Whitaker Trebella on Pivvot

As the GameAnalytics service has been around for quite some time now, there are already many wonderful games out there that are using it. So we thought of kicking off a new series of articles on our blog, which present how some of our clients integrated GameAnalytics in their design flow and eventually found success.

This first article focuses on Whitaker Trebella, who is both a prominent music composer for iOS games and a self-taught game designer. His first game, Polymer, was a critical success that confirmed his talent and enabled him to dedicate more time to programming. His second game followed closely afterwards. Minimalist yet intense, Pivvot, was listed as one of the top 100 apps in over 140 countries and even managed to reach second rank in the US app store. We thus decided to catch up with Whitaker to chat about the challenges of indie development and how using GameAnalytics makes the dev life a bit easier.

Tell us a bit about yourself and your games.

I’m a game developer and musician from Chicago. My goal is to produce artful games with the utmost integrity to inspire the best user and player experience possible. I want my games to be fun more than anything, rather than focusing on the business aspect. I love game design, and I constantly try to reevaluate the decisions I make while designing a game, iterating and simplifying, until something purely fun and artful is left.

Recently you launched Pivvot, can you describe this game a little more for our readers? 

Pivvot is a thrilling game of strategic avoidance that will consistently test and challenge your ability to make quick, impulsive decisions. As the game progresses, you will have to rely on your instincts and problem-solving skills to navigate down the winding path for survival. With its intensity, minimalistic design, and puzzling logic, Pivvot is sure to keep you guessing at every turn.

How did you come up with the idea ?

The idea was not a sudden realization. I set out on a mission to create a game that was beautifully designed and captivating in the best way possible. It was extremely important for me to steer clear of anything cluttered and to ensure that every gameplay feature had an intentional, vital purpose. Staying true to the gameplay style, Pivvot can be controlled with only two buttons. Rather than creating a game that requires in depth instructions and “hand holding”, I tried to create a game in which players are captivated enough to try to figure out Pivvot on their own.

It could be said that it all started with the idea of a ball on a path. I designed a basic prototype with a ball that moved along a path, with the goal to avoid obstacles that were flying around. It was okay but needed something more. So instead of the ball being stuck to the path, it was stuck to a *stick* that was stuck to the path, and it could pivot around. Then, instead of having obstacles that flew around, I changed it so that all the obstacles were stuck to the path. With those basic mechanics in place, I could just iterate a ton until it seemed finished.

Did you consider a freemium approach for your game? 

No. While I have nothing against the freemium approach morally (like some do), it’s just not for me, at least at this time. I believe that putting paywalls up is pretty much universally a detriment to game design. There are good and bad ways to do freemium, but even the good ways just feel like a distraction from good game design. I wanted to give players a premium experience. I wanted them to be able to buy the game and get everything from the start. Further, I wanted the unlock scheme of the game modes to feel like an intentional game design choice rather than a paywall. Additionally, I didn’t want the difficulty of unlocking modes to seem like a business decision rather than a design decision (if game modes could be unlocked by paying, the super hard unlocks might seem like an attempt to get users to pay.)

How did you decide to use GameAnalytics?

Honestly, I was just tweeting about the best things to use for analytics with Unity, and one of my friends (Brice Puls) recommended it to me. I used Flurry with my last game, but with Unity I figured there might be a better native solution. I didn’t want to have to buy a plugin. After looking at your site, it looked like the perfect solution, and it wasn’t too difficult to implement.

GameAnalytics is a very flexible and customizable service and GameAnalytics users can have very different approaches on the data they collect.  Can you describe what are you using GameAnalytics for inside Pivvot?

Mainly, I just collect info about how people are playing the game, and how often certain things happen. For example, which obstacles are the most difficult (high death/survival ratio), which game modes are played the most often, how often do people play, etc. I don’t even look at the data too often, but it’s really nice when I’m curious about how a certain concept is received by players or how many people are actually playing my game.

What are the most surprising things that GameAnalytics helped you discover about your game ?

It’s a simple thing, but I was pleasantly surprised to see how many people are still playing the game even though the Apple Free App of the Week promotion ended last week. 2.5 million people downloaded it during the promotion. While the daily active user count is dropping, it still makes me very happy to see that just yesterday, over 100k people played my game. Very good feeling.

Do you have any advice for developers who are new to game metrics and game analytics in general. 

Really think about how to submit events. There’s a great guide on your site that I should have reread and thought about more, because I am not happy with how I decided to submit stats about Pivvot obstacles.

The way it works is whenever someone dies or passes an obstacle, an event is submitted like this:

Spikeball->Failure or Spikeball->Success

While it seemed like a great idea at the time, it’s not nearly as helpful as I expected. Because now all I can do is look at individual obstacles and see their success rate. If I had done it like Success->Spikeball and Failure->Spikeball I would have been able to see which obstacles people died on or successfully passed the most often, which I think would have been more helpful.