Windows 365 Reserve and Disaster Recovery: A Business Continuity Guide

• 26 October 2026
• Windows 365
• Written by Editorial Team

Windows 365 protects a Cloud PC in four different ways: automatic in-zone recovery for compute failures, restore points that roll a single Cloud PC back, cross-region disaster recovery for regional outages, and Windows 365 Reserve, a separate license that gives a user up to 10 days of temporary Cloud PC access when their physical device is lost, stolen, or delayed. Each one solves a different problem, and none of them replaces backing up your data. Knowing which layer answers which scenario is the heart of a sensible continuity plan.

This guide is written for IT managers, business owners, and anyone responsible for keeping people productive when something breaks. It explains what each Windows 365 protection does, what it costs you in setup effort and limits, how recovery times compare, and how to combine them into a plan you can actually test. Everything here draws on Microsoft’s own documentation, which we link throughout, because the details of these features are where plans succeed or fail.

Key Takeaways

What Business Continuity Means for a Cloud PC

When a physical PC fails, the problem is easy to picture: a broken screen, a stolen bag, a laptop stuck in shipping. A Cloud PC adds a second set of failure types, because the desktop itself lives in Microsoft’s cloud (if the idea is new to you, here is how a Cloud PC works). In practice, the things that go wrong fall into four groups.

• The endpoint fails. The device a person uses to reach the Cloud PC is lost, damaged, or unavailable. The Cloud PC is fine, but the person has nothing to connect with, or needs a second Cloud PC while waiting.
• The Cloud PC itself goes wrong. A bad update, a faulty software change, or a user mistake leaves one desktop in a poor state.
• The compute behind it fails. A host, power, or network problem affects the infrastructure that runs the Cloud PC inside a datacenter zone.
• A whole region is unavailable. A large Azure regional event cuts access to every Cloud PC hosted there.

Windows 365 has a different answer for each of these, and the rest of this guide takes them in turn. Microsoft is also clear about where its responsibility ends: it provides infrastructure-level availability and resiliency, while your organization remains responsible for data protection, compliance, and operational recovery processes that suit your risk profile.

The Four Layers of Windows 365 Resilience

Think of the layers as a ladder. The lowest rung is automatic and needs nothing from you. Each rung above it handles a bigger or more specific problem, and costs more effort to set up.

Layer 1: Built-In Resilience and In-Zone Recovery

Every Cloud PC benefits from resilience you never configure. Microsoft’s business continuity and disaster recovery guidance for Windows 365 describes automated in-zone recovery for compute failures such as host failures, power failures, and networking issues. The service moves the workload to healthy compute resources in the same zone. Users may see a brief interruption and can reconnect once recovery completes. Recovery uses an up-to-date copy of the Cloud PC’s operating system disk, with a recovery point objective Microsoft describes as approximately zero.

The same page notes that Windows 365 is covered by a 99.9% availability SLA, and that Cloud PC OS disks sit on Azure storage that Microsoft rates at 11 nines of durability against data loss over a year. It also points to a quieter source of resilience: integration with Microsoft 365. OneDrive keeps user files independent of the Cloud PC, so if a Cloud PC must be reprovisioned, the files remain available. That last point will matter again when we get to backups.

Layer 2: Restore Points for a Single Cloud PC

Restore points answer a smaller question: what if one Cloud PC is in a bad state and you want it back the way it was? Windows 365 point-in-time restore lets an administrator, and optionally the user, return a Cloud PC to an earlier saved state, much like System Restore on a physical PC.

• Short-term restore points. For Windows 365 Enterprise, an administrator can set these to save every 4, 6, 12, 16, or 24 hours, with 12 hours as the default. Each Cloud PC keeps up to 10 of them, so a shorter interval means a shorter history.
• Long-term restore points. Enterprise also keeps four long-term restore points, saved every seven days. You cannot configure these.
• On-demand restore points. An administrator can create a manual restore point before a risky change.
• Windows 365 Business. Business saves short-term restore points every 12 hours, with 10 saved at that interval.

Microsoft’s restore overview recommends choosing the restore point closest to the current time, because restoring returns the Cloud PC to that moment and anything saved only on the Cloud PC since then is lost. There is also a safety net for one specific case: when an Enterprise Cloud PC is deprovisioned because its license expired, the service saves a retention snapshot that is available for 28 days. That snapshot does not apply when you remove a license from a user, a distinction worth knowing before you clean up licenses.

Layer 3: Cross-Region Disaster Recovery and Disaster Recovery Plus

Regional outages are rare, but they are the scenario that stops a whole company at once. Windows 365 offers two optional add-ons for them, and they trade cost against speed.

• Cross-Region Disaster Recovery

  This add-on periodically saves restore points of a Cloud PC and synchronizes them to a backup region you designate. If the primary region goes down, an administrator activates recovery and affected users receive a temporary Cloud PC in the alternate region. Microsoft’s target recovery time is about four hours, and the recovery point depends on how often restore points are saved. It is available as an add-on license for Windows 365 Enterprise and for Windows 365 Flex in dedicated mode.

• Windows 365 Disaster Recovery Plus

  Disaster Recovery Plus is built for people who cannot wait hours. It creates three copies of the Cloud PC OS disk in a different geography and proactively reserves capacity there, which brings Microsoft’s recovery time target down to about 30 minutes. It is an optional service for Windows 365 Enterprise. Microsoft’s own example gives it to a small group of executives and gives standard cross-region recovery to a larger call center, which is a sensible pattern for most organizations.

Several details shape how these behave in a real incident, and they are easy to miss.

• Activation is manual. An administrator activates and deactivates cross-region recovery, and bulk actions let you move many Cloud PCs at once. With Disaster Recovery Plus, Microsoft’s example also lets VIP users start the failover themselves from the Windows App.
• The recovery Cloud PC is temporary. When you deactivate recovery after the outage, the temporary Cloud PC is deleted and nothing from it is kept. Microsoft’s FAQ adds that temporary Cloud PCs must be deactivated within seven days after the declared outage ends, so save work to OneDrive or SharePoint.
• Capacity is not guaranteed for the standard add-on. If the backup region lacks capacity or is unhealthy during an event, the backup Cloud PC is not provisioned. Disaster Recovery Plus exists to reduce exactly that risk by reserving capacity ahead of time.
• The network matters. Microsoft’s scenario notes that using the Microsoft-hosted network simplified failover.
• You get warning, if you ask for it. Microsoft recommends enabling Service Health notifications for Windows 365 so you hear about an outage in time to start a failover.

Layer 4: Windows 365 Reserve

Reserve is the odd one out, and the easiest to misunderstand. The other layers protect the Cloud PC. Reserve protects the person. Windows 365 Reserve is a standalone license designed for people who mainly work on a physical PC. When that device is lost, stolen, broken, or stuck in shipping, or when an outage or a temporary staffing need arises, the organization can give that person a temporary Cloud PC with the corporate apps, settings, and policies already in place. It has been generally available since late 2025.

How Reserve Works

An administrator assigns each covered user a Reserve license and sets up a provisioning policy in Microsoft Intune ahead of time. When someone needs access, the administrator provisions their Reserve Cloud PC, and the user connects from any device, managed or personal, through the Windows App or a browser. Microsoft has also previewed letting users provision their own Reserve Cloud PC from the Windows App. If your people use tablets or phones as their fallback, our guide to running a Cloud PC on an iPad shows what that experience looks like.

The Numbers That Matter

• 10 days per user per year. Each license grants up to 10 days of access for one user. The days cannot be shared or extended, and stacking is not supported at general availability.
• The clock starts at provisioning. The 10-day period begins when the Cloud PC is provisioned and ends at deprovisioning, so returning the Cloud PC as soon as the person has a device again saves days.
• Seven days to activate. A user’s Reserve Cloud PC becomes eligible for provisioning seven days after the license is assigned. That delay applies the first time, or after a lapse in coverage, so you cannot buy a license in the middle of a crisis and use it that day.
• One at a time. A user can have only one Reserve license and one Reserve Cloud PC at a time.

What Reserve Does Not Do

Microsoft’s Reserve FAQ is refreshingly direct about limits, and they matter for planning.

• No guaranteed capacity. Reserve does not preallocate or guarantee Cloud PC capacity. If your chosen geography is short of capacity or has service health issues when you request a Cloud PC, provisioning may have to wait. During large outages, availability can also be affected, and Microsoft suggests provisioning your most critical users first.
• Not a copy of the user’s PC. A Reserve Cloud PC comes with corporate apps and policies from Intune, but local data from the user’s own PC is not restored to it.
• A fixed size. There is no choice of Cloud PC size. Microsoft documents a default of 4 vCPU, 16 GB of RAM, and 128 GB of storage.
• Standard images and networking. Reserve uses the latest three Windows 11 gallery images, with or without Microsoft 365 apps, on the Microsoft-hosted network. Custom images, Azure network connections, and GPU Cloud PCs are not supported, and reaching line-of-business apps requires a VPN or a similar solution.
• No safety net on return. Deprovisioning deletes the Cloud PC and all data that was not backed up elsewhere, and there is no grace period or snapshot when an administrator or user does it. When the 10 days run out, the service takes a retention snapshot and deprovisions the Cloud PC.
• It needs the internet. Like every Cloud PC, it depends on network connectivity.

Reserve also has the same licensing prerequisites as Windows 365 Enterprise and Flex: Windows Enterprise, Microsoft Intune, and Microsoft Entra ID P1 for each user. Our checklist of client and license requirements covers those, and it is worth checking them before you commit. Reserve works with Windows 365 Link and Windows 365 Boot, so a person can even start a local device straight into the temporary Cloud PC. For shared desks and spare devices, a purpose-built Cloud PC device pairs naturally with it.

Reserve vs a Loaner Laptop

The traditional answer to a lost or broken laptop is a loaner, and loaners have real strengths: they work offline and give a person a full device. They also need preparation, imaging, and shipping, they carry security risks while in transit, and you only have as many as you stocked. A Reserve Cloud PC trades offline access and a physical device for speed and consistency. Microsoft’s own continuity example has a call center that could not enter its flooded office provisioning Reserve Cloud PCs for its staff, while the administrators also sourced laptops from other offices for people without a usable device at home. The two approaches work best together, not as rivals.

Reserve vs Restore Points vs Cross-Region Recovery: Which Do You Need?

The quickest way to choose is to start from the problem. This table matches common scenarios to the right tool, and the image after it shows the same logic at a glance.

Scenario	Best Fit	Recovery Target	Watch Out For
A compute host or zone component fails	In-zone recovery	Automatic, brief interruption	Nothing to configure
A laptop is lost, stolen, damaged, or delayed	Windows 365 Reserve	Provision on demand after activation	Seven-day activation, no capacity guarantee, local data not restored
One Cloud PC breaks after a change	Restore point	As fast as a restore operation	Work saved only on the Cloud PC since the restore point is lost
An Azure region is unavailable	Cross-region disaster recovery	About four hours	Manual activation, backup region capacity, temporary Cloud PC
Key staff cannot wait hours	Disaster Recovery Plus	About 30 minutes	Add-on cost, Enterprise only

Two points deserve emphasis. First, these tools are complementary, so most organizations will use more than one. Second, they depend on your edition. Restore points exist in both Business and Enterprise, cross-region recovery is an add-on for Enterprise and for Flex in dedicated mode, and Disaster Recovery Plus is for Enterprise. If you are not sure which edition you have or need, our current plan pricing by edition shows the options side by side.

Windows 365 Is Not a Backup

This is the point that most disaster recovery plans get wrong. Restore points, cross-region copies, and Reserve Cloud PCs all protect the desktop. They do not protect the contents of your business. A restore point rolls a Cloud PC back, which can erase recent unsaved work. A temporary recovery Cloud PC is thrown away when the outage ends. A Reserve Cloud PC deletes its data when it is returned. In every case, the safe place for important work is somewhere other than the Cloud PC.

• Files. Keep documents in OneDrive, SharePoint, or your document management system, with known-folder redirection so Desktop and Documents sync automatically.
• Email and calendars. These live in Exchange Online, not on the Cloud PC, and need their own retention approach.
• Application data. Anything a line-of-business app stores locally needs a plan, ideally by moving it to a server or cloud service you back up.
• Identity. Recovery depends on being able to sign in, so protect administrator accounts and keep emergency access in place.

Security incidents are a continuity scenario too, and a clean Cloud PC is only useful if the controls around it are sound. Our steps for hardening Cloud PCs cover the policies that reduce the chance of needing recovery at all. Firms with preservation duties have an extra concern, since deleting a Cloud PC can delete evidence, and we cover how law firms should treat Cloud PCs as workspaces, not records in a separate guide.

Who Needs What: Matching Protection to Your Organization

You do not need every layer for every person. These patterns work for most organizations.

• Small Businesses on Windows 365 Business

  Business includes restore points every 12 hours, and the built-in in-zone recovery applies to every Cloud PC. The most valuable additions are discipline about where files live and a documented way to get a person working again if a device fails. Cross-region add-ons are an Enterprise and Flex feature, and Reserve needs Enterprise-level licensing prerequisites, so confirm what you hold before you plan around them.

• Call Centers and Shift-Based Teams

  Teams that cannot stop answering phones are the classic case for cross-region recovery, and Flex in dedicated mode supports it. Add Reserve for the scenario Microsoft’s own example describes: a building becomes unusable, and staff need desktops they can reach from home.

• Executives and Critical Roles

  For a small group whose downtime is expensive, Disaster Recovery Plus is the option that brings recovery down to about 30 minutes, and letting those users start failover themselves removes a dependency on the helpdesk.

• Hybrid and Remote Workforces

  People who mainly use physical PCs benefit most from Reserve, because it covers the most likely failure, a lost or broken laptop, at a modest annual cost compared with keeping spare devices. Combine it with good file hygiene and restore points.

• Teams Without Resilience Expertise

  If nobody in-house wants to own recovery planning, a partner can run it for you. Some organizations pair Windows 365 with fully managed cloud desktops so that one provider handles design, monitoring, and support end to end.

How Windows 365 Compares With Azure Virtual Desktop for Recovery

The Windows 365 approach is productized: you switch on defined features, and Microsoft publishes recovery targets. Azure Virtual Desktop usually leaves resilience design to you, for example by placing host pools in a second region and replicating profile storage, which gives more flexibility and more work. That difference follows the broader split between the two services. For the full picture, here is how the two Microsoft services compare on cost and control, and if you want AVD’s flexibility without carrying the operational load, managed personal desktops on Azure put a provider in charge of it.

What Windows 365 Continuity Costs

Costs differ by layer. Built-in resilience and restore points come with the service. Cross-region recovery and Disaster Recovery Plus are priced as per-user add-licenses, so the cost grows with the number of people you cover, which is the reason to tier your users instead of covering everyone. Reserve is an annual license, and Microsoft’s own continuity example describes it as costing less than $20 per user per year, though you should confirm current pricing with Microsoft or your cloud solution provider.

The better comparison is not the license price on its own but the cost of the alternative: a stock of spare laptops that need imaging and replacing, or hours of lost output when a region fails. Sizing matters too, because a recovery plan that doubles your Cloud PC spend is hard to defend. Our guide to ways to trim a Cloud PC bill shows how to find and remove waste so that resilience fits inside the budget.

How to Build a Windows 365 Continuity Plan in 7 Steps

A plan is only useful if it is specific, owned, and tested. These seven steps take you from a blank page to something you can rehearse.

• Step 1: Tier Your Users

  Sort people into groups by how costly their downtime is: critical staff, important staff, and everyone else. Be honest, because most organizations discover that fewer people are truly critical than they first assumed.

• Step 2: Pick Protections for Each Tier

  Match tools to tiers. Critical staff might receive Disaster Recovery Plus and Reserve, important staff cross-region recovery, and everyone else restore points plus good file hygiene.

• Step 3: Set Recovery Targets

  Decide how long each tier can be offline and how much recent work it can afford to lose. Compare those answers with the published targets, about four hours for cross-region recovery and about 30 minutes for Disaster Recovery Plus, and with how often your restore points are saved. If a target does not fit, change the tier or the tool.

• Step 4: Configure Everything Early

  Set up restore point frequency, cross-region settings, and Reserve provisioning policies before an incident, and assign Reserve licenses at least seven days ahead. For larger groups, automating provisioning with Power Platform removes manual steps when time is short.

• Step 5: Protect the Data Layer

  Confirm that files sync to OneDrive or SharePoint, that application data lives somewhere you back up, and that administrator access has an emergency route. The best recovery tools cannot restore what was never saved.

• Step 6: Write the Runbook

  Name who watches Service Health alerts, who declares an outage, who runs the bulk actions, how users are told, and who decides when to fail back. Put the Reserve steps in it too, including who provisions first when capacity is tight.

• Step 7: Test and Rehearse

  Run a failover and failback with a small pilot group, and practice provisioning a Reserve Cloud PC for a volunteer who connects from a personal device. Time each step, fix what was slow, and repeat on a schedule. A plan that has been tried is worth far more than one that has been written.

Common Mistakes in Windows 365 Recovery Planning

• Treating restore points as backups. They protect the desktop’s state, not the only copy of your files.
• Buying Reserve in the middle of a crisis. A new license needs seven days to activate, so it has to be in place before you need it.
• Assuming guaranteed capacity. Reserve does not preallocate capacity, and the standard cross-region add-on depends on the backup region having room. Disaster Recovery Plus reserves capacity, which is why it is the option for critical users.
• Forgetting to deactivate recovery. Temporary Cloud PCs must be deactivated within seven days after the outage ends, and nothing created on them is kept.
• Spending the 10 days carelessly. Leaving a Reserve Cloud PC provisioned after a laptop arrives burns days you may need later.
• Never rehearsing. Bulk provisioning has documented limits, such as 100 devices a minute for Reserve, and you want to meet them in a drill rather than a crisis.
• Covering everyone equally. Per-user add-ons add up, so tier your people and spend where downtime hurts most.

The Bottom Line

Windows 365 gives you a layered set of tools for the different ways a desktop can fail, from automatic in-zone recovery to restore points, regional recovery add-ons, and Reserve for the lost-laptop problem. None of them is a backup, and none works well when it is set up in the middle of an incident. The organizations that recover quickly decide in advance who needs what, keep their files somewhere safe, assign Reserve licenses early, enable Service Health alerts, and practice. Start with your critical users, choose the lightest tool that meets their recovery target, and test it on a quiet day.

Frequently Asked Questions

1. What is Windows 365 Reserve?

   Windows 365 Reserve is a standalone Windows 365 license that gives a user up to 10 days of temporary Cloud PC access per year when their physical device is lost, stolen, broken, or delayed. The Cloud PC arrives with corporate apps, settings, and policies from Microsoft Intune.

2. How many days does Windows 365 Reserve provide?

   Each license provides up to 10 days of Cloud PC access for one user per year. The days cannot be shared or extended, and the clock starts when the Cloud PC is provisioned and stops when it is deprovisioned.

3. How soon can I use Windows 365 Reserve after buying it?

   A user’s Reserve Cloud PC becomes eligible for provisioning seven days after the license is assigned. This delay applies the first time, or after a lapse in coverage, so licenses should be assigned before an incident.

4. Does Windows 365 Reserve guarantee capacity?

   No. Reserve does not preallocate or guarantee Cloud PC capacity, so provisioning can be delayed if the chosen geography has limited availability or service health issues. Microsoft recommends provisioning your most critical users first during large events.

5. Does Windows 365 back up my files?

   No. Windows 365 provides restore points and recovery options for the Cloud PC, but they are not a backup of your business data. Keep important files in OneDrive, SharePoint, or your document management system.

6. What are Windows 365 restore points?

   Restore points let you return a Cloud PC to an earlier state. Enterprise saves short-term restore points every 4, 6, 12, 16, or 24 hours with up to 10 kept, plus four long-term restore points saved weekly, while Business saves short-term restore points every 12 hours.

7. What is Windows 365 cross-region disaster recovery?

   Cross-region disaster recovery is an optional add-on that saves restore points to a backup region and, during a regional outage, gives affected users a temporary Cloud PC there. Microsoft’s target recovery time is about four hours, and an administrator activates and deactivates it manually.

8. What is the difference between cross-region disaster recovery and Disaster Recovery Plus?

   Disaster Recovery Plus creates three copies of the Cloud PC OS disk in another geography and reserves capacity there, with a recovery time target of about 30 minutes. Standard cross-region recovery targets about four hours and depends on capacity in the backup region.

9. Which Windows 365 editions support disaster recovery add-ons?

   Cross-region disaster recovery is an add-on for Windows 365 Enterprise and for Windows 365 Flex in dedicated mode, and Disaster Recovery Plus is an optional service for Windows 365 Enterprise. Windows 365 Reserve does not support the disaster recovery add-ons.

10. What happens to work done on a temporary recovery Cloud PC?

    Nothing from a temporary recovery Cloud PC is preserved when recovery is deactivated, so users should save work to OneDrive or SharePoint. Temporary Cloud PCs must be deactivated within seven days after the declared outage ends.

11. Is there an SLA for Windows 365?

    Microsoft states that Windows 365 is covered by a 99.9% availability SLA. It also describes automated in-zone recovery for compute failures and Cloud PC OS disks stored with 11 nines of durability.

12. Does Windows 365 Reserve work with Windows 365 Link and Windows 365 Boot?

    Yes. Microsoft lists both Windows 365 Link and Windows 365 Boot as supported with Reserve, so a local device set up with Boot can start straight into a Reserve Cloud PC. Reserve does not support custom images, GPU Cloud PCs, or Azure network connections.