Test Data - - Scramble them automatically! 

GDPR requires that all sensitive personal data are scrambled so that no test data can be attributed to a specific person.

In order to be able to do so successfully, you need to analyze your data first, and then you need to apply a mechanism for
doing so.

And whilst you are optimizing your test data, it is typically also a good idea to look at subsetting your data. Why? Because
you do not usually need to use all your data for testing. Fewer data makes testing quicker and can be just as valid, as 
long as you know that your test data are representative of your production data.

SOSY has a tool from Original Software which can do this automatically for you. Read more here.