The answer to this depends on how your particular instructor sees the process of historical research. Different people identify different numbers of steps in the process and give them different names. I will look at two versions of the steps of historical research, identifying which step in each is the one you are looking for.
In this description of the historical research method, there are eight steps. Here, you would decide what you think your result may be in Step 6. Of course, you would have been thinking about your result in steps 3, 4, and 5. As you read your sources, you would have to be thinking about how they apply to your question. However, you would probably not formally decide what your result is until analyze and interpret your sources.
In this other description of the historical research method, there are only five steps. Here, you would decide what you think your result may be in Step 4. In this step, you are taking and synthesizing all of the material that you have gathered. You are trying to figure out what it is that you have learned through all of your research.
In general, you should be deciding what your result is fairly late in the process of doing your research. You do not want to decide what the result is too early. If you do so, you may make up your mind before you have read all of the relevant information. Then, if you happen to come across information that should affect your thesis, you might ignore it because it goes against the conclusion that you have already made. So, whatever the steps that your instructor identifies, make sure that you are deciding what your result is toward the end of the process.