Writing is like… putting the peel back onto the onion.
Always start a writing requirement with a focused understanding of the task in the form of an outline: the onion already peeled. Who is my audience? What do I need to say? Though an academic writing assignment is generally geared toward a peer group, I always try to figure what the major topics and ideas the instructor grading the paper will be looking for and then consider the perspective of my audience-reader to look for a “hook”. Reading about baseball is more compelling than reading about physics, so sports games are an easy entre to explain some of the basic concepts of physical science. (I guess writing a blog post about an onion wins out over writing about writing.)
Whatever main ideas I am seeking to communicate, they will influence every element of my outline. Usually I form a thesis as I uncover these topics, but it tends to reveal itself before I can find it on my own. I disagree with the technique of focusing the entire effort around a predetermined thesis statement. Sometimes it is the very last thing I do!
Even though I know I am going to start with an outline, each one is different in size and shape before it turns into a real paper. Most of the time it begins with a few ideas numbered in an open Word document. If the ideas get prolific enough I will finally decide to save it – but deleting some of my very first ideas about a thing was the best thing I did for my writing -- and your reading.
When I was a kid we called this brainstorming and it usually looked like a big mess on the chalk board by the time we were all done yelling ideas out at the teacher. Now I call my process “iterative outlining” and it is a bit more refined, but I can't say my notebook doesn't look like a giant mess. Outlining is probably my biggest crutch, although I don't minding dragging it around. Creating a few topic sentences for each main idea with two or more supporting statements is a good way to know that you have enough evidence to support all of your main ideas. Adding layers of data and other supporting information like quotations from secondary sources affirming each topic fleshes out paragraph-like creatures that can be tamed into decent reading if you keep at it. 
As I discover new sources of information that I want to reference in my writing, I’m always revising my outline and leaving the prettying-up for a later draft effort. Outlines are perfect for organizing references to works you’ll need to cite later. Outlining is also a great way to work on a project with a distant due date, but waiting until the latest stages of a project to verify that you have connected all of the pieces with consistency and flow can be a harrowing effort with mixed results. But don’t cry, it’s just an onion.