I fell victim to that process when I designed a replacement for our paper-cone rice funnel. Rather than start with the problem (pouring rice into a specific bottle) I started by merely reproducing the makeshift paper-cone drinking cup I had been using. Let's ignore the fact that I left out the opening at the bottom of my first attempt, that was a result of my inexperience with the appearance of holes in Tinkercad.
|Let's just ignore this.|
When the rice clogged the first real (with a hole!) funnel I printed, I started making a series of test holes that began two millimeters larger than the first one. I didn't realize that I was trying to find the minimum-sized hole through which the rice would pass. I completely ignored that what I was trying to build was a funnel for a specific bottle. When I finally realized that (a real d'oh! moment,) I simply measured the inside diameter of the bottle's mouth and built the funnel spout to fit into it.
My earlier failures weren't a failure of physical design, they were a failure of my mental approach. It's mandatory to start by asking the question, "What am I trying to solve?"
When we see students run into that same mental trap, it's important to choose the right moment to ask them that question. Resist the temptation to stand in front of the room and preemptively "teach" them away from the experience. The "problem" we're are trying to solve is not to have students create perfect projects. It's to put them into the position of learning how to independently think their way through challenges.