A user changes the email address of a contact in your CRM - Let's say SigParser creates a contact john@example.com in your CRM. One of your users finds another contact john@fake.com and decides to change the email address on that contact to also be john@example.com. If your CRM doesn't have duplicate prevention turned on then your CRM will now have two contacts with the same email address.

Dual imports - Let's say SigParser is running a sync job but at the same time you're running a bulk import in your CRM already. If both SigParser and your CRM bulk import attempt to import the same email address for a Contact record within 1 second or less of each other a duplicate could be created. Duplicate detection on CRMs like Salesforce often catch catch this sort of scenario because the timing is too tight.

Using non-standard email address fields - If your CRM doesn't use the standard email address field in your CRM then SigParser won't know that a contact with that email address already exists and will attempt to add the contact. This is something you'd probably notice quickly while setting up your CRM.

CRM connection user can't see all Contacts in CRM - If for some reason the user that connected the CRM to SigParser cannot see all Contacts then SigParser won't know a Contact already exits with the same email address. Make sure the user that connects SigParser to the CRM has full visibility of all Contacts.