Does my company need to comply with GDPR even though we're not in Europe?
Yes. GDPR Article 3(2) applies GDPR to controllers and processors not established in the EU if they process personal data of individuals in the EU "in connection with the offering of goods or services" to them (regardless of whether payment is required) or "monitoring of their behavior" within the EU. If your website accepts EU visitors, processes EU customer orders, tracks EU user behavior through analytics, or provides SaaS services to EU businesses, GDPR applies. Non-EU companies subject to GDPR must typically designate an EU representative (Article 27) and may need to appoint a Data Protection Officer (DPO) if they engage in large-scale processing of EU residents' data.
What is the difference between a DPA and a BAA?
Both are data processing contracts, but they serve different regulatory frameworks. A Data Processing Agreement (DPA) is required by GDPR (Article 28) whenever a GDPR controller engages a processor to handle EU personal data. A Business Associate Agreement (BAA) is required by HIPAA whenever a covered entity or business associate engages another entity to handle Protected Health Information (PHI). If your business is both a GDPR controller and a HIPAA covered entity — for example, a healthcare company with EU patients — you may need both a DPA and a BAA with the same vendor. The documents have significant overlap in structure but differ in specific required content.
What happens if a vendor refuses to sign a DPA?
If a vendor refuses to execute a DPA and you are a GDPR controller processing EU personal data, you cannot legally engage that vendor for services involving EU personal data. GDPR Article 28 makes the DPA mandatory — there is no alternative mechanism. Proceeding without a DPA is a per-se GDPR violation. Practically, most reputable SaaS vendors and technology service providers now have standard DPA templates. If a vendor refuses entirely, it either does not process personal data in ways that trigger the requirement (in which case clarify the scope) or is not prepared to operate as a compliant processor (in which case consider alternative vendors).
Can a company contractually reduce or eliminate GDPR obligations?
No. GDPR obligations are imposed by regulation — they cannot be waived by contract. Contracts can allocate how regulatory obligations are met (which party is responsible for which tasks), but they cannot reduce the underlying regulatory requirements. A DPA that purports to eliminate the processor's security obligations, or that allows the processor to use personal data for any purpose, does not reduce the parties' regulatory obligations — it simply creates a contract that is incompatible with GDPR compliance. Regulators assess compliance with GDPR requirements, not merely with the contractual terms.
What are the CCPA thresholds and how do I know if my business is subject to it?
CCPA/CPRA applies to for-profit businesses (including non-U.S. businesses operating in California) that meet any one of three thresholds: (1) Annual gross revenues exceeding $25 million; (2) Annually buy, sell, receive, or share for commercial purposes the personal information of 100,000 or more California consumers or households; or (3) Derive 50% or more of annual revenues from selling or sharing personal information. If your business is below all three thresholds, CCPA does not apply — but California's data breach notification law (Cal. Civ. Code § 1798.82) applies regardless of size.
What is a Transfer Impact Assessment and when is one required?
A Transfer Impact Assessment (TIA) is required by the EDPB's Recommendations 01/2020 whenever a GDPR controller or processor uses Standard Contractual Clauses (SCCs) to legitimize a transfer of personal data to a third country. The TIA evaluates whether the legal framework of the destination country (e.g., the US) permits the data importer to comply with the SCCs' requirements, or whether surveillance laws and government access rights in the destination country undermine SCC protections. TIAs must be documented and available to supervisory authorities upon request. The complexity of the TIA is proportionate to the risk — high-volume transfers of sensitive data require more rigorous TIAs than low-volume transfers of non-sensitive data.
Can personal data be de-identified to avoid privacy law requirements?
De-identification can exempt data from some privacy law requirements, but only if done properly. Under GDPR, "anonymous information" — data that cannot "directly or indirectly" identify a natural person — is outside GDPR's scope. However, truly anonymous data is harder to achieve than it appears. IP addresses, device fingerprints, and behavioral profiles may be re-identifiable with auxiliary data. GDPR's Article 29 Working Party (now EDPB) has held that many "anonymization" techniques do not achieve true anonymization. Under CCPA/CPRA, "deidentified" information is exempt only if the business cannot reasonably link it back to an individual and has implemented technical safeguards, business processes, and contractual commitments to prevent re-identification. Using a vendor's "de-identified" data product to satisfy privacy obligations requires careful verification of the de-identification methodology.
What is the Global Privacy Control (GPC) and do contracts need to address it?
The Global Privacy Control is a browser-based signal that consumers can enable to automatically communicate their opt-out preference to websites and services — a "do not sell or share my personal information" signal sent automatically. California's CCPA/CPRA requires businesses to honor GPC signals as valid opt-out requests. Colorado, Connecticut, and other states have similar requirements. If your vendor provides an analytics, advertising, or data processing service on your behalf, the vendor's platform must be capable of honoring GPC signals appropriately. Contracts with analytics vendors and ad technology vendors should require GPC signal recognition as a standard capability.
What are the HIPAA breach notification requirements for vendors?
Under HIPAA's Breach Notification Rule (45 C.F.R. §§ 164.400-414), business associates must notify covered entities of breaches of unsecured PHI without unreasonable delay and no later than 60 days after discovery. The business associate must provide to the covered entity: the identification of each individual affected (if known), the date of the breach, the types of PHI involved, the unauthorized person who used or received the PHI, whether the PHI was actually acquired or viewed, and the extent to which the risk has been mitigated. Contracts should tighten the 60-day statutory deadline — 48-72 hours for initial notification, with supplemental information within 30 days — to give covered entities adequate time to fulfill their own notification obligations to HHS and affected individuals.
How long can personal data be retained after a contract ends?
Retention obligations post-termination depend on the applicable framework and the specific data. GDPR's storage limitation principle requires data to be deleted when no longer necessary for its original purpose — at termination, the original purpose (service delivery) no longer exists, so a deletion obligation arises promptly. Legitimate retention exceptions: legal hold obligations, ongoing litigation preservation, mandatory statutory retention (financial records, employment records as required by specific laws). Contracts should specify a deletion timeline (30-90 days after termination), enumerate any agreed exceptions with specific end dates, and require written certification of deletion. Under HIPAA, business associates must return or destroy PHI at termination of the BAA (45 C.F.R. § 164.504(e)(2)(ii)(J)).
Do data privacy clauses apply to employment contracts and employee data?
Yes. Employee personal data is subject to privacy law just as customer data is. GDPR applies to employee data, and several EU member states have employee-specific provisions (Germany, France, Netherlands, and others have stronger employee data protections under GDPR Article 88). California CPRA has fully phased in employee data rights — California employees now have access, deletion, and correction rights with respect to their personal information held by their employer. Contracts with HR vendors, payroll processors, benefits administrators, and employee monitoring software providers must include appropriate DPA/service provider provisions. Employee monitoring (communications, computer activity, location tracking) is subject to both state notice requirements and privacy law obligations.
What is "privacy by design" and does it affect contract terms?
Privacy by design is the concept — codified in GDPR Article 25 — that data protection is built into systems and processes from the outset, not added as an afterthought. Contractually, privacy by design requirements appear as obligations on processors to implement data minimization (collecting only necessary data), pseudonymization (where appropriate), default privacy-protective settings, and privacy-protective system architecture. When evaluating SaaS vendors and technology service providers, privacy by design contractual commitments include: default off for optional data collection; configurable data retention periods; role-based access controls as a standard feature (not an add-on); and audit logging of all access to personal data. Vendors who can demonstrate SOC 2 Type II certification or ISO 27001 certification have typically embedded privacy-protective practices in their operational controls.
What is the difference between GDPR's "legitimate interests" basis and consent, and how does this affect contracts?
GDPR Article 6 provides six lawful bases for processing personal data. Consent (Article 6(1)(a)) requires a freely given, specific, informed, and unambiguous indication of agreement — it must be as easy to withdraw as to give, and it cannot be bundled into contract acceptance. Legitimate interests (Article 6(1)(f)) allows processing where the controller has a legitimate interest that is not overridden by the data subject's interests or fundamental rights, but requires a three-part balancing test (purpose, necessity, balancing). Contracts matter here because: (1) a processor operating under a DPA cannot rely on legitimate interests independently — it processes on the controller's documented instructions and the controller bears responsibility for the lawful basis; (2) if a vendor's DPA permits it to process data for its own "legitimate interests," that is a signal the vendor is acting as a controller for that processing, not a processor; (3) consent-dependent processing creates contractual obligations to preserve and transmit consent records — the DPA should specify how consent signals are stored, transmitted, and honored upon withdrawal.
How do state data breach notification laws differ from GDPR's 72-hour rule?
The differences are substantial. GDPR Article 33 requires notification to supervisory authorities within 72 hours of the controller becoming aware of a breach — regardless of risk level, though notification can be phased if not all information is immediately available. Individual notification (Article 34) is required only for high-risk breaches. U.S. state laws operate differently: all 50 states require notification, but timelines vary from "most expedient time possible" (most common) to specific windows of 30 days (Florida, Colorado, Connecticut, Indiana) to 60-90 days in some states. U.S. state laws are triggered by access to or acquisition of unencrypted personal information — encrypted data is typically exempt. GDPR applies even to encrypted data that was accessed without authorization. U.S. state laws define "personal information" narrowly (name + SSN, financial account, health data, login credentials), while GDPR's "personal data" is broader (any information relating to an identifiable person). For contracts covering both EU and U.S. data, the GDPR 72-hour window drives the processor-to-controller notification obligation, while state law obligations layer on top for U.S. resident data.
What are "data protection impact assessments" (DPIAs) and when do contracts need to address them?
A Data Protection Impact Assessment (DPIA) is a structured process for identifying and minimizing the data protection risks of a project or processing operation. GDPR Article 35 requires DPIAs before commencing any processing that is "likely to result in a high risk to the rights and freedoms of natural persons" — including systematic profiling, large-scale processing of sensitive data, systematic monitoring of publicly accessible areas, and novel uses of biometric or genetic data. The EDPB's guidelines identify nine criteria that may individually or collectively indicate high risk requiring a DPIA. Contracts must address DPIAs in two contexts: (1) Controller obligations — when a controller plans to use a vendor for high-risk processing, the controller must conduct a DPIA covering that processing, and the vendor/processor must contractually commit to providing information necessary for the DPIA and cooperating with the process (GDPR Article 28(3)(f)); (2) Multi-jurisdictional scope — Virginia CDPA, Colorado CPA, Connecticut CTDPA, Oregon OCPA, Montana MCDPA, and Texas TDPSA all require "data protection assessments" for certain high-risk processing, modeled on the GDPR DPIA concept. These state law assessment obligations must be assigned contractually where processing is shared between parties.