In Interview, Tools/Technology

PlayFab Monetizes Back-End

PlayFab CEO James Gwertzman.

PlayFab CEO James Gwertzman.

OCT. 28, 2014 • PlayFab is looking to provide backend technology for MMOs.  From the early MMOs onto today’s huge number of freemium online titles, providing for the backend technology and ongoing service responsibilities has never been easy. Virtual item monetization has only added to the complexity. While the availablity of tools and services in this area has steadily grown over the years to make the job less of a hurdle, launching a freemium title can still be a daunting task, especially for small developers.Many developers and publishers are still forced to hunker down and create their own solutions.

Manging the backend proposition for freemium games was seen as a business opportunity by a new firm, Seattle-based PlayFab.  In launching the company this August, chief executive James Gwertzman’s goal was providing a comprehensive source for the building and managing of online games. After eight years with PopCap, Gwertzman was no stranger with the chores of building of 24/7 game services from scratch. Now that he is running a company dedicated to providing back-end services, DFC Intelligence spoke with Gwertzman about how PlayFab is set up and how the company can help developers.

DFC: Until this year PlayFab was in stealth mode. Please tell us more about the company, how it got started, how you developed your technology and how many people work at PlayFab. The last we heard there were under 15 people on staff.

James: PlayFab provides a complete Live Game Operations Platform, with all the tools and services game developers need to build and operate their games successfully. Think of us as Salesforce for gaming – we provide a great solution out-of-the-box, but we also partner with a variety of third-parties to extend our own solution to better meet the full range of needs of our developers.

We’re solving a huge unmet need in the market, since we believe the future belongs to game operators – operations is the new publishing – yet operating a live game effectively is hard and requires a huge amount of technology and know-how to do effectively. Most game developers are ill-prepared to compete effectively in the world of games-as-services, no matter how creative or fun their games. We think that’s a problem, and we see our mission as helping level the playing field – making it easier for developers of any size to launch and operate their games.

CLICK BRIEF ICON to register to get our latest Complimentary Brief

The inspiration for PlayFab came from my experience as an executive at PopCap where I saw first-hand how wrenching the transformation for PopCap was to be from building premium games to games-as-a-service. It was no longer enough to just build super fun games – now they had to worry about a whole host of other factors such as server operations, monetization, scalability, analytics, product management, etc. This transformation was made more difficult by a lack of off-the-shelf solutions. PopCap, like most other developers, was cobbling together their back-end from a mix of third-party “vertical” solutions not designed to work together, and home-grown code.

I presented my thinking at a small game conference in November of last year, and after my talk was approached by the CEO of Uber Entertainment, a Seattle-based game developer. They had developed a lot of the technology that I was describing, and were thinking of commercializing it, but had realized that to properly monetize their technology they were going to have to spin it out as a separate company. I was very impressed by the technology they had built, and PlayFab was born.

We founded PlayFab in January 2014. In return for an equity stake, Uber Entertainment transferred over all of their UberNet technology to PlayFab, as well as full operations responsibility for their existing live games. That meant that from day one of PlayFab we had the discipline of keeping live games operating 24/7.

We spent most of 2014 productizing the technology we had acquired – refactoring the API, creating documentation and SDK’s, enhancing the reliability and scalability, building a self-serve account creation tool, etc. The platform we are offering to developers today is about 50% UberNet and 50% new code we’ve written as PlayFab.

We opened to the public in August and since then have seen strong demand from developers, with more than 500 developers signed up already. We currently have 12 people on the team.

DFC: When you were posted in Asia for PopCap, what did you learn about the free-to-play model dominant in that market that helped you get PlayFab up and running?

James: Running PopCap Asia was perfect preparation for PlayFab, since I got to see first-hand all the challenges of building and then operating live free-to-play games.  There’s an entire life cycle there that all developers go through, but which is not well served today – and that we think PlayFab can address.

PlayFab Logo-SThink about all the stages and the unique requirements of each – no one else today is trying to provide a solution that covers every stage. There’s building the proof of concept and getting the green light, building the real game, testing it in one or two test markets, and tuning until it’s ready for launching in a major market like North America. Other factors are the possibility of rapid growth and the need for scalability post-launch, plus tuning and operations with frequent events and new content updates that need to go global with a successful game. There’s also the lining up of partners and launching in diverse markets like China, SE Asia, Korea, Europe, etc. – each with its own quirks, payment models, channels, etc. You also have to prepare for a sunset period after the game peaks when it’s still making money but can only be profitable if it can be operated by a skeleton crew utilizing as much automation and effective tools as possible.

I also became familiar with the full playbook of live operations, such as in-game offers, frequent time-limited events, monetization mechanics, A/B testing, cohort analysis, content updates, daily reports, and more. All of this has informed the features and design of the PlayFab live operations tools.

DFC: Does PlayFab have competitors, and if so, who are they?

James: Our biggest competitor, by far, are developers still building their own back-ends from scratch, though that’s changing as more and more developers realize the growing complexity of trying to do it themselves, and realize the huge ROI benefit of starting with a pay-as-you-go platform like PlayFab rather than having to make the upfront and ongoing engineering investment themselves.

Our closest corporate competitor is probably GameSparks out of Dublin, though we think we have a better solution on nearly every dimension. Sleepy Giant was also a competitor, though they seem to have recently pivoted themselves in a very different direction. That’s basically it since everyone else in this space is either narrow and vertical, or targeting a single platform. We’re the only company providing the full one-stop-shop solution, and doing it across all platforms including mobile, PC, and console.

PlayFab's Game Manager Interface.

PlayFab’s Game Manager Interface.

DFC: You say you want developers to concentrate on creative not the back-end. Freemium gameplay still demands a lot of design consideration up front to accommodate the monetization, however. You don’t want them coming to you with a finished title that won’t monetize well. How are you advising developers in this regard?

James: Exactly. That’s why it’s so valuable to have a solution like PlayFab available from the very start of development from day one of prototyping. We allow the developer to focus on the creative aspects of their game – and that includes the creative question of how the core game loop will include monetization – and not have to waste any time or energy building the backend, which is a distraction.

We believe that tools like our Game Manager are inherently creative. When you can quickly add a new in-app-purchase item directly from a tool, it allows rapid iteration because the product managers and game designers can quickly try things out themselves instead of having to wait for a programmer to implement their idea. That unlocks a ton of creativity right there.

I saw several games at PopCap that built their own back-end from scratch, and in all those cases the games would have a big buy now button that didn’t do anything until just a few months before launch because that’s how long it took to build the back-end. That’s a terrible way to design and build an online game. You need to test that stuff out from the very beginning.

DFC: How far into the development cycle can a studio bring you their game and reliably be able to have their online back-end handled by PlayFab?

James: Technically you can integrate PlayFab into a game at any stage of development, including after it’s already live, though for sure the greatest value accrues to games that decide to build on PlayFab from the very beginning.

DFC: Please expand on your tiered pricing structure. Is it true that Startups can utilize PlayFab’s platform free for up to 1,000 daily active users? After that threshold is passed, what is DAU cost for casual and core games, plus various server services?

James: Actually we’re exploring some changes to our pricing model. We’re exploring a new model where developers purchase DAU credits up-front, and then use up those credits as the game operates. When they run out of credits, they buy more. The more credits the developer buys at a time, the cheaper the price.

How PlayFab Works-SUnder this model, all developers get a certain number of credits for free and can also earn free credits by doing things like referring other developers, but when the initial free credits run out they will need to buy more.

We’ve also simplified our model so all features are included for the same low price. We just didn’t want to deal with the complexity of having multiple tiers and we never want a developer to avoid using some feature because they don’t want to trigger a higher price point.

DFC: As we understand it PlayFab builds, scales, operates, manages and monetizes online services for games. This can include customer support, in-game marketing, analytics, live customer service tools, in-game store purchases and virtual goods management, multiplayer game server hosting and scaling, software build management, data storage, plus social competitive features. Your goal is to provide these services at a low price but it sounds to us like there is sizable overhead cost. How can you afford to keep your client pricing low?

James: There are huge efficiencies of scale here that we can benefit from – that’s fundamental to this model and why developers are better off partnering with us than building themselves from scratch. Our biggest overhead cost is our own sunk cost of upfront engineering. That’s an investment we’ve already made – and a cost that we think our customers want to avoid having to make themselves.

We also have a monthly fixed operations cost, but as our scale grows, that becomes less and less material until the only cost that really matters is the incremental per-DAU cost – and we’ve set our prices so we can be profitable at scale, even while offering prices that are comparable or even cheaper than what our developers would see if they tried to operate themselves.

DFC: Back in August you said PlayFab had two-dozen clients. How many clients do you have now and please name some of them.

James: More than 500 game developers have signed up for an account. Of those, more than 40 game titles are generating the sort of daily API activity on our back-end services indicative of titles that we expect to launch in the next three to nine months. We are not yet naming individual new developers, however.

DFC: PlayFab received $2.5 million in funding this year. How many clients do you need to sign in order to break even and move toward profitability?

Loadout's in-game store.

Loadout’s in-game store.

James: It’s not a question of how many clients, it’s a question of how much total traffic, since we charge based on DAU. We can reach breakeven with a long tail of many small developers, a handful of big titles, or a blend. We’re pursuing both ends of that spectrum at the moment – we think we can and should satisfy both.

DFC: The PlayFab Game Manager supports iOS, Android, Facebook and PC content. What is the percentage breakdown between platforms of the games you are servicing?  

James: And don’t forget consoles. One of our marquee titles, Loadout, is launching on PS4 in December, and we’re powering the back-end features for that too. That said, mobile is for sure the biggest platform we’re seeing – about 80% of new titles we’re servicing are on mobile. And most of those mobile titles are on both iOS and Android.

DFC: What would it take you to support Linux and are you leaning in that direction?

James: We’re not seeing any requests for Linux support on the client side at all. The next platforms we add will probably be Kindle and Windows Phone – we do get requests for those.

DFC: PlayFab supports game development engines such as Unity, Unreal, Cocos2d-x and Xamarin. What does that mean and how do you integrate with these engines?

James: At a minimum it means we provide SDKs that can be included in those engines and make it much easier for developers using those engines to call and use our services from their games. We’re also looking at going deeper with some engines to make it even easier to use PlayFab – such as building in certain features automatically so you don’t even need to write any code at all.

DFC: Can your clients only use these engines in order to work with PlayFab? If not, what tools do you use to help client developers integrate with your services?

James: No, because we also provide a Restful Web API, our services can be used from pretty much any platform out there. We’re aggressively platform agnostic.

DFC: What international markets can you currently support?


Games from Uber Ent. were some of the first to utilize PlayFab.

James: Currently we have server hosting in the U.S., Europe, Singapore, Japan and Australia. We’re working on a project to include China early next year. China is the big one, both because of opportunity, and because of my own experience and relationships there.

DFC: To deal with different payments systems in different markets how will PlayFab have to adjust?

James: Since most of our traffic comes via mobile or Steam, the problem is simplified. Google Play, AppStore and Steam cover most of the revenue we take in. We do integrate already with AdYen, however, to cover payment methods on PC and web outside the U.S., and we’re looking at other third-party payment aggregators as well. Also for Android we’re looking at partnerships we can do with third-party Android stores – that’s especially important in China.

Start typing and press Enter to search