Sandbox seeding versus full-copy masking
Key Takeaway
To remove production PII that is already in a Full sandbox, create a Data Mask & Seed masking policy and run the masking job in that sandbox. Seeding works differently because it copies in only a chosen subset of records.
Testing and training sandboxes often need data that looks like production data. Copying production data can expose personal or sensitive information. There are two main ways to handle this risk. Masking replaces sensitive values with fake ones inside a sandbox that already holds copied data, such as a Full sandbox. Seeding copies only a chosen set of related records into an org and can anonymize them before they arrive.
Why a Full sandbox needs protection
A Full sandbox contains complete production data and configuration. Any personally identifiable information (PII) in production therefore lands in the sandbox too. Leaving that data open to the whole development team without access controls or data masking is a poor practice. Limit Full sandbox access to the roles that need production data.