I did one preliminary software patent application and I also found that it greatly helped me focus and "crystallize" what is the big idea here.
It's kind of like writing documentation but focusing on the core ideas. A patent should not only disclose the technical details of how it works but also explain why it is useful, why it is better than existing approaches.
That is a good way to evaluate the worthiness of the project. Why this, over software that already exist. A project is often useful even if it is not patentable. But writing or even thinking about writing a preliminary patent-application helps put it in context.
If you want to sell an idea to an investor it is not unreasonable that they would demand to know what exactly the idea is, which is best communicated by letting them see the patent application. No?