What went down?
This table follows a timeline compiled from each company’s own statements. Causes are what the company or an official said. Several are preliminary.
| Date | What | How long | Stated cause |
|---|---|---|---|
| Nov 12 | Microsoft 365: Outlook email, parts of Teams | About a day for email | A bug in a server update plus reduced capacity (preliminary) |
| Nov 13 | Workday | About 20 hours | A backup site connected to live systems during a test |
| Nov 13 | FAA nationwide ground stop | Departures halted over 4 hours | A contractor’s router during a network upgrade (under investigation) |
| Nov 15 | Amazon Web Services, eastern US region | About 3 hours | Not given |
| Nov 16 | Chase, with payment delays at other banks | About 6 hours | A software bug in its own hardware systems |
| Nov 16 | United Airlines departures | About 90 minutes | A “technology issue” |
The FAA stop alone led to more than 2,000 cancellations that Friday, according to the same timeline.
Was it a cyberattack?
Nothing published says so. Microsoft, the FAA, Chase and United each said they saw no sign of one, and Fox News reported on November 17 that companies and officials say cyberattacks were not behind the failures. The FBI is helping the Transportation Department look into the ground stop, which is routine after something that large.
Every cause that has been stated is a mistake: a bad update, a test that touched live systems, a piece of network equipment. That is worth knowing, because it changes what you should do. This was not a week for new security software. It was a week that showed how many businesses lean on the same few providers.
Why does this matter to a business with five employees?
Because your business runs on the same providers, a few steps removed. Your email is likely Microsoft or Google. Your payments go through a bank and a card processor. Your booking tool, your website host and your phone app each sit on a cloud provider. When one of them has a bad day, a piece of your business stops, and it is usually not obvious which piece or why.
Think about a Monday where your email is down for a day. Quote requests that arrive by email wait. Customers who reply to your estimate hear nothing. If email is the only way a lead reaches you, you are closed without knowing it.
What can I actually do about it?
You cannot stop a provider’s outage. These six things cost almost nothing and cover most of the damage.
- Write down what you run on. Ten minutes, one page: email, phone, website host, domain, booking, payments, accounting. Next to each, write what you would do for a day without it.
- Keep two ways for a customer to reach you that do not share a weak point. A form that emails you and a phone number that rings a real phone are separate paths. A form and a chat box that both end up in the same inbox are one path.
- Own your domain and your logins. If your web person holds your domain and something breaks, you are waiting on them. We covered what you should own in questions to ask before you buy a website.
- Keep your own copy of what matters. Export your customer list and open jobs once a month. If your scheduling tool is down, you can still make calls.
- Have fewer moving parts. Every tool you add is one more thing that can have a bad day. A simple site with a form and a phone number has less to break than one with eight plugins and three widgets.
- Check the status page before you change anything. Most providers have one. If it is their outage, the worst thing you can do is start resetting passwords and reinstalling things.
Are Atlas sites immune to outages?
No, and we will not pretend otherwise. Here is how ours hold up, plainly:
- The website itself is a set of finished files served from a hosting network with copies in many locations. There is no database or application behind each page that can fall over, which removes the most common way a small business site goes down. If the host has a major outage, the site is still affected.
- The form sends each request to a separate service. If that service is down, the form is too. That is why we recommend a phone number on the site as well.
- The phone receptionist runs on voice and phone providers, and it only picks up the calls you miss. Your own phone still rings first, so if a provider is down you lose the backup, not your phone line.
So the claim is modest: fewer parts, fewer ways to fail, and a second path when one does. That is the same thing we would tell you to do with any website, ours or not.
Is my website less safe during weeks like this?
Not because of these outages. The real website security story this month is a different one: a flaw in WordPress’s updater that attackers are using right now. That one does call for action if your site runs WordPress, and we wrote it up in WordPress sites got two urgent security fixes in a month.
Sources
- AppleScoop, Every major outage of November 2026: timeline, times and causes. Source for the dates, durations and stated causes in the table.
- Fox News, Tech outages live updates: Microsoft, banks and FAA (November 17, 2026). Source for services returning to normal, the FBI assisting, and officials saying cyberattacks were not behind the failures.


