Although often used interchangeably; masking, pseudonymization and tokenization are different. Tokenization is typically used to de-identify direct identifiers by replacing raw data with randomly generated tokens. The Privitar Data Privacy Platform supports tokenization with optional consistency, format preservation and reversibility.
Within a dataset or analysis, Privitar Protected Data Domains™ enable you to tokenize data consistently so you can maintain referential integrity and enrich data where needed. Consistency is maintained over time so that as new data becomes available or additional data is required, you can add it to the de-identified data in the Protected Data Domain.
Equally importantly, the tokens used by each Protected Data Domain are unique. This approach is exclusive to Privitar and minimizes linkage risks in your datasets, preventing employees or third parties from combining data across Protected Data Domains or linking outside data.
Privitar supports format and pattern preservation using a range of methods to maintain system compatibility and human usability. You decide which format preservation techniques to use in each dataset and context:
Whether it’s for a healthcare intervention, a personalized customer experience or suspected fraud; re-identification can be critical. Privitar lets you decide when de-identified data can be reversed. Re-identification is a privileged role in Privitar, so you can ensure proper oversight and governance over who can see personally identifying data.
Consistency and reversibility are crucial to de-identification. They require a Token Vault to securely store and map the original data values to the tokens that replace them. Privitar supports a range of Token Vault technologies to meet your enterprise standards.