Most data-protection remediation is aimed at the findings that read worst in a policy review. The enforcement record, and the new procedural machinery arriving in 2027, shows supervisors landing somewhere far more mundane, and far more expensive to fix once a complaint is already open.
I get asked a version of the same question in most data-protection conversations: where should the money go first. The honest answer is that it should follow the enforcement record, not the risk register. Those two point in different directions more often than anyone admits. The register is written by the people who will be blamed, so it foregrounds the failures that look worst on paper. The enforcement record is written by regulators, and it is remarkably consistent about which technical failures actually turn into findings.
Who this is for
- A CTO or DPO in a firm processing personal data across several EU member states, trying to decide where a finite remediation budget lands this year.
- A board being asked to approve a privacy programme without a clear view of which failures draw enforcement and which merely draw audit comments.
- A compliance owner in a UK and EU group now running two diverging regimes since the Data (Use and Access) Act, and unsure where the two frameworks have quietly parted company.
Where the findings actually land
The public enforcement record is now large enough to read as data rather than anecdote. Enforcement trackers put cumulative GDPR penalties past the seven-billion-euro mark, and Ireland, as the establishment of choice for the large platforms and the seat of the Data Protection Commission, continues to account for the largest share by value. The marquee cases against the platforms are the ones that make the press, but they are a poor guide for an ordinary regulated firm. Underneath them, the findings cluster on a short and unglamorous list.
Free · 4 minutes
When two of your systems disagree, do you know which one to believe?
Fourteen questions on ownership, lineage, and quality — the difference between a number on a dashboard and a number you could defend. Banded finding on screen, full sheet by email.
The first is the record of processing activities under Article 30. Almost every firm has the document. Very few have one that still matches what the systems actually do. A register maintained in a spreadsheet by the DPO, and a data estate that changes every sprint, diverge within a quarter. When a supervisor asks for the Article 30 record and then asks a data engineer to walk a real flow, the gap between the two is where the finding starts. This is a lineage problem wearing a compliance costume, which is why I keep pushing it back onto the technology function rather than the privacy team.
The second is storage limitation under Article 5(1)(e), and its operational twin, the right to erasure under Article 17. Retention sprawl is the default state of any system that was built to keep data and later asked to forget it. Deletion that runs in the primary database but not in the warehouse, the backups, the search index and the third-party analytics tool is not deletion; it is a deletion story. Regulators have become good at asking where else the record lives, and firms are consistently bad at answering it.
The third is international transfers under Chapter V, Articles 44 to 49. This is where the difference between residency and control stops being a talking point and becomes a finding. Standard contractual clauses on file mean little if the transfer impact assessment was never done, or if a sub-processor three tiers down moves data somewhere it never contemplated. The failure is almost never the paperwork; it is that nobody can produce a current map of where personal data physically resolves.
The fourth is breach handling under Articles 33 and 34, and specifically the 72-hour clock. The technical failure that shows up in the findings is not usually the breach itself but the inability to establish scope quickly, because the logging and evidence retention needed to answer “whose data, and how much” was never designed in. A firm that cannot reconstruct the blast radius inside three days will be judged on that, separately from the original incident.
The one-stop-shop was the bottleneck
For cross-border cases the real story of the last few years was procedural, not substantive. Article 60 built a cooperation mechanism between the lead authority and the concerned authorities, but left the actual mechanics of complaint handling to national procedural law, which differs in every member state. The result was cases that sat for years, and public friction between the Irish DPC and other authorities over how aggressively to act. Complainants learned that a finding in principle and a finding that produces a decision are not the same thing.
That is what the GDPR Procedural Regulation is meant to fix. It was adopted by the Council in November 2025 as Regulation (EU) 2025/2518 and begins to apply in 2027, after a transition period rather than on entry into force. It harmonises the parts that national law used to own: what a complaint must contain to be admissible, the information the parties are entitled to, the right to be heard before an adverse decision, and deadlines that force the lead authority to move. It does not change a single substantive obligation. What it changes is the probability that a cross-border complaint against your firm results in a reasoned, enforceable decision, and the speed at which it does. For anyone treating the one-stop-shop as a soft backstop, that assumption has an expiry date.
The UK has quietly diverged
The other structural change is that the UK is no longer running the same regime. The Data (Use and Access) Act 2025 received Royal Assent on 19 June 2025 and is commencing in phases, with the ICO issuing guidance through 2026 and a new statutory complaints-handling duty on controllers taking effect from 19 June 2026. The Act does not tear up the UK GDPR, but it reworks several load-bearing parts of it: a recast approach to solely automated decision-making that loosens the old Article 22 prohibition for most processing while keeping tighter controls for special-category data, a list of recognised legitimate interests, and adjustments to subject-access handling. None of this makes the UK a lighter-touch jurisdiction; it makes it a different one, and a group that runs a single control set across both regimes is now, by definition, out of step with one of them somewhere.
This matters most for firms using automated decisioning and AI on personal data, where the UK and EU treatment now genuinely differs, a theme I have picked apart in more detail on cross-border AI data flows. The practical instruction is to stop treating “GDPR” as one control surface and start maintaining a documented delta between the two.
What this means for the technology function
Read against the enforcement pattern, evidence-led remediation is a narrow set of engineering jobs, not a policy exercise. Reconcile the Article 30 record to actual data lineage, and keep it reconciled by automation rather than an annual spreadsheet review. Move retention out of policy and into the platform, so expiry is enforced by the systems and provable across every copy. Build a current transfer map that resolves to where data physically sits, sub-processors included. And design the logging that lets you scope a breach inside 72 hours before you need it, not during the incident. It is the same discipline I apply to where supervisory inspections actually land elsewhere: the paperwork is rarely the finding, the gap between the paperwork and the running system almost always is.
If you want that gap mapped before a complaint or an inspection does it for you, my advisory work is built around exactly this reconciliation between what a firm has documented and what its systems actually do. Engagements are scoped and fixed-fee against a defined estate rather than billed by the hour, so the cost is known before the work starts.
The firms that get caught out in the next two years will not be the ones with weak policies. They will be the ones whose policies were fine and whose systems never matched them, discovering the difference on someone else’s timetable.
Free interactive tool
Website compliance checklist
What your site has to do, based on what it actually does
Answer as much or as little as you like — the list builds as you go. Nothing is stored against your name and no email is required.
Everything that applies
Ordered by what to do first: legal requirements you can close quickly, then larger pieces of work, then what is expected rather than required. Not exhaustive, and not a legal audit.
Dated PDF, yours to keep or circulate.
Free interactive tool
Interactive deadline calculator
Check which regulations apply to you and when
Regulation across the EU, UK, US and Asia-Pacific has moved considerably in the past eighteen months, and several headline dates have shifted more than once. Twelve questions, about three minutes.
Results are shown on screen — no email required. A dated summary is available to download, and can be sent on if that's more useful. What we do with your answers.
Most technology problems are not technology problems. They are control problems.
The systems exist. The investment has been made. The question is whether leadership can understand, direct, evidence, and sustain what those systems produce. Find out where control exists — and where it only appears to.
Full Governance by Sixteen Pillars
Govern your business. Prove your compliance.
A board assurance cockpit for EU-regulated financial firms — tamper-evident, hash-chained proof of governance across DORA, GDPR, NIS2, ISO 27001, the EU AI Act and MiCA. In development.
See what's coming