

Most organizations run a single email platform with no tested fallback. When that platform fails, the gap between inconvenience and physical harm is narrower than most continuity plans acknowledge.
The November 13, 2026 Microsoft global outage made that point with unusual clarity. A hotel pool maintenance team could not receive dosing instructions because Outlook was down and no backup channel existed. Fifteen guests were poisoned by chlorine [1]. The email outage did not cause a data breach. It caused a chemical injury event. The cyber and physical domains are not as separate as security teams often treat them.
Microsoft 365 outages follow a consistent pattern. On January 25, 2023, a misconfigured WAN update knocked out at least 11 services simultaneously for over 5 hours, including Teams, Exchange Online, and Outlook [2]. Organizations that relied entirely on those channels to manage the response had no way to coordinate one. The November 2026 incident repeated the same failure at a different scale and with physical consequences.
NIST Special Publication 800-34 Rev. 1, the federal contingency planning standard, requires organizations to evaluate information systems and operations to determine contingency planning requirements and priorities [3]. A contingency plan that lists email as the communication channel for incident response does not satisfy that requirement. It assumes the system that failed will also coordinate the recovery. NIST's framework explicitly addresses the interrelationships between contingency planning and other emergency management plans, recognizing that communication is a dependency, not a given [3].
CISA's Emergency Services Sector Continuity Planning Suite covers 11 major continuity elements, one of which is specifically Continuity Communications, defined as systems that support full connectivity among leadership, internal elements, and other organizations to perform essential functions during an emergency [4]. The suite states: "Continuity planning is simply the good business practice of ensuring the execution of essential functions through emergencies" [4]. CISA developed this for first responders, but the operational logic applies to any organization where communication failure translates into physical or financial consequence. Hotels, water utilities, hospital networks, logistics operators, and manufacturing plants all qualify.
The backup communication architecture needs 3 layers, each addressing a different failure scenario:
All 3 layers fail if they exist only as entries in a SharePoint document that cannot be opened when SharePoint is down.
The authentication dependency compounds all of this. Azure Active Directory (AAD) is the identity layer for most Microsoft 365 environments. Azure AD Domain Services Replica Sets are designed to provide availability during an outage, but synchronization between replica sets stops when an outage occurs [5]. Authentication continues to function during that window, but no further syncs happen until the outage resolves [5]. Any backup system requiring a fresh identity verification against AAD during a concurrent Azure outage will fail at the authentication step. Backup communication access must be pre-staged with credentials that do not depend on a live AAD connection.
Outages also create a second-order risk. Business email compromise (BEC) attackers monitor organizational disruption, and the November 2026 Microsoft outage was no different. Security researchers documented a surge in themed phishing campaigns within hours of the outage beginning, with attackers exploiting staff confusion about which channel was authoritative [6]. Phishing and spoofing generated 191,561 complaints in the FBI's 2025 Internet Crime Report, the highest count of any reported crime type [7]. The average cost of a data breach reached $4.99 million in 2026 [7]. An outage that pushes employees to improvise communication channels, bypass normal verification, and accept urgent requests from unfamiliar addresses is exactly the environment BEC operators seek [8]. The attacker does not need to cause the outage. They wait for it.
The NCSC notes that implementing anti-spoofing measures including SPF, DKIM, and DMARC protects brand reputation and reduces costs associated with email fraud and service downtime [9]. Those controls matter on backup domains as much as on primary ones. A secondary email domain without DMARC enforcement is a spoofable attack surface that did not exist before the outage created a need for it.
The pool poisoning incident represents a pattern, not an outlier. Healthcare relies on email for medication reconciliation, test result routing, and care coordination. Water and wastewater systems transmit dosing schedules and compliance reports the same way. Industrial facilities route maintenance instructions and safety alerts through the same Microsoft 365 tenants that serve finance and HR. Each sector carries a version of the same exposure the November 2026 hotel incident made visible.
CISA's warning to emergency services applies equally here: "Without the planning, provision, and implementation of continuity principles, first responders may be unable to provide services to help fellow citizens when needed the most" [4]. Substitute "first responders" with "pool maintenance staff," "hospital pharmacy teams," or "grid operations centers," and the sentence holds.
The action list is short. First, identify every operational process that depends on email to deliver a time-sensitive instruction with physical consequence. Second, assign each one a documented backup channel that does not share infrastructure with the primary. Third, store that documentation offline or on a system outside the primary cloud environment. Fourth, pre-stage authentication credentials for backup systems so access does not require a live identity provider connection. Fifth, test the backup channel on a defined schedule, quarterly at minimum. A backup plan that has never been executed is a document, not a capability.
The November 2026 outage will not be the last. The technical cause of any given incident is Microsoft's to fix. The operational cause, running without a fallback, is yours.
Communication outages expose organizations to both operational failure and opportunistic cyberattacks. Anomali helps security teams stay ahead of both.
Does your current stack give you early warning when attackers pivot to exploit an outage your organization is already managing?
FEATURED RESOURCES


