Part of our guide to Best AI workspace for business in 2026: the landscape compared
On this page · 7 sections
- What stopped working on 13 November?
- Choose which part of Microsoft 365 to replace
- Run a continuity drill with two people and one customer scenario
- Price the move using one stubborn file and one difficult user account
- Keep a decision record that survives the next interruption
- Questions for the purchasing meeting
- Frequently asked questions
On Friday 13 November 2026, an Outlook disruption affected Australian users. Recovery was subsequently reported. Our account of the Australian Microsoft 365 and Outlook outage brings together the incident reporting and practical recovery steps. For a current service issue, start with Microsoft service status. An account of last Friday cannot tell you whether this morning's missing message is part of a new incident.
What stopped working on 13 November?
Imagine a service firm with a dispatcher, two field teams and a bookings mailbox. The dispatcher can still take calls. Both crews can finish their current jobs. The work that stops is accepting new bookings and telling a crew about an address change. Replacing the spreadsheet application would do little for either problem. The first purchase to consider might be a separately reachable bookings channel and a way to reconcile the work when mail returns.
Write down the last useful thing each person could do during the interruption. “Office was down” is hard to design around. “Could read yesterday's job sheet but could not receive a customer's key collection instructions” points to a concrete test. Include whether a phone had a cached copy, whether a person could authenticate again and whether someone on the other side actually received the message.
Ask the customer-facing team a different question from the technology team. Which customer promise could you no longer keep? An hour without email might be tolerable for a design review and unacceptable during a same-day dispatch window. The decision belongs to the person accountable for the promise. Let that person choose the first route to restore and the evidence they need before calling the route usable.
Choose which part of Microsoft 365 to replace
A second mailbox can help people exchange urgent messages. It does not move Office documents, change who owns a Teams meeting or recreate a SharePoint library. Conversely, buying a new suite does not tell customers where to reach the business during transition. Make a separate row for each dependency in the buying decision.
| Choice | When to shortlist it | What to demonstrate |
|---|---|---|
| Keep Microsoft 365 and add independent contact | Office files and collaboration work well; inbound enquiries matter during interruptions | A customer reaches the fallback and receives a reply while the main mailbox is unavailable |
| Pilot Google Workspace | The team can work in a browser and clients collaborate on shared documents | A real template survives a round trip, comments can be resolved and a guest loses access when removed |
| Pilot Zoho Workplace | Mail and collaborative office tools belong together for this team | Shared mail handling, calendar invitations and file permissions work for the roles in the pilot |
| Keep the suite and evaluate Cortex | The recurring cost is reading, routing and chasing work across tools | A specialist prepares a handover and a named person can approve the next action |
The Microsoft 365 business catalogue, Google Workspace product list and Zoho Workplace overview are starting points for a shortlist. Plan names and entitlements vary. Record the exact edition and the date of the quote alongside the pilot. A demonstration under the vendor's highest plan tells you very little about what staff will receive under the plan your business can afford.
Cortex belongs in the comparison when the problem includes work between tools. For example, a person can direct an assistant to assemble the day's outstanding callbacks from an accessible source. That still requires a working source, an available assistant and a human who can act. Treat every one of those as a dependency. Never count an assistant that reads the affected mailbox as an independent copy of the mailbox.
Run a continuity drill with two people and one customer scenario
Use a made-up booking. Ask one colleague to act as the customer and a second to cover the desk on a device they would actually have during an interruption. Give the customer the same public contact details a stranger could find. Have the desk work without its primary inbox. There is no reason to disconnect production or change mail routing to run this rehearsal.
Start the clock when the customer discovers the ordinary route is unavailable. Can they find the second route without receiving an email from you? A fallback address documented only in an Outlook folder is inaccessible to the person who needs it. A phone number on an independently reachable contact page might be sufficient. If evaluating Fulcrum Mail as a second mailbox, test how the operator signs in and retrieves its recovery information in the same circumstances.
Ask the customer to change the booking once. This is the part a simple send-and-receive test misses. The desk should be able to identify which request is newer and prevent both versions going to the field crew. Have the person covering the desk read back the latest agreed details, the time they received them and the name of the colleague taking the job.
Restore the normal route in the exercise and have the customer resend the first request. What stops somebody booking it twice? Give the fallback record a reference the ordinary bookings process can retain. Record the original timestamp, the action the desk already took and the person responsible for reconciliation. A fallback that leaves a queue of ambiguous commitments can cost more staff time to unwind than the original interruption.
Price the move using one stubborn file and one difficult user account
Choose a spreadsheet somebody would genuinely struggle to rebuild. Use a sanitised copy. Export, import, edit, share and return it to a colleague who stays on Microsoft 365. Ask that colleague to reproduce the number used for a decision, such as the available hours in next week's roster. Opening the file successfully is only the first part of the test. Look for changed formulas, broken references, lost protection and print output that no longer fits the worksheet.
Choose a second test around a contractor who works for several clients. Can the contractor join the right meeting and access the right folder without accidentally using another client's identity? Can the sponsor remove access at the end of the engagement? Test the invitation from the contractor's device, including what happens to an old link. A pilot run entirely from the administrator's browser tends to hide these costs.
Build the business case from the work these tests expose. List subscription overlap, the hours needed to repair templates, the time for guest onboarding, support arrangements and the cost of retaining an archive somebody can actually search. Ask the vendor to identify the paid licence required for each critical function. Leave estimates as ranges with an owner until a supplier or staff member can substantiate them.
Keep a decision record that survives the next interruption
Write the decision in a paragraph your service manager could read to a relief dispatcher. Include the route for new requests, the route for existing jobs, the point at which the desk changes over, the person allowed to announce recovery and the location of the reconciliation list. Record where customers will see the change. Have the relief dispatcher follow it without coaching and mark every place they had to guess.
Suppose the drill succeeds with an independent phone route and the spreadsheet round trip fails. Keeping the suite and funding the fallback is a defensible outcome. Suppose the documents work well on another suite, staff prefer its workflow and the cost of maintaining both during transition is acceptable. That is evidence for a limited migration. An outage can justify investigating either option. The results of your own tests tell you which purchase addresses the problem.
For a wider comparison, use the AI workspace and platform guide and the guide to Google Workspace alternatives. If the decision concerns coordination around the inbox, the Cortex platform overview describes that role. Keep your continuity drill in the incident folder even if you decide to move. Every new provider introduces its own service dependencies.
Questions for the purchasing meeting
Give a printed or otherwise independently accessible copy to the person who opens the office. Arrange a brief rehearsal during a normal shift change, when the outgoing and incoming dispatchers can confirm which unresolved requests have transferred.
Related guides: AI workspace alternatives for small teams: what belongs on your shortlist?.
Frequently asked questions
Does a second Outlook client provide an independent fallback?
Another client may help when one application has a local problem. If both clients use the same unavailable mailbox service or the same unavailable sign-in path, both can fail together. Test a separate route end to end, including how the customer discovers it.
Should we move our email domain during an active outage?
A rushed change can leave messages, identities and calendars split between systems. Follow the incident and continuity procedures appropriate to your organisation. Rehearse migration with a controlled pilot, clear ownership and a way to reconcile unexpected deliveries before committing the business domain.
What should we ask a supplier to prove before moving the team?
Ask a user to recover an old conversation, open an important attachment, share a calendar slot with an external guest and revoke access afterwards. Record which steps required administrative assistance. A suite that technically supports all four tasks can still be expensive for your team to operate.
How should we communicate recovery?
Name one person to announce it. First ask the desk to send and receive a fresh external message, then reconcile the fallback list and acknowledge any outstanding bookings. Publish a short status update using the same route customers were told to check during the interruption.
Sources
- Microsoft service status
- Microsoft 365 business catalogue
- Google Workspace product list
- Zoho Workplace overview
- Fulcrum Mail
General information only, not legal, financial or professional advice. Check the source for your circumstances.



