In short
Data privacy is the discipline of collecting and using personal data in line with the law and with user expectation: a lawful basis for processing, minimization to what is needed, transparency about what happens to the data, and respect for user rights. It is distinct from data security, which protects data from unauthorized access, and a marketing stack needs both, because lawful data that is poorly secured and well-secured data that is unlawfully used each create risk. For marketers the practical translation is that privacy is a design constraint rather than an afterthought. Decisions about what to collect, how to hash it, which consent to require, and where to store it are privacy decisions that also determine measurement quality. The stacks that handle privacy well tend to measure well, because the same disciplines, minimization, consent enforcement, and clear data flows, produce clean, defensible signal. Privacy designed in is compatible with performance; privacy bolted on afterward usually costs both.
Privacy versus security
Privacy asks whether you should hold and use the data at all; security asks whether anyone else can get at it. Encryption and access controls satisfy security. Only lawful basis, minimization, and honored user rights satisfy privacy.
Compliance by design
The cheapest time to build privacy in is at pipeline design: hash at source, collect consent with the data, keep field-level control over what each destination receives.
What does privacy-by-design look like in a stack?
A brand redesigns its collection so it gathers only the identifiers it activates, hashes them at source, and records consent per purpose. The result is a stack that is easier to defend to a regulator and cleaner to match on.
Reference: European Commission, data protection