Monday, March 31, 2008

QQ Games, the world's most popular multiplayer real-time casual game service is failing in the U.S. Why?

In China, QQ Games have reached up to 3 million concurrent players (that means players online at the same time).

In the U.S., this afternoon, after being live for over a year, QQ Games has just under 1500 concurrent users.


In the U.S., QQ Games takes the form of a plugin for AOL Instant Messenger. AOL has 30 million active users a month. So out of 30 million active users, QQ Games attracts 1500.

Wow. Again.

So what's the deal? How come one of the world's most popular multiplayer real-time game services is failing to take off in the U.S.?

You're expecting an's some possibilities:

1. It's a downloadable plugin. That usually hampers adoption somewhat, but it's a stretch.
2. AIM users don't know about it. Nope, it's prominently features on the AIM homepage. And it's offered as an option when you install AIM!!!
3. The games aren't fun. Maybe...but nope, they're fun. However, they are NOT social. They are multiplayer puzzle games that don't require any interaction between players.
4. Americans don't like playing against strangers. Maybe...if so, it's a problem for all multiplayer real-time games. And besides, they have an option to invite friends to play.

I think it just hit me. The games on QQ Games are not really any more fun in multiplayer mode then they would be in single-player mode. In fact, they feel like single-player games. For instance, they have a Bubble Bobble Clone called Robo. If you played Bubble Bobble in the arcade, it's awesome because you can hear your opponent groan when you send a barrage of bubbles down on her head. In the online version, you don't get any feedback, it's boring. You opponent isn't even going to send a chat message because they're too busy playing.

For multiplayer action games like the ones on QQ games to be as fun as the arcade originals that they're modeled after, it'll require audio chat.

However, I think the core reason they haven't had the user uptake of their parent company is because they have yet to fully implement their virtual economy. QQ Games does have gems that you earn by playing games, but at the moment those gems can't be used to buy anything, which makes them meaningless. Offering virtual goods in exchange for the gems gives people a reason to play over and over again. It gives purpose to gameplay. Purpose fuels retention. The problem, of course, is providing a context in which the virtual goods are meaningful. That's easy when you control the platform as in China, not as easy when you don't (a point I should probably touch on in a future post).

So while nothing can make QQ Games social (except perhaps adding some social games...duh), implementing the rich virtual economy of the Chinese parent will increase usage significantly, I suspect. I'd wish them luck, but their parent company, Tencent, pulled in a half a billion dollars last year, so as they say...who needs luck when you have 500 million dollars in the bank.

Monday, March 24, 2008

Is Mytopia the new Bunchball?

Dean Takahashi over at Venturebeat has a fairly glowing article about Mytopia, a new entrant to the casual games arena. Having played Mytopia, it strikes me as Club Penguin meets Club Pogo. Not a bad pitch, eh? No wonder they got angel money.

Mytopia's main differentiator: it's cross-platform, i.e. you can play simple casual games like Chess and Poker against friends on Myspace and Facebook.

Mytopia sounds a LOT like Bunchball, the real-time cross-network casual games platform (now known as Karma Games and Avatars on Facebook).

Bunchball is also built on Flash, also embeddable anywhere, and also cross-platform (Facebook, Bebo). It also has avatars, achievements, a currency, virtual items, and a leaderboard. So how successful has Bunchball been on the old socnets? Not very. It has never exceeded 11,000 daily active users on Facebook despite having the benefit of being one of the first apps on Facebook and the ideal url:

The fact is: creating a community around real-time games is hard. If Mytopia's strategy is to build that community on the back of social networks, then they have a difficult road ahead them. Only one real-time game has managed to have significant traction on the socnets, Zynga's Texas Hold-em Poker. And I would give my left eyeball (metaphorically, of course) to hear from Mark Pincus about how he managed that amazing feat.

If Mytopia wants to succeed if should take a lesson from World of Warcraft. The designers of WoW discovered quickly that not everyone enjoys playing with other people, especially not at first, so they put tons of single-player quests into the game. That way people could play WHEN THEIR FRIENDS WEREN'T ONLINE and still enjoy themselves.

It seems to me that providing a compelling single-player experience IN ADDITION to a compelling multi-player experience is the only way to get a world built around casual games off the ground.

Having said that, for a single-player game experience to succeed on the social network, it needs to have a significant social component, such as challenges (think Jetman) or a leaderboard(every other game on Facebook). And in the case, of a leaderboard, it's more successful if the game is built around a quality people want to brag about such as intelligence (such as in the case of Who has the Biggest Brain- currently just under 200,000 DAU).

To summarize: Mytopia, good idea, good execution...good luck.

(BTW, for readers of the blog, I assume a lot of things about you: 1. you know all the companies I talk about. 2. You know about the games I reference, and 3. You're currently running a game company or investing in one. If I'm wrong about this, let me know in the comments and I'll try and offer more introductory details about the things I'm talking about.)

Thursday, March 20, 2008

How to Scale a Facebook Game to 300 million pageviews a month

While tracking down a presentation that Friends for Sale did, I ran across this excellent interview they did about scaling their backend at It's fairly technical, but worth the read for anyone planning to handle 600,000 users a day (as I'm sure you all are).

Sample question from the interview:

What is your in/out bandwidth usage?

We used around 3 terabytes of bandwidth last month. This month should
be at least 5TB or so. This number is just for a few icons and

Yep, I just added 600,000 users = ~3 terabytes of bandwidth to my assumptions list in my budget projection, too. :)

Scrabulous back on top.

Friends For Sale's reign as #1 game on Facebook lasted all of one day. Scrabulous has been up by ~20,000 daily active users all week.

Scrabulous: 687,955 DAU (up 50,000 DAU from last week)
Friends For Sale!: 664,587 DAU (up 21,000 DAU from last week)

I don't plan to cover this story blow-by-blow, but since I did announce Friends For Sale's triumph last week, I figured I should keep you all updated.

In other news...I found an excellent presentation by the creators of Friends For Sale. Enjoy!

Tuesday, March 18, 2008

Friends For Sale! displaces Scrabulous as the top game on Facebook

That was quick. Last Tuesday I predicted that Friends for Sale! would be the top game on Facebook. At that point, Friends for Sale! was about 60,000 daily active users behind Scrabulous. As of today, Friends For Sale! is up by 11,000 users making it the top game on Facebook

Friends For Sale!: 643,877 DAU
Scrabulous: 632,372 DAU

In the long run, I think Scrabulous will return to the top. After all it's been on Facebook since last June and still holds on to 23% of its audience daily. It's around for the long haul.

It's yet to be seen if Friends For Sale! has that level of retention. It's still acquiring new users at a rapid rate which may could be masking any retention problems they might have (see Andrew Chen's excellent essay on this phenomenon).

So I until I see how things shake out, I'm not declaring a victory for social game design, but I will say this: if you're still porting old games from other platforms, you may want to rethink that strategy.

Here's an excerpt from an interview from 2006 with Trip Hawkins, founder of EA and mobile games company, Digital Chocolate, from the Hollywood Reporter that pretty much sums it up:

THR: But isn't it important for the cell phone deck -- or menu -- to carry instantly recognizable licensed game titles? Is anyone going to want to play, say, "Alien Shoot" when, say, "King Kong" or "Harry Potter" is available to them?
Hawkins: You know, we went through exactly the same phenomenon in the '80s. Atari paid $20 million for the rights to build an "ET" game for the Atari 2600 console figuring that such a game couldn't miss. Then they gave the programmers seven weeks to build the game and, of course, it never lived up to the public's expectations. It sounds like mythology but it's true -- Atari had so much excess inventory of the stupid game that it had to bury them all in an Arizona landfill site.
Okay, so I just like the E.T. reference. In any case, he has a lot to say about the folly of not designing for the platform (in his case, mobile). Read the interview and take heed.

Friday, March 14, 2008

Interview with the Agarwalla Brothers, creators of Scrabulous - Highlights

No, not my interview. The guys from Developer Analytics conducted it (I'm starting to feel like a pimp for those guys I mention them so much.)

Anyway, you have to be a member to read the interview(conducted March 8th), so I thought I'd post a few highlights.

On scaling: ...we have re-written the code from scratch exactly four times. When you have such a high traffic load, it’s not just about adding more hardware. You need to improve your software as well to complement that.

On metrics: ...Well, in the Facebook space, the number of installs isn’t really that meaningful. Instead, we really track app activity, such as return usage. In fact, even metrics measured on a daily basis really aren’t that useful. For instance, activity drops quite a bit on the weekends and also during holidays, so it’d be much more useful to see metrics on a weekly basis. Also, it’s interesting that Facebook tracks unique adds for the current day and include them as a ‘user,’ but if a user just added an app, is he/she really a user? What is a user? Someone who uses the app 5 times? 10 times? There’s really no standard there. Another thing I’d love to see is – sure, Scrabulous has ~3 million installs, but how many of those installed users actually have come back?

On marketing: [we] haven’t spent more than $500 on marketing.

On ad networks: We use four ad networks: VideoEgg, Google Adsense, SocialMedia, and With VideoEgg, they have the highest payout by far, but they don’t have the inventory to handle the kind of impressions a game like Scrabulous gets.

On social networks: Yeah, we’ve looked at Bebo and hi5 which is coming out soon. Though honestly, we really like Facebook. We think things on the other platforms are still a bit disorganized. We really think Facebook is very clean and organized, so we think we will be there for the time being.

My favorite factoid: We [Scrabulous] currently have about 25 full-time developers.

Twenty-five full-time developers!!!! Last I checked SGN only had six employees! Though I'm sure with that 10 million in funding SGN has been doing some hiring lately.

Tuesday, March 11, 2008

Friends For Sale: Designing a Game for the Platform v. the Channel

First, what is Friends for Sale?

Here's the description from the developer:

Buy and sell your friends as pets! You can make your pets poke, send gifts, or just show off for you. Make money as a shrewd pets investor or as a hot commodity! Friends for Sale is the bees knees!
Why you should care?

As of today, it has more daily active users (668,080) than Texas Holdem Poker and almost has as many daily active users as Scrabulous. If it is a game, then it's the second most popular game on Facebook.

So Bret, is it a game?

Here's the four key components of a game as defined by Wikipedia (feel free to dispute): goal, rules, challenge, interactivity. Let's break it down.

1. Goal: to have the highest worth among your friends.
2. Rules: you can only buy friends if you have enough money.
3. Challenge: To increase your worth you have to invest in the right friends
4. Interactivity: Other people can buy your friends from you and vice-versa.

And finally, my criterion: is it fun? Answer: ask the 668,080 people currently playing.

So yes, it is a game. In fact, I think it's an example of the new breed of social games emerging that will actually incorporate the social graph into their gameplay and not simply use the social graph as a distribution channel.

Platform v. Channel

In many ways, I think game developers are treating social networks as a channel for games rather than as a platform for games. Most games on Facebook only use the social graph to acquire users (i.e. Invite 40 Friends and get a Chuck Norris themed Jetman!). Sell your Friends! could not exist without the social graph.

A channel is just a way to distribute something. As a channel, Facebook is amazing, allowing any developer with a good game to *potentially* reach million of users. However, the vast majority of games don't. Roughly thirty games out of ~2000 games listed in the games sections of the app have reached one million installs (which, by the way, is more than I expected).

A platform has unique features that can be leveraged. Take the Nintendo Wii for example, it's unique feature is its motion sensing controller. Making a game for Facebook that doesn't leveraging the social graph (Facebook's unique feature) is like making a game for the Wii that doesn't leverage its motion-sensing controllers.

Just something to think about. There's a ton to say about *how* to design a game to take advantage of the social graph, but it'll have to wait until after my fever breaks.

Meanwhile, two predictions: Sell Your Friends! will soon be the most popular game on Facebook (as measured by DAU) and it's going to be copied endlessly.

Tuesday, March 4, 2008 Entering the Social Gaming Arena

Who the hell is you ask?

Epillager is a company that buys underperforming apps, mostly quiz apps, and resells them after acquiring tons of new users.

How do they acquire these users?
1. Cross-promoting the apps on their other apps.
2. Forced invites.

I've always thought their arbitrage strategy was brilliant, however it relied largely on forced invites which now have been outlawed by Facebook. That really threw a wrench into their business model.

Now, they're moving in the social gaming arena.

They say they're sitting on an install base of 3.5 million users. However, according to Adonomics, their daily actives is under 20,000 and shrinking. Will it be possible to bring their install base over to the new social games apps they plan to release without violating Facebook's TOS (i.e. using email to cross-promote other apps)? For their sake, I hope so.

I suspect we're going to see a lot of other Facebook app developers move into the social gaming arena. After all, what other kinds of app developers are getting millions of dollars in VC funding?