The Complete Guide to GDPR Compliance for Data Teams
Compliance
24 min read
Complete Guide

The Complete Guide to GDPR Compliance for Data Teams

A practical, engineering-focused guide to meeting GDPR's technical requirements — from Article 30 records to DSAR handling, data minimisation, and breach notification.

GDPR came into force in May 2018. In the years since, regulators across Europe have issued over €4 billion in fines — and the enforcement trend is upward. The most important thing for data teams to understand is that most fines don't result from dramatic breaches. They result from organisations that couldn't demonstrate basic hygiene: that they knew where their data was, who had access to it, and that they had honoured data subjects' rights. This guide covers GDPR from the data engineer's perspective: what the law actually requires technically, and how to build systems that satisfy those requirements durably.

01.GDPR's core principles for technical teams

GDPR is built on seven principles that have direct technical implications. Lawfulness, fairness and transparency requires a documented legal basis for every processing activity. Purpose limitation means data collected for one purpose cannot be repurposed without a new legal basis. Data minimisation means you should collect only what you need. Accuracy requires that data is kept up to date. Storage limitation requires deletion when data is no longer needed. Integrity and confidentiality requires appropriate security measures. Accountability requires that you can demonstrate compliance.

For engineering teams, these principles translate to: documented data flows, enforced retention schedules, access logging, encryption, and an audit trail that can answer 'what data do we hold, why, and for how long?'

02.Article 30: Records of Processing Activities

Article 30 requires every organisation with 250 or more employees (and smaller organisations that process sensitive data regularly) to maintain a Record of Processing Activities (RoPA). This is a documented inventory of every type of data processing your organisation conducts, including: the categories of personal data processed, the purposes of processing, the legal basis, the recipients, any international transfers, and the retention period.

The key word is 'maintain' — the RoPA must be kept current. Most organisations fail this requirement not because they never built a RoPA, but because it quickly became outdated as data flows evolved. The only sustainable solution is a RoPA built on an automated data inventory that updates when your actual data infrastructure changes. DBsafe's RoPA module generates Article 30 records directly from live database scans.

03.Data Subject Access Requests (DSARs)

Under GDPR Article 15, data subjects have the right to request a copy of all personal data you hold about them. You have one month to respond. For organisations with data spread across dozens of systems, responding manually is both slow and error-prone — and regulators treat incomplete DSARs as violations.

A scalable DSAR process requires: a mechanism to identify all systems that might hold data about an individual, a way to query each system for data matching an identity, a process to review results for exempt categories, and a secure delivery mechanism for the response. DBsafe's DSAR module automates the cross-database search: provide a name or email address and the system returns all matching records across every connected data source within minutes.

04.The right to erasure (Article 17)

The right to erasure — also called the right to be forgotten — requires that you delete personal data when a subject requests it and no exemption applies (such as legal obligation, freedom of expression, or archiving in the public interest). The technical challenge is that modern data systems are remarkably good at making copies: backups, replicas, analytics snapshots, audit logs, message queues, caches, and third-party integrations may all hold copies of a record you've deleted from your primary database.

A complete erasure process must identify every location where a data subject's records exist, delete from each location, verify deletion, and produce an auditable record of the erasure. DBsafe's erasure module maps all downstream copies of a record before executing deletion, so you can audit and confirm complete erasure.

05.Data minimisation and retention schedules

GDPR's data minimisation principle requires that you don't store data you don't need, and its storage limitation principle requires you to delete data when the purpose for which it was collected no longer applies. In practice, most organisations are storing significantly more historical data than GDPR permits — because deletion is hard, and no one has explicitly owned the responsibility of scheduling it.

Building a retention schedule means: defining the retention period for each category of personal data, implementing automated deletion when that period expires, and being able to demonstrate to a regulator that deletion is actually happening. DBsafe's retention module lets you define per-column retention policies, automatically purges expired records, and logs every deletion to the audit trail.

06.Breach notification under Article 33

GDPR Article 33 requires that you notify your supervisory authority within 72 hours of becoming aware of a personal data breach that is likely to result in risk to individuals. Article 34 requires notification to affected individuals when the risk is high. The clock starts when any member of your organisation becomes aware of the breach — not when you finish your investigation.

Preparing for breach notification before a breach occurs is not paranoid — it's required. Your incident response process should include: a documented definition of what constitutes a reportable breach, a chain of escalation that reaches the DPO within hours, template notification letters, and a process for quickly querying your data inventory to determine the scope of what was exposed. Teams that have tested this process before an incident respond in hours; teams that haven't respond in weeks.

07.Technical and organisational measures (TOMs)

GDPR Article 32 requires 'appropriate technical and organisational measures' to protect personal data, proportionate to the risk. The law doesn't specify exactly what those measures must be — but regulators, particularly in Germany, the Netherlands, and France, have issued guidance that points to specific technical controls: encryption at rest and in transit, access control and authentication, regular testing and evaluation of security measures, and the ability to restore data availability after an incident.

For most data teams, the priority technical measures are: column-level encryption for sensitive PII, role-based access control with least-privilege policies, database audit logging retained for a minimum of 12 months, regular vulnerability scanning, and tested backup and recovery procedures.

Build your GDPR compliance programme with DBsafe

Get started with a free trial — no credit card required.