"The readiness is all."
My advice is to always decide the night before what you will be writing on any given day. Further, I propose not to bring any source materials, whether that be books or articles you've read or primary sources you're writing about, with you to the writing session. (I've noticed that this is an important difference between my approach and, say, Jonathan Mayhew's.) You should only bring the notes you've made in preparation for the writing you are about to do. When you write, you should be writing down something you know, indeed, something you already knew quite comfortably at least last night. You should not be discovering what you want to say in the act of writing. This requires a clear separation of the research process and the writing process, i.e., the process that generates ideas and the process that commits them to the page. Many scholars find this to be quite a radical idea.
Authors I work with sometimes say that my approach very quickly draws attention to what happens in preparation for writing. To plan their writing also means to plan the work that happens the day (or days) before any particular paragraph is to be written. And I am sometimes asked whether I have any advice about how to organize that time as well. I am always hesitant to say something about this, however, because in the end nothing should depend on advance preparation. You've decided to write something you know down, that is all. How you came to know it is none of my business. At the end of the day, no matter how you conduct your research, there are things you know, and all I am saying is that you should devote between one and six half hours to writing them down in coherent prose paragraphs.
If one of those sessions does not go well, some writers tell me that this reveals their lack of preparation. They thought they knew but had not worked hard enough, or effectively enough, the day before to make sure. They articulated a claim, but they discovered in the act of writing it that they were unable to support it adequately. This week, I will offer some suggestions for how to structure the non-writing research time (ideally, work done in the afternoon). But I want to emphasize that I do not, in fact, believe that such organization is necessary. If you choose something you think you know well enough to write a paragraph about every evening, and then write that paragraph every morning, you will become better and better, not just at writing such a paragraph, but at choosing which paragraph to write.
My aim is Socratic. I don't want to help you become more knowledgeable. I want to help you better distinguish what you know from what you don't know.