Developer Roundtable: Triple-A, Free-to-Play

Author: by David Daw

A reprint from the February 2013 issue of Gamaustra's sister publication Game Developer magazine, this article examines the growth of premium free-to-play games. You can subscribe to the print or digital edition at GDMag's subscription page, download the Game Developer iOS app to subscribe or buy individual issues from your iOS device, or purchase individual digital issues from our store.

Free-to-play is nothing new for many core PC game audiences, with Valve's Team Fortress 2, Wargaming.net's World of Tanks, and Riot Games' League of Legends championing microtransaction-based models for a few years now.

However, in the latter half of 2012 we saw three new entrants in the free-to-play market, each bringing back classic boxed PC franchises in a new free-to-play format: Tribes: Ascend (Hi-Rez Studios), MechWarrior Online (Piranha Games Inc.), and PlanetSide 2 (Sony Online Entertainment).

Game Developer caught up with Todd Harris (Hi-Rez COO), Matt Higby (PlanetSide 2 creative director), and Bryan Ekman (MechWarrior Online creative director) to try to unpack each of their strategies for cracking the core PC game market.

Is there anything about your respective IPs that made them particularly conducive to a free-to-play (F2P) business model?

Todd Harris: Um, no. I would actually say it was probably not that expected that we would make it free-to-play with that particular franchise. I think it kind of had maybe the opposite reputation. We're the fourth Tribes game for the PC, but it had been a while since there had been a Tribes game.

They were known for being quite hardcore and for having a pretty passionate group of veterans that still played the old game, and also known for being all about the multiplayer, not single player, so things like balance and any perception around pay-to-win would be a big deal. So I actually think Tribes was not an IP that people would have expected to go F2P, and we saw that as a challenge.

Matt Higby: One of the best things about it, I think, is that a lot of times you're making an F2P game, and a lot of the people coming in and getting enjoyment out of your game aren't really doing a lot for you unless they're buying stuff from the store. As a developer, you're not getting much out of the people that are playing for free, unless you can entice them into buying something. With PlanetSide, that's not true; everybody that jumps in and plays PlanetSide is actually providing content for all of the people that are playing PlanetSide with them.

Bryan Ekman: The nature of the MechWarrior IP allows us to attract a large player audience; who doesn't like giant robots? Once a player gets through the basics of learning how to pilot a walking tank, they will find a very deep and engaging experience that allows you to tinker and customize your BattleMech (avatar) endlessly. The nature of a BattleMech gives us plenty of opportunities to monetize non-pay-to-win (P2W) concepts, such as time-savers and cosmetics, that add real value to your battlefield persona.

Was the game originally conceived as a free-to-play game, or was that model added during the development process?

TH: Step one was that we wanted to make a game that had the gameplay of Tribes -- specifically, a fast-paced shooter with jetpacks. We liked that gameplay, so first we wanted to make that type of game. Second, we were fans of the game, so we looked into buying the IP and we ended up doing that so we could make a Tribes game. It was originally envisioned as a one-time purchase, but then throughout the development cycle we shifted toward making it free-to-play.

MH: When we first set out to make PlanetSide 2, we knew that F2P was on the table as a possibility, and as we built the game out more and more, we found all the ways that it fit, and I think one of the things that's fascinating about free-to-play is how well it fits a lot of different types of games. So as we were building the game out it became more and more clear that free-to-play was the best option for us.

BE: When Piranha Games first started working with the MechWarrior IP back in 2008, our intent was to make a traditional console product. In early 2011, we acquired the Xbox and PC licensing rights, and quickly decided to scrap the old brick-and-mortar design in favor of taking MechWarrior in a new direction. And thus MechWarrior Online was born.

How does your game convert a free-rider into a paying player?

TH: What someone gets value-wise is two categories of things. One of them is cosmetics. We have cosmetic skins, so you can make yourself look a little more badass, but it doesn't affect your actual game. We also have voice packs, which have been kind of nostalgic for the Tribes players; the early games had these built-in voice quick-key commands that let you taunt or call out tactics, so we offer custom voice packs for those.

Then the second thing is you get more variety faster by paying. You can unlock additional classes, so if you want to play a stealth infiltrator class, or a technician who deploys things, you can unlock new classes and new weapons for those classes much more quickly if you decide to pay.

PlanetSide 2

MH: With PlanetSide 2 one of our keystones is ensuring that we have fair competitive gameplay. So one of the things we decided is that anything that can affect gameplay in any way can always be unlocked through gameplay. You can go and purchase items, but you can unlock those items through gameplay too. I think if you're making a competitive free-to-play game, that's a must-have.

The main thing we're doing with monetization right now is convenience. People who might not have 40 or 50 hours to play games anymore (like you might have had if you were me when I was in college) can use microtransactions or purchase a membership to unlock items more quickly. So the convenience factor is really key, and also we have lots of cool cosmetic things.

Since those don't actually affect gameplay, those are sold pretty much exclusively for Station Cash (currency purchased with in-game money - ed.) in the game, and we find a lot of people actually have those so they can be a little more distinctive. For us, though, the most important way to turn a free player into a funded player or a paid player is to just have a fun game.

We know that having a game that people can log into every day and have fun in every day -- have an enjoyable experience -- that's the thing that convinces them to spend money more than anything else, including even what we can offer them to spend money on. The players feel like the game is fun, and they want to be able to support the developers of the game, and we see that being a true thing within our community.

BE: First, we focus on getting the player engaged and teaching them the mechanics of what makes MechWarrior Online fun and refreshing. Then, after a player has learned the basics of piloting, tinkering in the MechLab, and customizing their BattleMech, they discover a cash store filled with fun or advantageous items to purchase. That said, it's still possible to have a free and fun MechWarrior Online experience.

What kind of in-game metrics and analytics tools do you use to measure your game's health?

TH: I think what's most interesting, from a developer standpoint versus a business view, is all of the game design metrics that we collect and that any F2P developer can collect. So yes, we look at retention rates, and monetization rates, and what's selling, but our designers have access to really, really detailed data on the strength of every weapon. They can look at, for instance, the kill-to-death ratio of the nine different classes in the game, and whether those ratios are actually bearing out in reality as we would expect from the game design.

With a single day's worth of data, our design team can see enough statistically relevant data to see if any design changes are working as intended or not. So that's really what's most exciting to the game design team.

Every match you play in Tribes, that data, in terms of how many kills, what weapon you used, how effective those weapons were, how effective your team was -- all of that is being captured in a persistent database, and our designers can use that data to improve the game.

MH: We have very extensive monitoring and metrics tools in PlanetSide 2 for us to figure out stuff like how many people logged in today, logged in yesterday, and percentages of falloff of people. Also every kill that happens, every death that happens, we track and we can filter that through a variety of tools to figure out balance -- figuring out which areas of the game people play and stick around with.

So yeah, data gathering and metrics for a game like this, where we're planning on making changes for years to come, being able to track all our metrics and what people are doing and what we can do to make people keep playing is really, really important.

BE: We designed our own proprietary telemetry system that logs pretty much every user action in the game, from where they click, to how well they do in each match, to how much they spend and when, to their average FPS. Our community of players also gives us regular feedback and has been a huge asset in the Closed and Open Beta phase.

How do you decide what to charge for and how much to charge? Is there a coherent philosophy behind your monetization design?

TH: The philosophy that we started with in Tribes was that we wanted it to be relatively less expensive in terms of time or money to unlock different classes, so that various roles on the battlefield would be filled up pretty quickly. Then, in terms of weapons, we wanted there to be more progression involved in terms of player time or money. So classes first, weapons second, just so there would be diversity on the battlefield. Beyond that, it's fairly metrics-driven, and we do a lot of experimentation in terms of price points.

MH: I think it's a feel thing. I don't think there's really a formula that you can plug stuff into to figure it out exactly. It has to do with how many items you're going to allow people to unlock. What sort of progression is involved in unlocking items? What's the gameplay associated with unlocking?

For us, as you're playing PlanetSide and getting kills and capturing bases and all that stuff, you're earning stuff that you need. You don't really need to go out of your way to do stuff that's not fun, or not part of the core game, to get the points that you use to unlock new items. So the very core of the entertainment experience of the game is also helping you progress your character and unlock new stuff.

But we set kind of a wide range of prices from things that are like 50 cents to, I think our highest-priced items right now are bundles that give you multiple items for around 10 bucks. At the end of the day, with a free-to-play game, the best possible thing you can do is make people feel good about the purchases that they're making. Make them feel they got a good value for what they're spending, and that they're supporting a game they enjoy. If you can accomplish those two things I think you can be successful in the free-to-play space.

BE: We're still working out how elastic our economy is, testing a variety of price points, value propositions, and rarity. We generally start with a theory on value, and the player confirms (through a purchase or not) the value of an item, and how much would they be willing to pay for it. Then we test the theory and analyze the results. Based on early Closed Beta data, we tuned our prices and content toward the results of these tests. Now that we're in Open Beta and seeing true user-buying habits, we've tweaked a few of our original theories, most notably by adding both temporary and permanent buying options.

Lots of rather vocal players toss "pay-to-win" accusations around. From a game-design perspective, how do you monetize your game effectively without being labeled as P2W?

TH: I think at the end of the day it's all a matter of degrees and perception, because in the West, the client downloadable games at least are rarely in-your-face pay-to-win, so it's a matter of degree and perception. And for us, because Tribes is known as this high-skill-level game, there are really three rules that we had.

First, you can acquire anything that affects the gameplay by playing the game, so it's really a time-for-money tradeoff. Second, we wanted anything that you could unlock to really be sidegrades -- not necessarily better weapons, but just a different play style. Third, just the way the game is designed, whether it's free-to-play or not, it does really depend a lot on player skill.

Very specifically, you're moving around like crazy, you're having to lead your opponent with most weapons over a very, very large battlefield. So a skilled player with just the free weapons can and will beat a new player that has all the items unlocked just because of the nature of the game.

MH: Nobody wants to be a pay-to-win, but it's almost impossible to define what pay-to-win is, because it's a really personal decision that you're going to make. Some people are going to be real hardliners about it, and will call your game pay-to-win if it has anything that costs real money that will give you a boost to how much experience you earn, or unlock an item that does anything noncosmetic for you.

Other people, I guess, are far more liberal in their definition of what they think pay-to-win is, to where if I have to pay money to buy a tank that does twice as much damage, then that is pay-to-win. The tank, to us, would be a really egregious example of P2W, and we would absolutely avoid doing anything like that. But because it's such a personal decision for the player, it's really hard to make those kinds of determinations.

We've done our best to make sure our business model is completely fair, and I think we have a really fair non-pay-to-win business model that still allows people to make shortcuts and unlock items in a fair way. But of course, since it's your own personal opinion, there are still going to be people out there who say that we're a pay-to-win game. You can't please all the people all the time.

MechWarrior Online

BE: You have to be very creative and disciplined about adding perceived value, without adding too much of any power. We found a nice balance with our Hero Mech design; a unique BattleMech variant can be designed with special properties, which are not viewed as overpowered, and thus not P2W.

When are you "done" building an F2P game? Do you ever really get finished?

TH: That depends. The thing that's the biggest change is that in the packaged-game business, what people defined as "done" actually equals "start" for the free-to-play game business. You're really only starting once you have real gamers using your system. The analogy that I think is accurate that people talk about a lot is that packaged games are like movies and free-to-play games are more like a TV series. So when's a TV series done? Well, sometimes it's a commercial decision, sometimes it's a creative decision, sometimes it's a production-related decision.

MH: For certain aspects of the game, I think you certainly do. We're always striving to achieve balance and create new exciting things for our players to enjoy. So nothing ever really finishes. One of the things I was telling people a lot just before the game shipped is that I have no idea when the game's going to be done. And that isn't really an important question to me because we're continually adding stuff, constantly thinking about things that can go in the game.

So "finishing" the game isn't really a concern. We have years of tasks in the backlog right now for things that we plan on adding to the game, and that can be a bit intimidating at first when new people are coming from more traditional game development into making these longer-term games... It takes people a little while to get used to the fact that you don't just ship a project and then just move to the next product; you're continually working on building out and enhancing the game that you just made.

Once you've gotten used to that, though, it's a great feeling because the sky's the limit. There's never a "We only have a few months to get this game finished" thing. It's a "Well, maybe next year we can really work on that cool thing that's going to take a lot longer than two months out" thing. There's never a closed door.

BE: I'll let you know when it happens! (Perhaps never…) MechWarrior Online is a living, persistent game.

How does a free-to-play game's dev cycle compare with a more traditional boxed/one-time purchase dev cycle?

TH: I think that with free-to-play games you typically go out into the marketplace earlier. Once you feel like your most critical mechanics are there, then you're more likely to go out and have players earlier in the cycle, then you're adding scope, both features and content, for a much longer period of time. The most successful free-to-play games out there continue to get content updates four or five years after the players first started playing them.

MH: One of the things that I love as a designer working on a free-to-play game is that my job every single day is to come into work and figure out ways to make the game awesome, and make the game more fun. Because at the end of the day, if people are logging in every day and enjoying it, then we have a chance to maybe get them to buy some cosmetics, or buy a membership because they love playing the game. That's a really great feeling. I don't really need to worry about making a demo that's going to trick people into buying my $60 game, I just have to worry about making sure that the game is really, really fun -- and as a developer, that's really, really fun.

BE: So far, the free-to-play dev cycle is completely different than the traditional boxed/console model. There's always more work to be done. As time goes on, further refinement of our processes will lead to segregation between live ops and feature development. Eventually a portion of the team will be shared to work on regional versions, and new game concepts.