Pick one task your team has to finish even when a cloud service is unavailable. Write down what it needs, who owns it and how they’ll do it if they can’t sign in or open a file. Then have someone try the fallback on a managed device before you need it.
The recent AWS and Dropbox disruptions are useful reminders to test two different things: whether people can get into the tools they rely on and whether they can reach the information they need once they’re there.
What happened this week
November 15: AWS reported connectivity issues in Northern Virginia. The AWS Health Dashboard lists an incident affecting multiple services in its US East (N. Virginia) region. A November 19 Cloud Security Alliance analysis says AWS IAM Identity Center sign-in was among the services affected. That matters because a team may need to sign in to manage its systems even when parts of the underlying application still work. The AWS entry doesn’t establish the impact on any particular business.
November 20: Dropbox reported trouble across several products. Its incident report lists the website, desktop and mobile apps, API and other products among the affected components. Dropbox posted its first update at 2:28 p.m. Eastern and marked the incident resolved at 3:13 p.m. Eastern. That’s the time between the two updates, and individual experiences can differ.
These were separate service reports. Neither one establishes a common cause. Sources checked November 21, 2026, at 2:08 a.m. Eastern.
Run a 15-minute practice exercise
Start with something ordinary. Imagine a dispatcher needs the next two job addresses and must tell the field crew which stops are confirmed. The scheduling app normally holds the assignments, the shared folder holds the work orders and the same company sign-in opens both. A backup app that uses that sign-in may leave the dispatcher with the same problem.
- Name the decision and the owner. The dispatcher decides which already-confirmed jobs can proceed. Choose a backup person and give both the same short instructions.
- Find the minimum information. Keep only the current shift’s approved details in an encrypted location on a company-managed device. Mark when the information was last refreshed. Use sample records for the exercise.
- Make the practice realistic. Start from a fresh session and assume the primary apps or sign-in are unavailable. Can the backup dispatcher open the approved file and reach the crew using the designated phone channel? If opening the file still needs the same sign-in or an online permission check, revise the fallback.
- Set a stop rule. When a job change, a payment instruction or a customer record can’t be confirmed, hold that action and call the person responsible. Keep existing access controls in place.
- Agree how to catch up. Record changes made during the interruption with a time and owner. When the main system comes back, have one person reconcile the log before automated messages or updates resume.
For a household, the same exercise can be smaller. On a trusted device, check that an essential travel itinerary or emergency contact is available offline before you leave. Choose the one item you would actually miss and keep access limited to the people who need it.
Put the result where the team can find it
A useful note can fit in four lines: the task, where to find the approved fallback, when to stop and who reconciles afterward. Set a refresh date so people can spot stale information, and repeat the exercise whenever a sign-in provider, app or process owner changes.
If the exercise exposes a shared dependency, you have a concrete question for your IT team. Our guide to business continuity and recovery gives you a broader planning framework. If you want help checking which systems your team depends on, book a Business IT consultation.

