October 2012 marks my 4 year mark at Zynga. One of my most prominent memories is when FishVille launched. I figured it would be appropriate to reminisce on the experience from a developers standpoint. Common legalese here – these are my personal experiences, viewpoints, etc and not those of the company, your results may vary, don’t try this at home, etc, etc… 🙂
Launching a product is one of the greatest feelings an engineer can experience in their career. Whatever product you are on – strive to understand what it will take to get you to launch standpoint and work hard towards that achievement. It is a great experience.
I had been on a team that was responsible for business partnership type work. We tested out new social networks, we built new products such as toolbars, we ran feature tests on older games, etc. Around September of 2009 our group was asked to work on an aquarium sim game. We took up the challenge and laid out a plan.
In early development the game went through several iterations. Designs changed multiple times and the economy received constant tweaks to ensure we had a healthy fish and decoration ecosystem. Believe it or not – in just 3 weeks we had a playable candidate for launch. On the day we were scheduled to launch we met together and made the decisions to wait a few more weeks. A previous game had just launched a few days ago and it was decided to take some key learnings from that game and apply it to ours.
We used the extra time to build out a larger infrastructure. Recent games at Zynga had been growing faster then expected and we needed to ensure we launched with enough capacity. We also polished, polished and polished the game as much as possible. Can we make the fish movements more natural? Is the economy properly balanced? What errors are we facing in the logs?
After a couple more weeks we were ready to launch. We all gathered together in the room we were located in (nicked named the “Fish Tank”) and wen through our final checklist. Turn alerts on, warm the caches, push the latest build, etc.
With a simple push of a button we changed the game from Sandboxed mode to open on Facebook. Within ~15 minutes (and no press) we had several hundred users already starting to play the game. Feeds began posting, game invites started sending… the game was now live.
I wish I could say that we simply kicked back and enjoyed our accomplishment – but like most product launches, several hiccups were encountered. The most prominent issue occurred with our blob encoding strategy which was causing some users to get in a corrupted data state. Through some Herculean efforts our devs were able to muscle through issue after issue and by the end of the day the game was declared stable, healthy and growing.
During the next several weeks our team grew rapidly. We added numerous features based upon user feedback and ultimately got the game up to a peak of 7 Million Daily Active players. I have one memory of my nephew playing the game with his grandmother. They bought and sold fish, helped them to grow. These types of memories made the game special and unique for me – not to mention proud of the great game that we produced.
Well it’s been almost three years for FishVille and Zynga is getting ready to sunset it. It’s been a great game and a blast to be a part of.
It’s amazing to see how much effort has been put into the launching process for games here at Zynga. So much time and effort goes into launching a new game that now allows Zynga to have successful game launches like CastleVille, FarmVille 2, ChefVille and many others.