What I learned at Casual Connect

August 09, 2010 - by badnima

In mid-July the Joyent crew travelled to Seattle for this year’s Casual Connect conference.  This was our first time attending so we partnered with our friends at tmpSocial and descended upon Benaroya Hall. The weather was gorgeous, the venue perfect and it gave us a chance to meet Social Connect’s Jessica Tams as well as a lot of talented people from WildTangent, Playdom, Big Fish Games, Unity 3D, Hi5, Dime Rocker and more.

Taking a break from booth duties, we jumped at the chance to sneak in a few presentations and wanted to share what we learned, along with a bit of our own color commentary.

1. Measure twice. Code once.

Albert Lai from Kontagent gave a concise presentation on the relevant social gaming and web metrics. The message of the day was that viralality, as we know it, and viral growth was dead. With Facebook’s continuing changes to its privacy policy, game developers no longer have carte blanche with viral mechanism and should not expect that success is guaranteed. Instead, as Albert noted, they should focus on ARPU (average revenue per user, or per paying user) and how they can maximize the value of a user.  While this may smack of the old capitalist mantra entering the wonderful freespace of the interwebs, Albert has a point. Gaming is a business, businesses need to make money (would you rather have the biggest game on Facebook, or the most profitable?) so it only makes sense that social gaming developers should track the user actions that most affect their bottom line.

I won’t go into the detail of each metric, but here are my top 5:

  • Viral Conversion – This is a simple measure of the click-through and/or install rate from a single message
  • Entry Event Distribution – The first event or action that a user takes in a session. This single metric tell you more about what got their attention that any other type of market research
  • Engagement – Time on site and page views are both good universal metrics, but if you can normalize it to a ratio of DAU/MAU, then you have a better sense of engagement. DAU/MAU on Facebook can range from 5% (bad) to 50% (awesome!)
  • Exit Event Distribution – This is the last thing a user does before they leave the session (e.g. why did they leave. Why are people churning)
  • Lifetime Network Value – Related to ARPU, this is total dollar value of a user in terms of revenue and opportunity cost and a prerequisite when trying to understand the cost of a particular feature in relation to it’s marginal profit function

The short of it is, if you’re going to create cool new games, make sure you think about what you should track to ensure your game is successful! And if you need help with your metrics, the folks at Kontagent will be more than happy to help.

2. Gaming is dead. Long live gaming.

In a packed room at the Triple Door across from Benaroya Hall, the folks at Playdom presented this year’s Social Gaming trends. They covered broad categories of what’s new and hot, what’s on the edge and what’s waning down in popularity and also shared their thoughts on why and what to look out for.

Gaming genres on the decline included Farming, Pets and Aquariums while Business Sims (restaurant city, café world, nightclub city, baking life) were flat with equal gains and losses. As the first wave of simulation games decline in popularity, the next generation is gaining prominence with genres such as  City Building (My City, MyTown, Social City, My Empire, Millionaire City), Resorts (Happy Island, Tiek Resorts), Tile Exploration (Treasure Madness, Treasure Isle), Casual to Social (Chocolatier), and Sports (Bola, EA FIFA, Playdom’s deal with ESPN titles).

Aside from the obvious market shifts, the speakers explored some of the drivers affecting gaming dynamics. Specifically, they highlighted the problems of rampant cloning and how that affects and ultimately diminishes each genre. In addition, they pointed out to the increasing rate of spoilage (games get old very fast) and discussed some of the unique problems facing developers as they try to get new games to market faster while depending the engagement mechanism to ensure better retention and stickiness.

On game design, the speakers had very concise feedback to developers, some of which I’ve noted here:

  • Engage users with to-do checklists and quest logs to mitigate the jarring transition from tutorial to on-your-own game play
  • Less text. Focus on robust animation and good graphics (albeit this also increases production costs significantly)
  • Flash is not dead (apparently) as it appears almost all games are being presented in flash (I openly wonder how HTML5 will play out on Facebook and other gaming platforms – maybe we just have to wait and see)
  • To deepen engagement, consider mixed genres (e.g., farm + pets + business sims) to keep the game new and fresh. Take existing proven mechanisms, combine in a never before seen way, add a new thing and polish
  • Consider charitable tie-ins as an alternative hook for engagement
  • Look at automation (monetization and offer systems) for a way to give users to bypass boring or rote parts of the game (e.g., buy points instead of earning them with labor)
  • Consider neighborgating: recruiting friends to expand game play across platforms
  • Keep your eye on your profile to extend to mobile as well, with Facebook Connect crossing-over into mobile to provide a better web-to-mobile connected game play

Ultimately, the one thing everyone agreed on was that next year’s boom games are yet to be discovered and that any new viral mechanism created today will certainly be banned or removed by the major social networks this time next year.

We’re looking forward to the fall conferences, so if you’re going to any that are particularly interesting, drop us a line and let us know. Just send us a note to @joyent on twitter.