Topics Tools Books & Guides Talk to Securafy

Knowledge Hub / Business Continuity

Business Continuity

Your Business Continuity Plan Probably Ignores the Outage That's Most Likely to Hit You

Most continuity plans cover fires and ransomware but not cloud provider outages. Learn what the 2025 AWS, Azure, and Cloudflare outages teach SMBs.

Randy Hall By Randy Hall Updated Nov 2026 12 min read Share

What would your team actually do for the next six hours if Microsoft 365 sign-ins stopped working at 10 a.m. on a Wednesday? Not your IT provider, your team: the person taking orders, the person running payroll, the person answering the client who is waiting on a document. Most business continuity plans we review have a confident answer for a server failure or a ransomware attack and no answer at all for this question, even though it is the outage businesses have repeatedly lived through.

A business continuity plan for cloud outages assumes there is nothing for you to restore, because the failure sits inside a provider you do not control. It describes how the business keeps operating, communicating, and deciding while someone else fixes the problem, and it identifies in advance which of your tools share the same upstream dependency.

Why is a vendor outage the one your plan is missing?

Because most continuity plans were written around assets the business owns: the server room, the office, the backups. Today much of the work runs through SaaS applications and cloud platforms that sit on a small number of shared providers, so the most frequent disruption is often a provider failure that your own IT team cannot repair, only work around.

That is the belief worth challenging. Owners tend to rank continuity risks by how dramatic they are, so fire, flood, and ransomware get the planning time. But for a company whose email, files, phones, CRM, and accounting all live in the cloud, a multi-hour provider outage is a far more ordinary event, and ordinary events are the ones a plan should rehearse. A plan that only says "restore from backup" has nothing to execute when the system that is down belongs to Amazon, Microsoft, or Cloudflare.

We made a version of this argument after the CrowdStrike outage, where a faulty vendor update took down Windows machines worldwide. The pattern since then has been consistent: the trigger changes, but the business problem is the same. Work stops because a dependency you rarely think about stopped.

What the late-2025 cloud outages showed

Three incidents in October and November 2025 make the case better than any hypothetical, because each provider published its own account of what failed. None of them was a cyberattack. Each was an internal change or defect that cascaded through infrastructure thousands of unrelated businesses depend on.

Provider and dateWhat failedWhat customers lostContinuity lesson
AWS, October 19–20, 2025A DNS automation defect left the DynamoDB endpoint in the US-EAST-1 region with an empty recordErrors across dependent AWS services for roughly 14 hours, including console sign-in through IAM Identity CenterOne regional failure can disable SaaS tools you never knew ran on AWS
Microsoft Azure, October 29, 2025An incompatible configuration in Azure Front Door crashed edge serversTimeouts and DNS errors for services including Microsoft 365 and parts of Microsoft Entra ID for about eight hoursIdentity and productivity can fail together, from the same root cause
Cloudflare, November 18, 2025An oversized Bot Management configuration file caused proxy failuresHTTP errors for sites behind Cloudflare and widespread Cloudflare Access login failuresA security layer in front of your apps is also a dependency in your plan

The details are worth reading in the original sources. AWS's post-event summary of the US-EAST-1 disruption traces how a latent race condition in DynamoDB's DNS management spread to EC2 launches, Network Load Balancers, Lambda, Amazon Connect, and sign-in services between 11:48 PM PDT on October 19 and 2:20 PM PDT on October 20. Microsoft's post incident review for Azure Front Door lists Microsoft 365, Dynamics 365, Power Platform, and Entra ID components among the affected services, with impact from 15:41 UTC until mitigation was confirmed at 00:05 UTC the next day. Cloudflare's own write-up of the November 18 outage explains that failed Access authentication blocked users from reaching their applications until a fix rolled out, and that most dashboard users could not log in while Turnstile was down.

The shared lesson is about login. In all three events, the authentication or access layer was either directly affected or sat behind an affected service. If your continuity plan assumes people can sign in to something, it needs a section for when they cannot. We looked at the AWS event specifically in what the October 2025 AWS outage meant for Ohio businesses; the Azure and Cloudflare incidents show that the same exposure exists regardless of which provider you favor.

What is the difference between disaster recovery and business continuity during a cloud outage?

Disaster recovery restores your systems and data after a disruption. Business continuity keeps the business operating while the disruption is happening. During a provider outage you usually cannot perform disaster recovery at all, because the systems are not yours, so continuity is the entire response: workarounds, communication, and decisions about what waits.

That difference matters for how you spend planning effort. A strong disaster recovery program, with tested backups and documented restore order, is essential for incidents inside your environment. But it does not answer what the front desk does when the scheduling platform is unreachable, or how finance approves a payment when email and Teams are down at the same time. Those answers are owned by the business, not the IT team, which is why we think this is a leadership planning exercise first and a technical one second.

Regulated businesses already have a hook for this. The HIPAA Security Rule contingency plan standard requires, separately from backup and disaster recovery, an emergency mode operation plan that lets critical business processes continue while operating in emergency mode. That is a continuity requirement, and a provider outage is exactly the kind of event it should cover.

How do you build a business continuity plan for cloud outages?

Start by listing every cloud and SaaS service the business touches daily, rank each by how long you can work without it, and record which upstream provider it runs on. Then write a short, specific workaround for the top tier and test it. The goal is not to prevent provider outages; it is to make the next one an inconvenience instead of a lost day.

NIST's Cybersecurity Framework 2.0 frames this well in its supply chain outcomes: suppliers should be "known and prioritized by criticality," and relevant suppliers and third parties should be "included in incident planning, response, and recovery activities." In practical terms for a 20- to 300-person company, that translates into a sequence we walk clients through:

  1. Inventory the services people actually use. Include identity (Microsoft Entra ID or Google), email, file storage, phones, payments, CRM, line-of-business SaaS, and anything sitting in front of them such as a DNS or security proxy.
  2. Set a tolerable downtime for each. Ask the department head, not IT: how many hours until this becomes a customer, revenue, or compliance problem? That number becomes the planning target.
  3. Trace the upstream provider. Ask vendors which cloud and region they run on, and note where several of your "diverse" tools share the same provider. Concentration you did not choose is still concentration.
  4. Write the manual workaround. For each top-tier service: who does what, using which alternate tool or paper process, and how work entered during the outage gets reconciled afterward.
  5. Plan communication that does not depend on the outage. Keep a staff contact list and a client notice template somewhere reachable if Microsoft 365 or Google Workspace is down, plus a named person who decides when to send it.
  6. Rehearse once a year. Run a one-hour tabletop where a specific provider fails at a specific time, and fix whatever the team could not answer.

The third step is the one most plans skip, and the one these outages exposed. Our guide to protecting your business from vendor vulnerabilities covers how to ask vendors the right questions, and the same discipline is spreading into AI tools; our AI Lab piece on fourth-party risk in your vendor's vendor shows how quickly hidden dependencies stack up. If your business now relies on AI assistants or agents for real work, apply the same thinking to what happens when your AI vendor goes down.

What should you ask your IT provider?

Ask whether your provider has mapped your dependencies, whether it monitors provider status for the services you use, and what it will do in the first 30 minutes of a provider outage. A good answer includes who confirms the outage, who notifies your staff, and which workarounds it can switch on. Contracts matter here as well; several of the MSP SLA traps that still cause downtime for SMBs come down to service levels that quietly exclude third-party outages.

What this costs a business that skips it

The impact of a provider outage is rarely the outage itself; it is the confusion around it. Staff wait instead of switching to a workaround. Clients hear nothing and assume the worst. Managers make ad hoc decisions about what to delay, and work entered on paper never gets reconciled. Eight hours of provider downtime can easily become a day and a half of lost productivity and cleanup when nobody knows the plan.

A plan that names the dependency, the tolerable downtime, the workaround, and the person who communicates does not make the provider more reliable. It shortens the business impact, protects client trust, and gives leadership a clear basis for deciding where real redundancy, such as a second internet path or an alternate phone route, is worth paying for. If you are refreshing the whole document, our walkthrough of building a business continuity plan that works covers the foundations this section plugs into.

How Securafy helps with cloud outage continuity

For clients, we start with a dependency map built from what your people actually use, not from a vendor list, and we trace each critical service to the provider and region it depends on. Department heads set the tolerable downtime, and we document workarounds and communication steps that do not rely on the services most likely to fail together, especially identity and email.

When a major provider has an incident, our team confirms impact against that map, tells you which of your tools are affected, and helps your staff switch to the documented workarounds. Afterward, we review what happened and update the plan, so each outage makes the next one shorter.

Where To Go From Here

If your continuity plan has a chapter for a flood but not for a morning without Microsoft 365, start with the dependency inventory above. It usually takes a few hours and immediately shows where a single provider failure would stop your business.

To see how we build and maintain continuity plans that account for third-party outages, review Securafy's business continuity services.

If you would rather talk through your specific environment first, book a strategy call with Securafy and we will walk through your critical cloud dependencies and the workarounds your team needs before the next outage.

Tagged Under Business Continuity

Join The Conversation

Have a question or perspective on this topic? Add it below.

Randy Hall

About The Author

Randy Hall · CEO & Founder

Randy Hall is the CEO and Founder of Securafy, with decades of experience helping organizations make smarter, safer decisions about technology.

A frequent speaker and instructor at national IT events, Randy has advised thousands of organizations, from startups and SMBs to large enterprises and U.S. government entities, on secure, practical technology adoption. He writes about the decisions business leaders are often expected to make without enough context, including cybersecurity, compliance, AI, cyber insurance, IT strategy, and business resilience.

Outside the office, you’ll often find Randy on Lake Erie enjoying time on his 38-foot Chris-Craft.

Writes about: Cybersecurity strategy, compliance, AI security, business resilience, cyber insurance, SMB risk, IT leadership

More From This Author →

Get Practical Cybersecurity Field Notes

Monthly cybersecurity, compliance, and IT strategy updates from Securafy, written for business owners who need clear next steps.

  • Practical security tips from our Cyber Security Drip series
  • The Securafy Times, our monthly roundup on compliance and IT strategy
  • Occasional updates on new tools, guides, and research
  • No spam — unsubscribe anytime