If using a mailing with a personalized link then each record in the dataset will have a SERIAL that identifies the respondent. In the default opt-in setting the addresses are stored pseuonymous, that means, you cannot see which address belongs to which SERIAL, but this information is still stored on the server.
The obvious option is to work with "personalized" address entries. However, note that personal data requires some additional efforts regarding GDPR. Also note, that pseudonymous data may also be considered as personal data. The same is true for bank connections.