@MontyJ - sorry I haven't looked at your CSV file but glad you got it figured out.
As for the process - you have two choices. If you use it on your current library, then at step 2 via the matching process you should be able to identify which books are in your library and which are not.
Alternatively, you can firstly create a new library, run the CSV import creating empty books for all those entries, and then use the Find Duplicates "in another library" feature. It really comes down to what you actually want to achieve and personal choice...