Every genealogist starts out in a “collection” phase. You just gather up anything and everything that pertains to your family tree, with very little focus on a particular project. I think this is the right way to do it. You have to have a base to start from and collecting and gathering important items and information from family members now, while they are still alive, is so important. But after a while, you may have noticed, you have quite a “pile” (whether physical or digital, or both if you are like me).

I’ve talked about organizing those piles before in my series about getting organized, so I won’t belabor it again. But you have to get organized before you can really move forward. OR you’ll do what I did and order the same records several times, download the same wills, search for the same obits, etc. Don’t do that. Get organized now, while the piles are small-ish, so you can be more efficient later.
Once you’ve gotten organized, you can start to see where your challenges are, the proverbial “brick walls,” or the research projects that are going to take more effort to solve. Filling in the gaps will require more structured research. When you get to this phase, it is helpful to define your research by establishing good research questions.
A good research question is a well-defined research goal. It needs to be as specific as possible, defining exactly who you are looking for, but not so specific that it isn’t possible to solve. The question has to be answerable for it to work. Here is an example from my own research:
Too broad: Who was the father of Fred Miller?
Well, which Fred Miller. There are about a billion of them in the world. And about a hundred in Wood County, Ohio in 1850. (Ok, I might be prone to exaggeration, but it sure feels like that sometimes!)
Too narrow: What is Fred Miller’s exact date of birth?
Well, for this one, we may never be able to find an exact date of birth due to record loss or just no records at all. The question may not be answerable so it is better to broaden the question a little bit. Perhaps to simply say “when was he born” which could be answered with a date range or something like “about 1805.”
Just right: Who were the parents of Fred Miller, living in Perrsyburg Township, Wood County, Ohio, who was 45 years old in the 1850 census?
With this one, I have identified a unique person in time and place, and I am asking an answerable question, “who were his parents?”
NOW, you can start planning your research to answer this question. Next we will talk about research plans and how to create and execute them.