The data you wish to import must first be in a CSV (comma-separated values) file. This is the most generic and universal standard for getting a lot of text-based information from one system to another. In CSV, each record occupies one 'row' of a text file, with all the distinct values in that record separated by a comma (or sometimes a tab character). Pretty much any system you use will allow you to export your data as a CSV, including Microsoft Excel.
Unfortunately not all data will be importable into Magnetracks from your CSV file. Every system stores data differently, and it's one of those things you'll need to accept. For instance, if you're importing metadata info from SourceAudio, that application may store musical genres as a two digit code rather than as text ("happy"). That data will have no meaning - and may not be mappable - to Magnetracks.
You'll save yourself a lot of time and aggravation if you make sure the file you are working with is as "clean" as possible, prior to importing. That means any empty rows should be deleted, any empty values that can be filled in have been, and the formatting of the information all looks consistent. It's much easier to do this cleanup in a spreadsheet program like Excel, than trying to do it one record at a time in Magnetracks.
When importing tracks, make sure any albums and the catalogs they belong to are created prior to importing. That way, when you begin mapping the fields in Step 2 of the import process, Magnetracks will be able to properly connect each track to the album and catalog they belong to.
Note: If a track's album doesn't yet exist, Magnetracks will create a new albums for that track, but it will set default values that you may not like or want. It's much better to set them up beforehand.