The five questions that will make your puzzle game better

Schell Games is demoing a devious puzzle title called I Expect You To Die at Oculus Connect, and I was able to solve one of the game’s puzzles — which included dodging a laser aimed at my head, defusing a bomb and then driving a car out of a plane — but became stumped with the second.

It’s a fascinating game, especially with the touch controller. You have to move around and grab things from the environment and see how they work together. There is a sense of play and experimentation in the design; you’re expected to fail and then learn and try again. There is a video of an earlier build below to give you a sense of the game.

In fact, the team told me there were five questions they asked everyone who tested the game. These questions helped them adjust the puzzles and interactions into something that feels natural and, impressively, they promise three solutions to every single puzzle in the game.

What are the questions?

After someone plays the game, they are asked:

1. What was your favorite moment or interaction?
2. What was your least favorite moment or interaction?
3. When did you feel the most clever?
4. Was there anything you wanted to do that the game wouldn't let you do?
5. If you had a magic wand and could change any aspect of the game or your experience, what would it be? Unlimited budget and time.

“I think it was a Marc thing,” Jesse Schell told me I when I asked about it later. “I think it was initiated by Marc [Tattersall, project director at Schell Games]. But it really was out of a desire to get the most useful information out of our play testers.”

The wording of the questions is very particular, in fact. After speaking with Tattersall at the show he insisted on emailing me the list of questions in order to get the wording correct.

“Things like, if you had a magic wand and could change one thing, what would it be,” Schell said. “What I love about that is that it’s very concrete, it’s a very different question from ‘what do you think we should change.’ That’s kind of a very abstract question. But when I put the magic wand in your hand and say how would you change the game, you can imagine yourself doing it, and something about that lets us get better responses out of people.”

These questions, when asked consistently to a large number of people, will tell you what you’re doing right as well as what you’re messing up when tweaking the game. Schell stresses that change is the goal of the process, not a symptom that something is wrong.

“That’s the point of play testing,” he explained. “You want to change things, to figure out what needs to change. And having a fixed number of questions allows you to have some consistency from play testings, and you can see if you’re progressing. Because you’ll have things people bring up all the time as problems, and then when it goes away it’s a big relief, and then it’s replaced by something else, and that’s what progress is like.”

That incredible level of play testing, and Schell said this was the most-tested game he has ever worked on, allowed the team to create a somewhat astounding three solutions for every single puzzle in the game.

“We came up with a rule that every puzzle must have three solutions,” he explained. “We had to push for it, too. You can come up with one, because that’s puzzle design. If you work at it you can come up with a second, but coming up with that third one is tough.”

The trick was to let the solutions come to them through the players.

“Very often we would let it emerge,” Schell stated. “We would someone try something and think that would be a brilliant way to solve it, but our system doesn’t support it. Let’s put in support for that. Usually we would design one [solution] and let the other two grow out of play testings.”

And it all starts with five simple questions.