A regulator asks where your data is held. Somebody checks Atlassian Administration, sees the app pinned to a region, and reports that the data is in country. That answer is incomplete, and occasionally it is wrong in a way that matters.

Data residency does something specific and worth having. It also leaves categories of data outside the boundary by design, covers five products out of the suite, and for two of the four markets we work in it offers no region at all. All of that is documented. None of it is obvious from the admin screen.

Where you can actually pin data

Atlassian offers twelve locations, one of which is Global. The eleven real regions are Australia, Canada, the EU, Germany, India, Japan, Singapore, South Korea, Switzerland, the United Kingdom and the USA.

Read that list for what is missing. There is no Middle East region. No Saudi Arabia, no UAE, nothing in the Gulf. An organization in Riyadh or Dubai that needs Atlassian data held in country cannot do it through data residency, because the option does not exist.

There is a second trap for Indian organizations, and Atlassian flags it twice in its own documentation: India is not assigned as the default data residency location, even for organizations based in India. It has to be set manually. An Indian company that assumed geography produced the right default has data sitting wherever the Global setting put it.

Everything unpinned sits in Global, where Atlassian determines the location dynamically and moves data between regions as needed. That is the default for every app until somebody changes it.

What stays global whatever you choose

This is the part that surprises compliance teams. Several categories of data cannot be pinned at all, in any region, on any plan.

User account information is the largest of them. Names and email addresses are held globally by design, so that account lookups work for users in any region. If your regulator's concern is personal data rather than business records, the thing they care about most is the thing residency does not move.

The rest of the out-of-scope list is long and specific. Data in transit, for up to 30 days. Cached content such as emails and notifications, for up to 30 days. App, audit and operational logs. App analytics and user analytics. Team profile information. Third party app integration data. Automation flow configuration data. For Jira, connected DevOps data including commits, branches, pull requests, builds and deployments. For Jira Service Management, customer accounts. For Confluence, permission and restriction configuration data, and space keys. Atlassian Guard's audit log events cannot be pinned either.

Atlassian states the boundary plainly in its Isolated Cloud documentation: on Commercial Cloud plans, data residency coverage typically only extends to in-scope data stored at rest for more than 30 days. That sentence is worth taking to whoever signs off your compliance position.

The products that are not in the scheme

Data residency covers Jira, Jira Service Management, Jira Product Discovery, Confluence and Loom, on Standard, Premium and Enterprise plans. Products outside that list are outside the scheme.

Rovo has no residency setting of its own. It follows your Jira data, or your Confluence data if you do not run Jira, which means your AI chat session logs and agent configuration land wherever that app is pinned.

One more constraint catches multi-entity groups. Residency is configured at the app level and cannot be set per project, per client or per user. If one business unit has a residency obligation and another does not, the app moves as a whole or not at all. Organizations using units manage residency at the unit level instead, and any app that does not support the unit's pinned location hosts its data outside it.

What moving actually costs

Changing region is not a toggle. Atlassian states that moving data involves downtime of up to 24 hours depending on data size, and that search may be disabled during re-indexing for up to 3 days.

Plan that as a migration rather than a setting change, with a maintenance window and a note to users who will find search missing for days afterwards. Backups follow the data into the chosen location.

One region carries a warning worth repeating. Pinning to Switzerland makes the Atlassian Data Lake unavailable, which removes Atlassian Analytics charts and dashboards and breaks Assets reports embedded in Jira Service Management. For Opsgenie reports embedded in Jira Service Management, Atlassian states the limitation persists and that you will not be able to access that data even if you move back to the previous location. Read the limitations for a region before you pin to it, not after.

When there is no region for your jurisdiction

This is the live problem across the Gulf, and the honest answer is that it has three imperfect routes. The first is to accept Global and build the compliance position on contractual and certification grounds rather than geography. That is viable in more cases than people assume, and it should be tested with counsel before it is dismissed.

The second is Isolated Cloud, where residency is architectural rather than configured. An instance is provisioned in a region and all Jira and Confluence app data is pinned there from creation, the boundary extends to data in transit rather than stopping at data at rest, and the instance cannot later be moved to another region. The constraint is that it is fixed to the region of creation, so the region has to exist in the first place.

The third is self-managed Data Center, which puts the data wherever you put the hardware. That has been the Gulf answer for years, and it now has an expiry date: Atlassian ends Data Center sales to existing customers after 30 March 2028 and the products reach end of life on 28 March 2029. We set out what that timeline means in our Data Center end of life roadmap.

Organizations that chose Data Center for data location reasons now have a decision with a date on it. It is a different decision from the usual migration conversation, because the driver was never cost or features.

The pattern underneath

Residency is a control, and like every control in this set it is only as good as the understanding around it. Knowing which data it covers, which products it reaches, what moving costs and what your jurisdiction actually requires are four separate questions, and the admin screen answers none of them. That is the argument we make across the estate in our guide to enterprise security posture.

Case studies

Tawuniya. Saudi Arabia's first national insurance company, regulated by the Saudi Insurance Authority and operating more than 60 insurance products. Alongside an ITSM and ITIL 4.0 maturity engagement, we closed security vulnerabilities from outdated product editions and brought its data hosting into line with Saudi Arabian regulations.

ZATCA. The Zakat, Tax and Customs Authority is Saudi Arabia's regulatory body for zakat, taxes and customs, and a Vision 2030 digital transformation program. We migrated its Jira, Confluence and Bitbucket data out of a third party's environment into the authority's own Atlassian ecosystem while maintaining data integrity and confidentiality, then configured and optimized its disaster recovery system for swift failover.

If your compliance position rests on a region setting, talk to us about an estate review before an auditor tests it.

Frequently asked questions

Which regions does Atlassian data residency support?

Twelve locations including Global. The eleven regions are Australia, Canada, the EU, Germany, India, Japan, Singapore, South Korea, Switzerland, the United Kingdom and the USA. There is no Middle East region, so organizations in Saudi Arabia and the UAE cannot pin Atlassian cloud data in country.

Which products support data residency?

Jira, Jira Service Management, Jira Product Discovery, Confluence and Loom, on Standard, Premium and Enterprise plans. Rovo has no setting of its own and follows your Jira data, or Confluence if you do not run Jira. Atlassian Government Cloud does not offer the setting because all its data is resident in the US region.

Does data residency cover all of my data?

No. User account information such as names and email addresses is held globally by design. Data in transit and cached content are out of scope for up to 30 days, as are audit and operational logs, analytics, third party app integration data, automation configuration, connected DevOps data in Jira, customer accounts in Jira Service Management, and permission and restriction configuration in Confluence. Atlassian states that coverage extends only to in-scope data stored at rest beyond 30 days.

Is India the default location for an Indian organization?

No, and Atlassian says so explicitly. India is not assigned as the default data residency location even for organizations based in India, and it has to be set manually in Atlassian Administration. An organization that never set it has its data in the Global location.

What happens when you move data to another region?

Expect downtime of up to 24 hours depending on data size, and search unavailable during re-indexing for up to 3 days. Treat it as a planned migration with a maintenance window rather than a configuration change.

What are the options if there is no region for your country?

Three. Accept Global and build the compliance case on contractual and certification grounds. Use Isolated Cloud, where residency is architectural and extends to data in transit, but the instance is fixed to its region of creation. Or run self-managed Data Center, which is now on a timeline ending 28 March 2029.