Comparison
Data maskingvsTokenisation
Data masking
the analyst sees the email column as four asterisks and an at sign, while the support tool sees the real thing, from the same table.
Obscuring values at read time based on who is asking, so one table can serve different audiences. It is convenient and it is a control at the query layer, which means it protects nothing against someone with access to the underlying files. Treat it as a way to reduce casual exposure among people who already have legitimate access, not as a boundary against someone determined.
Full entry →Tokenisation
the card number is replaced by a meaningless token, and the only system that can turn it back is one you deliberately cannot query from the warehouse.
Replacing a sensitive value with a surrogate that has no mathematical relationship to it, keeping the mapping in a separate vault. Because the token carries no information, the analytical system can join and count on it without ever holding the real value. It is stronger than encryption for this purpose precisely because there is no key in the analytical environment that could reverse it.
Full entry →