I like to start a design problem with context, understanding the present situation, taking note of the surroundings. Context tells me what "kind" of solution the problem "wants". From there I often look at constraints next, zoning, codes, approvals needed. I want to know where the danger zones are, where I might waste time looking for solutions that will never be "clean". You can see that my overall approach is about elimination. I am much more comfortable knowing up front where the boundaries are. Infinite possibilities is not reality. There are always boatloads of solutions that won't work well. I like to get rid of them up front.
Next is a quick peek at schedule and budget followed by a rough estimate of size and needs. Then I like to mull the whole thing over while I dig into the size and needs issue in more detail.
I had few projects early in my career where there were serious misfits that took way to long to understand. One was a suburban library that wanted to be one-story because it just wasn't big enough to justify two stories. The site was a truncated wedge shape. The building wanted to be in the narrow end of the wedge and the parking in the wider end. About twenty schemes later I realized I had two problems where I thought there was just one. I had been working on the 10 lbs. in a 5 lb. bag problem without realizing it. Not only did the shape of the site make things difficult, the SIZE of the site was just large enough to accommodate building, parking and setbacks. Things started to fall together once I realized the real limitations.
There is no point in trying to hurry the process while I am working on space an needs (or watching someone else work on it). The whole thing needs to simmer for a while. Something critical to the process happens here. If you rush it, you don't really make any progress. The bigger the challenge, the longer it takes. This probably looks like procrastination, or a learned disfunction from college days. I prefer to think of it as time needed to marinate.
My method of design works much better when I can follow these steps at my own pace.
Take a look at this article to see some other benefits of Unique Methods and a tool that I use. Two other articles on documenting methods are here and here.