Skip to main content

When Microsoft Stops: A Case Study in Single Supplier Dependency

Resilience
Be Ready Business Continuity Microsoft 365 Single Supplier Case Study

Last Friday, Microsoft 365 stopped working. A configuration failure took down email, Teams, SharePoint and OneDrive for 130,000 organisations across 74 countries. UK police couldn't access records. Schools couldn't take registration. NHS trusts couldn't see patient histories. The FAA grounded 11,000 flights.

Microsoft called it a config error. They resolved it. Everyone moved on.

But for a small firm that depends on Microsoft for everything — email, files, invoicing, communication — an 11-hour outage is not a news story. It is a day where the business does not work.

This is what that looks like.

The case study

Sarah runs a 12-person logistics firm in Gateshead. They move specialist medical equipment for NHS trusts across the North East.

At 8:15 on Friday morning, Sarah opened her laptop and saw the M365 error. Not a cyber-attack. A Microsoft configuration failure. The same one that would make the news that afternoon.

Her drivers couldn't access their delivery sheets (SharePoint). Her dispatch team couldn't update customers (email down). Her invoicing system — also in SharePoint — was gone.

By 10am, she had six missed calls from NHS trusts asking where their equipment was. By noon, two drivers were sitting idle because they couldn't get their routes. By 3pm, she'd lost a day's billing and was writing manual delivery notes on paper.

The outage lasted 11 hours.

Sarah didn't have a backup plan. She had Microsoft's cloud, and when it stopped, she had nothing.

The recovery wasn't the hard part

When Microsoft fixed the config and the systems came back, Sarah's email arrived, her files reappeared, her invoicing resumed.

The hard part was the 11 hours where her business couldn't function because one supplier stopped working.

Six missed calls from NHS trusts. Two idle drivers. A day's billing lost. Manual delivery notes on paper that someone now has to transcribe. Customer confidence dented.

None of that was Microsoft's problem. Microsoft's problem was a config error. Sarah's problem was that she had no plan for when her supplier stopped.

What Sarah didn't have

When the systems stopped, Sarah didn't have:

  • A restore plan for when M365 goes down. Not "if" — "when". Every supplier has an outage. The question is not whether, but when, and for how long.
  • A way for drivers to access delivery sheets offline. The sheets existed in SharePoint. When SharePoint stopped, the sheets stopped. An offline copy — a PDF, a printed route sheet, a second system — would have kept the drivers moving.
  • A manual invoicing process she could switch to. The invoicing system was in SharePoint. When SharePoint stopped, invoicing stopped. A paper invoice template, or a second invoicing tool, would have kept the billing flowing.
  • A single document her team could open and follow when the systems stopped. Not a plan buried in a shared drive (which was also down). A one-page document, printed, by the phone, that said: "When X stops, do Y."

She had Microsoft. She didn't have a plan.

The pattern

Sarah's situation is not unusual. It is the default for any firm that has moved everything to a single cloud provider without asking: "What happens when this stops?"

The question is not theoretical. Microsoft has had outages before. Google Workspace has had outages. AWS has had outages. Every major provider has had outages. The question is not whether your provider will stop. It is when, and what you will do when it does.

The answer is not "use a different provider". The answer is "have a plan for when your provider stops".

The plan

The plan is not complicated. It is the same discipline as the Be Ready guide for power outages, applied to a single supplier:

  1. Name the single point of failure. "We depend on Microsoft for email, files and invoicing."
  2. Ask: what happens when it stops? Not "if". "When".
  3. Name the critical processes that stop. Email. Files. Invoicing. Communication.
  4. For each one, name the backup. Offline access. Paper process. Second system.
  5. Test the backup. A backup you have never used is a hope, not a control.

None of this costs money. It costs thought, written down before the bad morning rather than during it.

The ten checks

The free Be Ready guide has ten checks for digital resilience, none of which cost anything. They cover:

  1. Naming single points of failure
  2. Testing a backup restore
  3. Keeping an offline copy of money documents
  4. Having a second communication channel
  5. Being able to take payments offline
  6. Three days of supplies at home
  7. The one-page plan
  8. Power and battery audit
  9. Cold chain readiness
  10. Supplier and customer continuity

The M365 outage hit four of these: single points of failure (1), backup restore (2), offline money documents (3), second communication channel (4).

The full printable guide is free. Keep it by the till. For a hands-on walkthrough, book a resilience session and complete the ten checks for your firm on one page.

Frequently asked questions

What caused the Microsoft 365 outage?

Microsoft have attributed the outage to a configuration failure. The details are still emerging, but the pattern is familiar: a change to the system, unintended consequences, widespread impact. The same pattern has caused AWS outages, Google Workspace outages, and every other major cloud provider outage. The cause is less important than the response: what happens when your supplier stops?

How long did the Microsoft 365 outage last?

The outage lasted approximately 11 hours for most organisations. For some, it was longer. For a firm that depends on Microsoft for everything, 11 hours is a day where the business does not work.

What should I do if Microsoft 365 goes down?

The answer depends on what you have planned before it goes down. If you have offline access to critical files, a manual invoicing process, and a second communication channel, you can keep trading. If you do not, you stop. The plan is the difference between an inconvenience and a day where the business does not work.

How do I prepare for a Microsoft 365 outage?

Start with the ten checks for digital resilience. Name your single points of failure. Ask what happens when they stop. For each critical process, name the backup. Test the backup. The plan is not complicated. It costs thought, written down before the bad morning rather than during it.

Is Microsoft 365 reliable?

Microsoft 365 is as reliable as any other major cloud provider, which is to say: very reliable, but not infallible. Every provider has outages. The question is not whether your provider will stop. It is when, and what you will do when it does.

Where this sits

This is the same discipline as the Be Ready guide for power outages, applied to a single supplier. The exposure is the same: a single point of failure, a bad morning, a business that stops. The response is the same: name the exposure, plan the response, test the plan.

If you run a firm of between five and fifty people, the ten checks are on one printable page, free, at fleurlamont.com/guide.html. Keep it by the till. The point is not to predict the bad day. It is to make the bad day short.

Is your firm ready for three days of disruption?

The free ten-check guide covers backups, suppliers, power, cold chain, and the return. Ninety-minute resilience sessions available in the North East.

Download the free guide · Book a Guided Session