At the moment I am nearing the end of writing one duoethnography and just about to submit another to a journal. One thing that has made things a lot easier for me has been reverse outlining, but in a spreadsheet.
As you can see, if you have images enabled in your browser, I have columns titled Section, Para(graph), Overall (paragraph idea), then three columns labelled Idea. The section is the section header and underneath I have a rough word count. Para is the number of the paragraph, with the Overall and Idea columns being the main contents of the outline.
This approach works for me because it means I can find any topics that are recurrent or tangential, or abrupt subject changes and work them into the piece better or cut them out without losing coherence. This is going to be something that I come back to again and again, I think, particularly in my PhD work.
Reverse outlining might not work on a text that is moribund but for something that needs a polish and a tidy up, it has saved more time than it took to set it up.