we sell this life-saving drug for

just enough money to cover the cost of creating the drug that's the best for society because the least number of people die I'm using the pharmaceutical example because calculating this calculating social cost is really really complicated and unfortunately one of the easiest ways to understand it is when we calculate death as the cost that's actually kind of it's a little more binary and cut and dry there are other industries of this effects like telecommunications and utilities providers that is not so dire but it's really complicated to actually calculate social cost here it's a little bit easier to understand if we just say yeah this is the point where the least people die I'm not going to talk about the fair return price because then I want to that this talk to get political you can look into it but in video games we're not talking about a social cost that's death thank goodness but that doesn't mean social cost it doesn't exist in your game and one of the biggest things that

you need to think about when you are pricing something is can you calculate your social cost you should be asking this question you should be thinking about even if you decide that you don't care about your social cost you should at least be considering it it's not as obvious and severe this social cost as it would be with you know life-saving drugs and all of that but sometimes players will pay the revenue optimized price but they will feel taken of advantage taken advantage of every time they pay it it's really easy to say that if someone bought something they must have thought that it was worth the price they paid but in a monopoly that's just not true and charging prices that players perceive as too high can cause a perception problem and it can cause community backlash and it can build up over time if you take one thing away from this talk it is that your games are not free markets and if you are doing economic evaluation of your games and you use free market economics to do it you're lying to yourself you're not

getting accurate information you need to be using monopoly economics because that's the actual market that you exist in one example that I see of this in games a lot is very high prices for things like name changes and server transfers it is true that those are incredibly inelastic and people will pay a really high price for them because when they need them they really need them but they have a tendency to create a situation where the player feels taken advantage of like their account or their players being held hostage because they really feel like they need these things you know it's it's not a mistake that when players are most likely to purchase a server transfer is also correlated with when they are most likely to leave your game forever because it's a really high-pressure situation so social costs can also you know that's a pretty specific example but social cost can kind of creep in to your entire micro transaction catalog because players have different purchasing power you got to

understand that players have spending limits at some point no matter how high their demand is they're going to run out of disposable income so if the only thing you think of is how enticing is this item how much do people want this item you're going to end up cutting a lot of people off because when you you price something at the revenue optimizing price you are cutting some people off that's sort of inevitable especially like in in free-to-play title somebody is going to be on the other side of that lawn but they can't really afford the thing that you're buying there that you're selling but you have to also account for the fact that when you price potential customers out of an item that incurs a social cost and if players continually run into a situation in which everything they want in your game is out of their reach because they can't afford it they're probably going to become frustrated and they'll probably leave because by because in a lot of games the the loop of purchasing things near microtransactions is a core fundamental part of the game and they in players see it that way so if you are saying you can't participate in this you're never going to be allowed into this part of

the game players do you see that as being cut off from things and they don't love it further if you attempt to squeeze players right to their purchasing limit over and over and over again they're eventually going to feel taken advantage of you'll basically optimize revenue on a series of individual items but you'll end up cutting your potential revenue pool significantly right you'll just leave off a big group of them and you'll probably burn some of your players out you know spending burnouts very real has a very like intense negative emotional effect and it also leads to a lot of people leaving games and turning out permanently um and this is kind of silly right because micro transactions have such a low continued production cost which like we don't even have to like think of supply I don't even have to put it on charts because it's so negligible to what we do we have incredible control over pacing and variety of items offered and pricing that like no one in physical goods can ever have so we can actually optimize for this we can care about

social cost too and actually balance against that and hopefully have it longer engagement with our players and more potential customers with a wider or spending pool when we launched Guild Wars 2 we thought about this a lot we were creating a premium game when we launched Guild Wars 2 it cost $60 to get the game to get in the door but we knew we'd plan from the beginning that we were going to support the game through micro transactions we had a live business that we were preparing and we had plan for right from the beginning but then again like I said $60.00 to get in the door everyone who is coming into the game had just paid us 60 bucks and we were creating our first catalog the very first thing we showed players in our cash shop we were constantly asking how will this store look to someone who just paid us $60 they just got in on 60 bucks we needed to be aware how these items values were going to be compared to that initial $60 cost even if maybe in a

vacuum we would have done some things differently we had to be aware of this sort of community and social situation in the game so we wanted the game to feel accessible and sustainable to those players even though they just paid a premium price and we are essentially saying you're signing up for a game in which there's going to be a live business that's the way that we're going to support it so this is going to be a part of that business and a part of your gaming experience so we focused on having a lot of sort of high-value permanent unlocks so set a standard of high value in high quality because that we felt paired better but the fact that they just paid $60 to us and they were and they were interested in a premium experience and they were interested in things that were gonna last the lifetime of of their account but we also put a big variety of low cost consumables into the game because we wanted to make sure players were primed for the fact that there was going to be a lot of accessibility to the cash shop and there was going to be a lot of options for all of our players and we also have a dual

currency system that's freely tradable so players can actually trade they're our virtual currency for the in-game currency at a floating rate that's determined by demand so for us also very important to say hey even if you're not going to be buying our virtual currency you can be trading your gold in for that virtual currency and still use this system there's going to be something cheap enough to make that realistic for you okay so just a few more things about pricing remove on the price you set for your items needs to reflect the value we sort of talked a lot about that about what value is and how to determine it but it also the price you set has a large impact on the players perception of an item's value and first impressions are really important human beings have this thing with numbers there they're sort of special in our in our brains when we attach a number to something it's really hard to dislodge it's harder than an image it's harder than a sound I guess it's not as hard as smell for some

weird reason that's like the most difficult one but numbers are really sticky and once we associate that with them with something that's really hard to change it so once you put an item out there with a price you're kind of stuck with the perception of that price even if you change it later I think this probably goes to that saying but you should never raise prices on items that's you know that's obviously not going to go over really well discounts go over better than raising prizes of course but it's very hard to permanently change that Association so even if you eventually lower the price the price of an item it's never going to have the effect that it would have if you launched with that price you're still going to be sort of fighting against that initial perception for similar reasons frequent polite price fluctuations are not great this even I mean I even mean running frequent sales events can be significantly detrimental to the way players interact with your store it arose player's confidence when

they're purchasing so if prices are constantly changing you're introducing a new question the player has to answer which is is this the right moment for me to buy this should I wait an hour should I wait a day should I wait a week I don't know because the price is always changing and that's an additional question you don't really want them to have right you want to be the easiest possible decision they make to make that purchase the last thing we're going to talk about pricing is indirect pricing it mostly just talked about putting prices on items but there's also a very common practice in microtransaction design to do indirect pricing so we're gonna go back to talking about random boxes I talked a little earlier about them we talked about jackpots and we talked about grab bags they both are also in direct purchasing systems but they do it a little bit differently so jackpots are usually built around the idea of items that derive their value from rarity of course this requires as I said your game to have an actual concept of rarity and for your game to actually put value

on rarity but let's assume that it does and you have jackpots um but if it doesn't you're probably not going to do super great with jackpot items jackpot random boxes basically allow you to price an item by calculating the amount of money a player will spend in order to get the jackpot you don't want to fall into the trap of trying to calculate the price by determining how many of the items you wish to exist in the world that just isn't going to translate we talked about a little before players just don't care they only care about their personal experience and so you want to anchor this as much as you can to if I am an individual player buying this thing how am I gonna feel about its value also hey watch out for random number generators I know they're really simple and they're like now they're just everywhere you can just grab them basically for free so people use them all the time but if you have an item that comes out at a1 and 50 to operate that means 36% of your players aren't gonna get it in 50 drops it's gonna be more than 50 yes 36% of then we're gonna get it in less

than 50 as well but that's not going to really placate the ones that had to pay more than that random number generators are not the only way to go I would implore you to think about other things like set schedules tokens systems anything that you can do to gain a little bit more control as a designer over crafting the experience for your players so that you can have more control over their perception of value random number generators sometimes they're totally fine but you know humans are really bad with actual random okay so grab bags they're not built to route rarity markets so that's not the way that they use they create indirect pricing grab bag function actually function the best when the items come out of them at a about the same rate so even that even though that's true they still are an indirect pricing model let's use an example so let's say you've created a cosmetic item and for instance it's a monocle and then let's say you

plotted your demand curve for that monocle and you looked at your art AR PPU and you looked at your potential market you did all of this research and he found that the monocle is really niche and really an elastic and the optimal price for the monocle was $80 you'd probably look at that and think man that's a lot of money for a monocle that's way out of whack with what most people will pay for cosmetics and other games that might not look so good in our catalog if we actually charged $80 for this monocle a possible solution for this problem is to release the monocle in a grab bag because you did all this work to plot the demand curve for the monocle but there's a second demand curve that exists for players that just want cosmetic options and they'd be happy to have the monocle but they're not going to pay a premium price for the monocle so you can put the monocle on a grab bag with a lot of other cosmetic options you're not gonna be able to charge anywhere near $80 for it but your markets gonna be so much bigger that you'll probably actually do better in the long run games like overwatch basically employ this strategy exclusively and it actually has a pretty

high customer satisfaction over all we're finally through pricing let's talk a little bit about feedback and data so feedback is really really important to design we are all full of cognitive biases and screw skewed perspectives and personal heuristics and we rely on other people to give us feedback and provide a diverse set of voices so that we can understand how to do good design that isn't just for us but it's actually maybe going to be broadly appealing but what do you do when you're designing something that is almost impossible for people to speak openly and honestly about well traditionally for microtransactions the answer is you turn to data and that is great I love data collection the collection of data on a large scale has changed the video game industry right we have amazing insights into the activities of our players now it's really really incredible but data has a really big weakness when it comes to making design decisions and that's

that data is analytical but it's not predictive data can be used to make predictions but all by itself it isn't predictive and that sucks because we really need it to make predictions we need to understand you know how players are gonna react to our designs what we should we focus our efforts on next we need all these predictions and we need them really fast forecasting an actual predictive analysis is really time-consuming it involves people with like PhDs and stuff and it's really hard it's complex and so full it's a really bad substitute for feedback because it's so slow data is fast and now it's really easily obtainable so what I see happen for a lot of developers is they decide that since they need data to be predictive then it just is so you know we why not it's the magic of design we you know you create items and features and changes and then you look at data and you react to the data and that strategy is like sort of effective you know data

sometimes does line up with predictive analysis and so you're going to get some hits and the nice thing about this strategy is that it moves really fast and creates a lot of iterations so it covers up all its misses so it just rolls right over them and you never find out about it but from a design perspective this is super super dangerous even though it creates a lot of work and sometimes that's very satisfying vast majority of outcomes play out over a long period of time right in this strategy guarantees that only a really small subset of possible design decisions are ever going to get made because they are the design decisions that will produce positive data in the short term and your designers will definitely start optimizing for that because their designers and they do that all the time that game every system they can even if they don't realize they're doing it this is basically the equivalent of all your design decisions coming from brainstorming alone only ideas that can be thought of explained and defended in a single sentence are ever gonna survive and that's not great it shouldn't be shocking that short term gain is often

paid for with long term loss if your relationship with your players lasts longer than a single data cycle then making design decisions based on a single data cycle is ludicrous it's just not good enough and you're not going to really gain any insight into why players are doing anything you know this is all one item from Guild Wars 2 it's just looking at it from different perspectives the hourly sales daily sales weekly sales they all tell a different story and I mean I just from this information I couldn't even tell you if this item was successful it sort of depends on what you meant to do with it a good example of this is for the up cell example right you might decide that when it would be most advantageous for a player to buy something we're just gonna pop a message up and remind them that hey they could buy it and it'd be really easy to just click this button right now and purchase it and usually that's going to produce this really big spike in sales but then that spike is sort of destined to fail and go down it almost certainly will go down below your sales before you did this and that's because

people do respond to immediate attention seeking calls to action but they can't keep responding to them our brains are set up to filter that out like the reason we respond to them is they seem like danger and after we realize oh this isn't dangerous and I know what it is we just start tuning it out and we can't really stop that or hardwired for it so at this point you're tying the the items success entirely to the ability for this attention seeking call-to-action to work and it's destined to get less and less effective over time this is exactly the same kind of short-sighted number chasing that leads games to fall into an endless cycle of sales events hey we need to meet our goals this week let's have a sale great we pumped up the numbers oh but now we're doing really poorly the beginning - next week let's have another sale let's do it again and again and again and again this slowly eats away at your average revenue per paying user it has - just the math means that you are basically lowering the price of all your items over time and you really can't stop that and looking

at the big picture can possibly get you out of this cycle so how do you avoid this sort of short-term data being used as a substitute for feedback well you can try to get the data to kind of work for you right set proactive goals that aren't relevant to design I said know what you want to be looking for maybe you're not looking for just raw sales numbers maybe you want to actually be looking at a RPP you maybe you want to be looking at repeat purchasers actually decide what you want to be looking at before you actually put the item out there another way to avoid data replacing feedback is to get the feedback you know I said earlier that people aren't really capable of talking about microtransactions honestly and that's pretty true it's a really difficult conversation to discuss because you know basically it's a highly emotional kind of because it involves money and that makes sense actually because game designers are super scary game designers spend a lot of time learning how to manipulate human nature and encourage specific behavior so we actually have a really

high capacity for doing evil and when you add money to the mix like the stakes are really high and it gets kind of frightening and in fact like if unchecked it can be easy to just start chasing numbers and forget like wait what am i doing like I'm I'm manipulating people into doing you know maybe something that they'll regret or something that they won't like and you have to actually watch out for that you need to check yourself from time to time right are your designs pushing or manipulating people to do things they regret or that make them unhappy your job is to entertain entertain people and make fun even as a micro transaction designer and no one needs to play your game to live we talked about that so if you're creating situations that make players unhappy they're going to leave and preventing that from happening as part of your job as a micro transaction designer not just selling the items but also making sure you're not erode in the quality of your game and now we're out of time basically I'm gonna try to get through this feel free to stay if you want I've like maybe 10 more slides but we are at 4:30 so I apologize so the majority of unstructured my car has micro transaction feedback just comes down to I don't want to have to pay for

things and I'm pretty sympathetic to that because capitalism is inherently unfair but engaging with that feedback isn't really going to help you do better design so when you're soliciting feedback on micro transactions you need to focus on the feedback on things that are relevant to design and open-ended questions like would you buy this are not going to give you what you want because they make people feel pressured to answer in a way that makes them feel virtuous or clever because it's a performative experience to be asked that kind of question and it doesn't really have you're not going to get a lot of relevant information that way instead you want to focus on asking questions like how would you use this item or rank these items in order of value to you things that are more quantitative than qualitative or better yet you know don't ask questions just watch people use your item that's the gold standard for feedback and all an all design it's totally true for micro transactions you can learn a lot by looking at someone's eyes as they use your item

and and watching the way they react okay we are in the homestretch microtransactions and design this is just my little soapbox - in this I strongly believe that the mechanisms by which your game generates money should be the purview of game designers game designers should have the same vision control and implementation responsibilities when it comes to micro transactions as they do with any other system in your video game furthermore game designers should Suze a Stickley participate in the vision control and implementation of micro transactions because they matter a great deal to your game and they matter a great deal to your players if your attitude around micro transactions is I don't want to have to think about that or I personally don't like micro transactions I'd never buy one of those so I'm not going to do this you're doing yourself a huge disservice and you're really doing your players a disservice game designers are uniquely equipped to understand what players will value and how that value should be presented in the most effective way as micro transactions

become more and more a part of all types of video games game designers of all kinds are going to have to get comfortable designing micro transactions and being part of the discussion about micro transaction strategy so if you're a game designer I'm employing you get involved you don't have to have a business background you you just have to do the things you do you're uniquely equipped get involved demand the high quality standards are placed on micro transaction design stand up and confidently declare you are the right person to make sure that those quality bars are met and then do it deliver really great design with micro transactions thank you very much you