Explore our new Cybersecurity Terms Glossary, explained by our experts.

View the Glossary View the Glossary
PRIVIA

Cybersecurity Terms Glossary

What Are Primary and Secondary Banking Systems?

Turkish banking regulation requires banks to run a primary system and a separate secondary system capable of taking over, and to keep both inside Turkey, which has no UK equivalent.

Illustration for the cybersecurity term Primary and Secondary Banking Systems
Home / Blog / Cybersecurity Terms / What Are Primary and Secondary Banking Systems?
17 November 2026

What Are Primary and Secondary Banking Systems?

What Are Primary and Secondary Banking Systems?

In Turkish banking regulation, the primary system is the environment in which a bank actually conducts its operations, and the secondary system is the separate environment maintained to take over when the primary cannot run. The distinction is not merely architectural vocabulary. It is a supervised obligation under the banking regulator’s regulation on information systems and electronic banking services, which has applied since 2020, and it comes with two requirements that surprise UK institutions: a geographic separation rule, and a localisation rule placing both systems inside Turkey.

For a UK group with a Turkish bank or a Turkish financial subsidiary, this is the point where a global consolidation strategy meets a wall. The group’s preferred answer, a primary region in Ireland and a secondary in Frankfurt, is not available.

What the Turkish Regime Requires

The regulation sits alongside other prescriptive duties, including two-factor authentication for banking transactions and an obligation to commission penetration testing at least annually across a defined scope, discussed under mobile banking app security.

How the UK Approaches the Same Problem

British regulators pursue the same objective through outcomes rather than architecture. The operational resilience rules introduced by the PRA and the FCA in 2021 require a firm to identify its important business services, set an impact tolerance for each describing the maximum tolerable disruption, map the resources those services depend on, and test whether it could remain within tolerance in severe but plausible scenarios. Firms were expected to be able to remain within their impact tolerances by 31 March 2025.

There is no UK requirement to hold data in the United Kingdom, and no prescribed secondary site. A UK bank may run critical services from another country entirely, provided it can evidence that it meets its impact tolerance and complies with outsourcing expectations in PRA Supervisory Statement 2/21 and the equivalent FCA material. Since 1 January 2025 a further regime has applied to critical third parties designated under the Financial Services and Markets Act 2023, which brings major cloud and technology providers directly within the regulators’ reach instead of relying solely on contracts between firms and suppliers.

TurkeyUK
Regulatory stylePrescriptive architectureOutcome-based resilience
Data locationPrimary and secondary inside TurkeyNo localisation requirement
FailoverSecondary system required and testedFirm’s choice, judged against impact tolerance
Third partiesOutsourcing rules with regulator access rightsOutsourcing rules plus the critical third parties regime

Making Failover Real

Whichever regime applies, the recurring failure is the same: a secondary environment that exists on paper and has never carried production traffic. Four habits separate the banks that recover from the banks that discover problems during an incident.

  1. Fail over for real, on a schedule. Run live from the secondary environment for a defined period, including a month-end, rather than running a tabletop.
  2. Include the dependencies. Payment scheme connectivity, hardware security modules, card authorisation, identity services, and the network paths to branches all need to work from the secondary site, and they are the things that are usually missed.
  3. Protect the secondary from the primary. Synchronous replication copies ransomware and corruption faithfully. Immutable backups and a recovery point that predates the incident are what actually defeat a destructive attack, as covered under immutable backup.
  4. Treat the secondary as in scope. It holds the same data and often receives less patching, monitoring, and access review attention. Attackers have noticed.

Our emergency action plan services cover failover design and rehearsal, including the dependency mapping that both the Turkish and UK regimes expect to see evidenced.

FAQ – Frequently Asked Questions

Can a Turkish Bank Use a Cloud Region Abroad?

Not for the primary and secondary systems supporting its activities in Turkey, which must be located in the country under the banking regulator’s rules. Local regions operated by major cloud providers have been established partly in response to such requirements. Supporting workloads that fall outside the defined scope may have more flexibility, and that scoping question should be settled with the regulator rather than assumed.

Does the UK Require a Secondary Data Centre?

No. UK regulation sets outcomes rather than architecture: identify important business services, set impact tolerances, map dependencies, and demonstrate through testing that you can remain within tolerance during severe but plausible disruption. Many firms meet that with multi-region cloud deployments, others with traditional paired sites. The regulator cares about the evidence, not the floor plan.

Why Is Synchronous Replication a Problem for Ransomware?

Because it does exactly what it is designed to do. Encrypted or corrupted data written at the primary site is replicated to the secondary within seconds, so both copies are unusable. Resilience against destructive attacks requires a recovery point from before the compromise, held in immutable storage or offline, with a tested path to restore from it.

Is the Secondary System in Scope for Penetration Testing?

It should be. It holds the same data, is often patched less frequently, and frequently has weaker monitoring and access control because it is considered dormant. Turkish banks must commission annual penetration testing across a defined scope, and any firm should include the recovery environment in that scope rather than testing only what is currently live.

Related Terms

You May Be Interested In These

What Is an Alert Prioritisation Model?

Cybersecurity Terms

17 November 2026

What Is an Alert Prioritisation Model?

Read More
What Is Payment Gateway Security?

Cybersecurity Terms

17 November 2026

What Is Payment Gateway Security?

Read More