I have a number of manuscripts (3) saved onto Scrivener. I compiled an ebook from one of my book/manuscript as a test, seeing that editing was not available. I worked on a second book, 60,000 words, and compiled to a Word document. I compiled the second book to ebook, and Scrivener kept compiling the first book. I repeated this 5 times, and each time I tried to compile the current book, the first one came up in the results.
I have not addressed the editing of the ebook because I cannot get the correct ebook to be compiled.
Do you have all three ebooks saved into the same Scrivener project? If so, the way a Scrivener project is structured is that there is one folder that is where Compile starts looking. What this folder is named can be changed depending on what template you used (if any) and if you have re-named it along the way, but it’s a special folder and that’s the root of where Compile will look.
Below is an example from my Scrivener project – I started with the short story template so my folder’s name is Manuscript. Everything under there is part of the actual potential output for the story. Even though I created a folder called “Manuscript 2” at the same level, the icon is different and Compile will never recognize that folder or any of its contents.
It sounds like you may have done something similar – placed a new folder for each new manuscript into the same Scrivener project, as peers to the Manuscript folder.
Now, you can totally have separate sets of output in a single project, but they need to all reside within that special Manuscript folder. You would then probably use collections or your “Include” checkboxes in the compiler to switch between the sets of output. Some people do that because they share a lot of research material and other documents between multiple books and they don’t want to have to replicate that between projects.
