-;; If things don't work, you can probably fix it with a field mapping variable.
-;; By default, we use a combination of all predefined mappings, and look for
-;; every known field. If you have data that is from something we've already
-;; tested, try using it's specific mapping table in case that works better.
-;; Here is a handy template to set each of the predefined mapping tables:
+;; Tested to work with thunderbird, gmail, linkedin,
+;; outlook.com/hotmail.com. For those programs, if it's exporter has an option
+;; of what kind of csv format, choose it's own native format if available, if
+;; not, choose an outlook compatible format. If you're exporting from some other
+;; program and its csv exporter claims outlook compatibility, there is a good
+;; chance it will work out of the box. If it doesn't, you can try to fix it as
+;; described below, or the maintainer will be happy to help, just anonymize your
+;; csv data using the M-x bbdb-csv-anonymize-current-buffer (make sure csv
+;; buffer is the current one) and attach it to an email.
+;;
+;; Duplicate contacts (according to email address) are skipped if
+;; bbdb-allow-duplicates is nil (default). Any duplicates found are echoed at
+;; the end of the import.
+
+;;; Custom mapping of csv fields
+;;
+;; If a field is handled wrong or you want to extend the program to handle a new
+;; kind of csv format, you need to setup a custom field mapping variable. Use
+;; the existing tables as an example. By default, we use a combination of most
+;; predefined mappings, and look for all of their fields, but it is probably
+;; best to avoid that kind of table when setting up your own as it is an
+;; unnecessary complexity in that case. If you have a problem with data from a
+;; supported export program, start by testing its specific mapping table instead
+;; of the combined one. Here is a handy template to set each of the predefined
+;; mapping tables if you would rather avoid the configure interface: