Soon, BulletFish will enter a large playtesting phase in it’s development. Feedback has been essential for this game. Early on in design I pushed out a prototype I had been sure was a winner, a guaranteed proof of concept. I mostly just expected to be able to share something cool and new I thought I had made.
That prototype was this:
Hard to recognize, I know.
Despite initial encouragement from playtest results, it became clear this project had design flaws that didn’t come through as concerns during concept. Play-testers would make statements along the lines of “these levels don’t really feel like they’re increasing in difficulty” or “there’s only a few levels and I expected to be introduced to more.” The easiest thing would have been to write off their feedback as presumptuous reactions to the limitations of the feedback build. Part of me wanted to say, “well, we aren’t testing all the planned levels or the difficulty curve, we’re testing the core gameplay” and continue from there plugging away on the next build. However, instead of dismissing this feedback I began to think more closely about how my future levels would look and play, fighting the initial feeling that it was somehow premature to analyze that step at such an early stage.
It ended up causing me to look at my design plans more critically and I quickly discovered my design wasn’t going to scale as I intended it to. The extent of my prototype brought to light issues that proved the breadth of features I would rely upon to produce more game content wouldn’t be achievable, all for reasons inherent in the core gameplay. So, I guess the golden rule I took away from the experience is that playtesting is in many ways testing the entirety of the game, not just the feature or system of immediate focus driving the test forward.
Once I accepted that direction was futile, things became a lot better. I reevaluated based on the to player experience I wanted to try and capture I realized I could rework the existing project and the lessons learned into a more proven structure and experiment from there, rather than run with a disaster waiting to happen. Practically a whole different game came out of the experience, but it has made for a better game so I couldn’t be happier.
The Next Round
This time around I’m focused on obtaining more meaningful feedback by making the process passerby-friendly, increasing accessibility for kind, unbiased, internet strangers to play my game. I’m writing a feedback form, one that can be printed or completed digitally. Of course, I’ll still need in-person playtesters so I can see the game being played physically to observe nonverbal tells, and to watch and potentially record live gameplay.
After collecting the data it’s a matter of exploring each bit of feedback to the fullest extent possible. I saw in my earlier efforts how I almost dismissed an important concern as a known problem, as I was biased as the designer and developer. I now know it would be too expensive a mistake to make in the future. Additional work will include analyzing the data to establish a base metric, to see if future builds are making steady improvements to the game experience.