Now, this might be a subjective answer, so take it with a pinch of salt. If you have never written a research paper before, I would advise to start writing now - don’t worry about the best way to do it yet, as you would be stuck at step 0. Some of the ideas here might sound counter-intuitive - they are, the focus is on to get started and to learn by doing.
Start by writing. In this amazing talk, Simon Peyton Jones from Microsoft Research explains how writing can “quick start” your thought process on your research. Personally, I have experienced writing as a way to think deeply in any area I want to build my knowledge on.
Don’t worry about “path breaking” ideas. Ideas especially at early stages are raw. You will have to spend the time to nourish and develop them. Just start with an idea that can be validated and will be useful.
Don’t try to put in too many concepts in one paper. Focus on one simple construct and build your arguments around it.
Make it interesting! Academic papers need not be boring. If you want your knowledge to spread around, take the time to construct your narrative in a way that is easy and interesting for readers to grasp the information. Introduce your concept, tell why it’s interesting, the idea, how it works and how it will solve the problem. An original and unconsumed idea is a huge loss to science.
Write the abstract at last. By the time you are done with your research paper, you will have a better grasp on what you are talking about. It’s far easier to write the abstract at the end than at the beginning.Use your writing as a way to question your own assumptions and inferences. This way your writing can sharpen your content for the better.
Reach out to people who have already published, understand what it takes to be published. Share your paper with reviewers whom you trust. Usually we spend a lot of time writing and less time for feedback. Take lesser time writing your content, and more time getting it reviewed. Especially if it’s your first paper, you might spend more time on aspects of paper that are not important. A good reviewer can help you shape your paper for the better.
Identify parts of research paper that are more important. Yes all parts are equal, but some are more equal. For example - your idea, its introduction, the problem you are solving - these areas will have more eyeball views than say details or bibliography. Focus on making the argument strong in these areas.
Get it read by your friends and see how easy it is for the information to be grasped and absorbed. Are they getting the idea, the problem? are they seeing the solution directly linked to the problem? Presenting your knowledge in an easy and simple way is as important as the idea itself.
Use tools available to weed out unwanted parts of your research process. For example, use Mendeley to discover and collect all your references. So when you are ready to write your paper you can quickly get your sources in a click. If your research workflow is online you can also use Paperpile. For following author guidelines and formatting, try Typeset.io- it automatically applies all the rules on your content in a click.