Before molecular docking, any compound needs to optimize. But I want to know how the structure of compound can be analyzed before optimization i.e. which method needs to apply and so on.
If you are using the three-dimensional structure from any reliable Database such as ZINC or PubChem then you can use these structures as it is for screening as well as molecular interaction studies (docking), in case you don't have an available structure for a particular compound. You have to draw its 2D structure in chemsketch or any other software, you need to convert it into 3-D using any appropriate tool such as Openbabel or any other tools, that time you need to minimise the structure in a specific force field for that you can use simple UFF (Universal Force Field) or any other.
In order to find out about a ligand structure , you can use several open source platforms e.g., pubchem database, drugbank etc. If you want to know about several pharmacological properties of the structure, which can be very important to prone to be a drug, you can check ADMET properties of the structure by using Swissadme server , procheck server, protox platform etc. You can do this things before doing the molecular docking study of the ligands structure with the receptor molecule.
You need the structure either from a database (such as PubChem) or you can draw it in any free software designed for this purpose (such ChemSketch, Avogardo, ChemAxon -which I personally use).
If you found your compound in PubChem as 3D structure, then it's ready for docking studies. 3D structures in PubChem are already minimized.
If the 3D structure is not available, use OpebBabel to generate and minimize the 3D. It has a graphical user interface for Windows users. Hence, you don't need to write commands. Just few clicks.
But if you are planning to do virtual screening studies, then you really recommended to learn how to use the command prompt to work with OpenBabel through commands. You should read the official documentation of OpenBabel to do things right.