Understanding the Fourth-Party Problem
A shared mail and document provider experienced a breach affecting data linked to 22 million people. These individuals weren't direct customers of the provider; they were patients, tenants, and employees of organizations that had outsourced their correspondence, sometimes through another intermediary. This highlights the fourth-party problem: personal data flows through a chain, and your contract only covers the first link.
The Maileva incident serves as a wake-up call. It's not about assuming your processors are weak, but about identifying which downstream parties handle your personal data and what you'd need from them in an emergency.
This guide provides a Fourth-Party Exposure Register and a processor incident notification clause. The register identifies who handles your data, and the clause specifies how quickly you'll be informed of a problem. Together, they help you meet obligations under Article 28 (processor contracts and sub-processor authorization), Article 32 (technical and organizational measures), and Articles 33 and 34 (notification).
Preparing to Use the Template
Before you begin, gather:
- Your Article 30 records of processing to ensure each register entry ties back to a processing activity.
- Your current list of processors and signed Article 28 agreements.
- Each processor's published sub-processor list, if available.
- The name of the person on your team who can authorize a supervisory authority notification quickly.
If your Article 30 records are outdated, update them first. The register's effectiveness depends on an accurate processing inventory.
The Template
Part 1: Register (one row per processor)
Copy this into a spreadsheet or your GRC tool.
| Field | What to enter | Why it matters |
|---|---|---|
| Processor name | Legal entity, not brand | Contracts bind entities |
| Processing activity | Link to Article 30 record | Shows the purpose and lawful basis in play |
| Data categories | e.g. names, addresses, financial details, health-related data | Drives severity if exposed |
| Data subject groups | e.g. patients, tenants, employees | Shows who you'd have to notify |
| Volume and retention at processor | Approximate records; how long held | Long retention widens exposure |
| Sub-processors (known) | Names and processing location | Your fourth-party view |
| Sub-processors (unconfirmed) | Parties you suspect but haven't verified | Flags gaps to chase |
| Article 28(2) authorization | Specific or general; date approved | Evidence you authorized the chain |
| Flow-down confirmed | Y/N: same obligations imposed on sub-processors (Art. 28(4)) | Closes the contract gap |
| Concentration flag | Y/N: does this processor serve many of your peers or sectors? | Spots systemic dependency |
| Notification SLA | Hours to tell you of a personal data breach | Your Article 33 clock depends on it |
| Last assurance review | Date and evidence type | Shows ongoing diligence |
| Owner | Named person | Someone has to chase it |
Part 2: Processor Incident Notification Clause
The processor shall notify the controller of any personal data breach without undue delay and in any event within [X] hours of becoming aware of it. The notification shall describe the nature of the breach, the categories and approximate number of data subjects and records concerned, the likely consequences, and the measures taken or proposed. The processor shall provide further information in phases where it is not available at once, shall require each sub-processor to notify the processor on equivalent terms, and shall not itself notify data subjects or any supervisory authority in respect of the controller's data without the controller's prior written agreement, except where required by law.
Customizing the Template
Set the hours deliberately. As a controller, you must notify the supervisory authority within 72 hours of becoming aware of a personal data breach (Article 33(1)), where it's likely to result in a risk to individuals. Your processor's window must be shorter than that. Choose a timeframe your team can manage, then test it.
Prioritize by consequence, not count. A processor handling sensitive data, like treatment details, needs more scrutiny than one dealing with low-sensitivity data. Sort by data category and data subject group first.
Use the concentration flag. When one provider serves many organizations, an incident can impact you simultaneously with others, dividing the provider's attention. For flagged processors, ask about retention (can old data be purged sooner?) and how they'd communicate during a multi-client event.
Investigate the unconfirmed column. Many controllers know their processor but not the processor's intermediaries. Ask directly, in writing, and record the answer. If a debt collector or platform sits between you and the final provider, that's a fourth party you need to document.
Align with Article 34. If a breach poses a high risk to individuals, you must inform them without delay. Your register's data subject groups column helps you identify who to notify and how. Pre-draft a notice for your highest-risk processing activities, including guidance on spotting impersonation attempts.
Retain control over notifications. The clause above ensures you handle contact with the supervisory authority and data subjects for your data. This prevents conflicting messages and keeps your transparency obligations under your control.
Validation Steps
- Completeness check. Every processing activity in your Article 30 records should map to at least one register row. Gaps indicate missing data.
- Evidence check. For each flow-down confirmed "Y", locate the contractual wording. A tick without documentation isn't evidence.
- Clock test. Run a scenario for one high-concentration processor. Assume you're notified of a breach at hour [X]. Can you gather decision-makers, assess risk, and draft a notification in time? Note any delays.
- Contact test. Ensure each processor's incident contact is reachable outside office hours. A specific contact is better than a generic mailbox.
- Review schedule. Plan a quarterly review for flagged processors and an annual one for others. Revisit any row when a processor changes its sub-processors.
A register won't prevent a breach at a provider you don't control, but it ensures that when one occurs, you know what data was involved, who was affected, and how quickly you'll be informed.




