KVKK and Data Residency When Choosing a Domestic Cloud Service
Nanda Teknoloji Engineering Team · · 8 min read
In brief
When choosing a domestic cloud, evaluate together where the data is physically held, who can access it, the provider's status as data processor, the written agreement, security measures, and the conditions for transfer abroad. Being domestic does not on its own mean KVKK compliance; the data controller's obligations continue.
When choosing a domestic cloud service, what is decisive from a KVKK perspective is not the provider’s “domestic” label but where the data is held, who can access it, in what capacity the provider processes data, and whether these are secured by a written agreement. An organization that processes personal data remains the data controller; in most cloud infrastructure relationships, the provider is the data processor acting on its behalf. A domestic infrastructure that guarantees data residency narrows a risky area such as transfer abroad, but it does not remove the obligations of notification, legal basis, security measures, and retention and destruction. This article is not legal advice; it gathers the technical and contractual checkpoints for the selection process. Work with your legal and compliance team for your specific situation, and consult the resources of the KVKK Authority for the official framework.
The relationship between data residency and KVKK
Data residency concerns the geography in which data is physically stored and processed. KVKK defines separate conditions for the transfer of personal data abroad; for this reason, every scenario in which data is hosted abroad or can be accessed from abroad must be assessed separately. Data kept domestically simplifies this assessment but does not close the issue entirely: access to the management console from abroad, backups being kept at another location, or remote access by the support team can each raise an “access” and “transfer” question.
In addition, some sectors (such as public sector, finance, and healthcare) may have extra conditions on data location in their own regulations. Because these conditions can change over time, you need to confirm the current legislation for your own sector with your legal counsel.
Roles: data controller and data processor
In cloud use, the distribution of roles must be in writing.
- Data controller (you): Determines the purposes and means of processing; responsibility for notification, legal basis, data subject requests, and security measures rests with you.
- Data processor (the provider): Processes data in line with your instructions; applies the security measures set out in the agreement, prevents unauthorized use, and reports incidents.
- Sub-processors: Third parties the provider uses for the service (for example, a data center, support, or monitoring service). It must be known who is involved and within what scope.
The agreement should cover these roles, the categories of data processed, the purpose and duration of processing, security obligations, the right to audit, the incident notification process, and the return or destruction of data at the end of the service.
Questions to ask the provider
The table below is a set of questions you can use during the selection and quote process.
| Topic | Question to ask | Why it matters |
|---|---|---|
| Data location | At which locations are production data, backups, and logs kept? | Backups and logs can also contain personal data |
| Access | Under what conditions do provider personnel or third parties access the data? | Risk of unauthorized and foreign access |
| Sub-processors | Which sub-processors are part of the service, and is notice given of changes? | Chain of responsibility |
| Agreement | Is the data processing agreement ready, and what is its scope? | Written instructions and obligations |
| Security | How are encryption, access management, network separation, and logging applied? | Evidence of technical measures |
| Incident management | What is the notification process and timeframe in case of a data breach? | The data controller’s ability to meet its obligations |
| Backup and DR | Where are the backup and the disaster recovery location? | The location of the second copy of the data |
| Exit | How is data returned and destroyed at the end of the contract? | Retention and destruction obligation |
Technical measures: beyond the contract
Written commitments alone are not enough; the measures must have a technical counterpart. The main headings to check:
- Encryption. Data must be encrypted both in transit and at rest; it must be known who holds key management.
- Access management. The principle of least privilege, multi-factor authentication, and logging of privileged accounts are needed.
- Network isolation. Separation of customer environments from one another and private network options should be evaluated.
- Logging and traceability. Who accessed what and when must be traceable, and logs must be protected against unauthorized change.
- Backup and recovery. The backup must also meet the same security and location conditions, and rollback must be tested. For details, see the backup solutions and disaster recovery center pages.
- Security updates. Patch responsibility at the infrastructure layer lies with the provider, while responsibility at the operating system and application layers lies with you, according to the split in the contract.
Example scenario: a decision to move a CRM that contains personal data
An educational institution wants to move its registration system, which contains student and parent information, to the cloud. In the first step, a data inventory is prepared: which fields are personal data, which are special categories, and who accesses them? The providers submitting quotes are asked about data location, backup location, the list of sub-processors, and the incident notification process. It turns out that one keeps production data domestically but stores the backup at another location; the institution assesses this with its legal team and prefers the option in which the backup is also kept within the same scope. The data processing agreement is signed, access permissions are restricted, and the migration plan is prepared accordingly. For the migration itself, the steps in our article on the data center migration plan can be used.
Notification, legal basis, and retention: their relationship to cloud selection
The choice of cloud provider affects the organization’s other KVKK obligations. Moving personal data to the cloud may count as a new processing activity and may require updates to the notices given to data subjects, to records, and to the data inventory. How your retention and destruction policy will be applied in the cloud environment (for example, deletion from backups) should also be considered in advance. Data in backups should be deletable when the retention period expires, or at least should drop out on its own through a time-limited cycle.
A short preparation list for the organization:
- List the personal data categories and special categories of data in the systems to be moved.
- Verify the purpose of processing and the legal basis for each category.
- Update the data inventory and notice texts if necessary.
- Review access permissions again on a role basis.
- Define the retention period and destruction method for the cloud environment.
- Write down what will be done internally in case of an incident and what is expected from the provider.
Relationship to security frameworks
Technical measures for KVKK compliance can also be supported by generally accepted security frameworks. For example, the NIST Cybersecurity Framework, which organizes the functions of identifying risk, protecting, detecting, responding, and recovering, offers a common language you can use when evaluating a cloud provider. These frameworks do not replace a legal obligation; however, they make the questions to ask the provider systematic.
Data residency from a continuity and disaster recovery perspective
The data residency decision covers not only the production environment but also the backup and disaster recovery environment. If the recovery location is abroad, it should not be forgotten that backups and copies are also subject to transfer and access conditions. For this reason, when designing disaster recovery, where the second location will be, how data will be protected there, and what the recovery objectives are should be considered together. When setting recovery objectives, the concepts of RTO and RPO are the basis.
Additional criteria when comparing
When comparing domestic providers with each other, look not only at data location but also at operational maturity. Redundant infrastructure for service continuity, a support team that is reachable in Turkish and, when needed, on an on-call basis, transparency in maintenance notices, and the habit of post-incident reporting are indicators of reliability. In the contract, pay attention to how the service level is measured and what happens in case of a breach. The right to audit, meaning the ability of the organization or an independent auditor to examine the provider’s measures, should also be on the evaluation list. These criteria form a much sounder basis for decision than a selection that looks only at price and location.
Common mistakes
- Mistaking the “domestic” label for compliance. Without a contract, measures, and process, location alone is not enough.
- Not asking about the location of backups and logs. Even if the main data is domestic, copies may be elsewhere.
- Ignoring sub-processors. Everyone in the provider’s chain is part of the evaluation.
- Deciding without a data inventory. Without knowing which data will go to the cloud, risk cannot be measured.
- Skipping exit conditions. Data return and destruction can become a problem later.
- Involving the legal team late. Contract clauses reduce bargaining power after selection.
Decision summary
In choosing a domestic cloud, do the following in order: prepare the data inventory, classify workloads by sensitivity, obtain in writing from the provider the information on data location, access, sub-processors, security, and exit, and review the data processing agreement together with the legal team. For situations where a hybrid placement may be appropriate, see the article on hybrid cloud architecture, and for technical selection criteria, see the article on cloud server selection criteria.
Moving forward with Nanda Teknoloji
On projects like this, Nanda Teknoloji carries out the discovery, design, migration, and operations steps; it clarifies your data location, access, and backup requirements together with you and reflects them in the technical design. For infrastructure options, you can look at the cloud server page, and evaluate your backup and disaster recovery requirements together with the backup solutions.