Business Continuity

    The Week Everything Broke, Even the NSA (and Nobody Got Hacked)

    Hawkins ConsultingNovember 19, 20266 min read

    If the week of November 12 felt like the internet was held together with tape, you weren't imagining it. In six days, Microsoft 365 went down for users in 74 countries, a nationwide ground stop canceled about 12,000 flights, people couldn't move money, and the National Security Agency's classified network went offline.

    The twist: none of it was a cyberattack.

    Key takeaways

    • Five major outages hit between November 12 and 17, 2026. Each one was blamed on an upgrade, a bug, routine maintenance, or a technology issue, not hackers.
    • Scammers didn't need a hack. They started sending fake "outage" emails the very next day.
    • When Microsoft 365 went down, Copilot went with it. If your team runs on AI, that's a new single point of failure.
    • The NSA kept critical work running because it built in redundancy. Most small businesses have none.
    • A one-page outage plan covers most of what matters, and it takes an afternoon to write.

    The scorecard

    DayWhat brokeWhat caused it
    Thu, Nov 12Microsoft 365: Outlook, Teams, SharePoint, and Copilot, with users in 74 countries reporting problemsAn issue from an earlier network capacity upgrade
    Fri, Nov 13A nationwide FAA ground stop, with about 12,000 flight cancellationsA malfunctioning router at a facility being upgraded from copper to fiber
    Mon, Nov 16Chase online banking, with delays on Chase payments at other banksA software bug
    Mon, Nov 16United Airlines departuresA technology issue United said was not cybersecurity-related
    Tue, Nov 17The NSA's classified internal networkA "temporary degradation" during scheduled maintenance

    Five outages. Zero hackers. Here's what that means for a 20-person office.

    Lesson 1: Your biggest outage risk is an update, not a hacker

    Microsoft traced its outage to an issue from an earlier network capacity upgrade. The FAA's ground stop started with a router at a facility mid-upgrade. The NSA went down during scheduled maintenance.

    The scariest thing in your IT environment might not be a hacker in another country. It might be Tuesday night's update. That goes for your updates and your vendors' updates. The fix is unglamorous: test changes before rolling them out everywhere, schedule them when a failure hurts least, and know how to undo them.

    Lesson 2: Scammers don't need a hack, just a headline

    The day after the Microsoft outage, security researchers at Trellix caught a wave of fake Microsoft notices claiming that messages had been "held after the outage" and would be deleted unless you clicked to restore them. They even quoted the real Microsoft incident number, which is what made them believable.

    Security researchers also warned that scammers switch to phone calls when email is down, posing as IT support, and may keep doing it after services come back.

    The rule for your team: during any outage, nobody enters a password from a link and nobody gives access to a caller. Check the vendor's official status page, or call your IT provider at the number you already have. We covered how attackers impersonate trusted services in Your Email Security Might Have a Door Propped Open.

    Lesson 3: Your AI assistant goes down with everything else

    Copilot was part of the Microsoft 365 outage. That's worth sitting with. If your team now leans on AI to summarize email, draft proposals, or take meeting notes, an outage doesn't just stop email. It stops the way your people work.

    That doesn't mean you should avoid AI. It means you should know which work depends on it and make sure people can still do that work by hand for a day. If you don't know which AI tools your team relies on, start with Your Employees Hired an AI Assistant. Nobody Told IT.

    Lesson 4: A backup isn't a file. It's a way to keep working.

    Even the NSA's network went down. But the agency said its mission-critical capabilities remained fully operational, because its infrastructure is "designed with multiple layers of redundancy."

    Most small businesses have one internet line, one bank, one email system, and one person who knows the passwords. The banking problems on November 16 showed how far one failure spreads: Chase confirmed a software bug, and Zelle payments to and from Chase were delayed even at banks whose own systems were working fine. If your payroll ran through the wrong bank that day, it was your problem too.

    Lesson 5: What your customers hear matters

    When Chase's online banking slowed, customers didn't know whether it was an attack. Chase said publicly that a software bug was to blame and that it found no evidence of a cyberattack. Clear, early answers leave less room for the worst guess.

    When your systems go down, your customers will hear something. Make sure it's from you.

    Your one-page "the tech is down" plan

    Write this down before you need it. One page, printed, in a drawer.

    • A backup way for staff to talk: a phone tree or a group text, not something that depends on Teams or Outlook.
    • A second way to get paid and pay people: a backup bank account or payment method, and a plan for payroll if your primary bank stalls.
    • A manual fallback for AI work: what people do for a day if Copilot or other AI tools are down.
    • A customer message, ready to send: a short note for your website, phone greeting, and social accounts.
    • The "verify first" rule: no passwords from outage emails, no access for unexpected callers, ever.
    • Who decides: one person who can say "we're down, switch to plan B."

    Backups are one of the five cybersecurity essentials. If you haven't checked yours lately, the free Security Checkup takes about two minutes.

    Frequently asked questions

    Were the November 2026 outages caused by cyberattacks?

    No. Microsoft, the FAA, Chase, United, and the NSA all attributed their outages to technical problems such as upgrades, software bugs, or maintenance, not cyberattacks. Scammers did, however, use the outages as cover for phishing emails and fake support calls.

    What should employees do if they get an email about an outage?

    Don't click links or enter passwords from it. Check the vendor's official status page directly or call your IT provider at a number you already have. Outage-themed phishing often looks real and may even quote real incident numbers.

    What is a business continuity plan for a small business?

    It's a short, written plan for how your business keeps operating when key systems fail. It covers how staff communicate, how you get paid and pay people, how work continues without specific tools, what you tell customers, and who makes the call to switch to backup plans.

    Should small businesses stop relying on cloud services and AI tools?

    No. Cloud services and AI tools are usually more reliable than what a small business could run on its own. The goal is to know which work depends on them and to have a simple fallback so one outage doesn't stop the whole business.

    Get a Business Continuity & Security Review

    The outages of November 12 to 17 weren't anyone's fault at your company. The next one might not be either. What you control is how ready you are.

    Our Business Continuity & Security Review maps which systems your business can't work without, finds the single points of failure (including AI tools), checks how your team would spot outage scams, and leaves you with a finished one-page outage plan. No 80-page binder. Just the page you'll actually use.

    Book your Business Continuity & Security Review

    Sources