Insights

When the platform everyone uses stops

On 12 November 2026, Microsoft 365 started to fail for organisations around the world. For close to a day, many people couldn’t open their email, and parts of Teams and Copilot stopped working. Businesses, schools and governments were all affected. Microsoft says the cause was a problem introduced during an earlier upgrade to its network, not a cyberattack.

Outages like this are not unusual, and they are not unique to one company. The dependence isn’t new, but each outage like this shows again how much rests on so few providers. Email, chat, files, sign-in and the systems behind them often come from one or two suppliers, so a single fault can stop several critical activities at once, for thousands of organisations at the same time.

Few organisations can, or should, stop using large providers. Their services are usually more reliable and better protected than most could run themselves. The aim is different: to know what depends on each provider, to decide how long you could manage without it, and to have a plan ready for the day it stops. This Insight article does that with six requirements from the RiskWitness Blueprint.

What matters most

  • Your critical activities, such as serving customers, taking payments and paying staff, in an order leadership has agreed RW-RES-18
  • What each one relies on: the systems, information, suppliers and people behind it

What they rely on

  • Email, chat, files and sign-in, often all from the same provider
  • One provider behind several critical activities, so one outage stops them all at once
  • Each critical supplier's risks in your risk register, with an owner and a decision RW-SUP-07

The day it stops

  • A continuity plan for each critical activity, saying how it carries on RW-RES-08
  • Personal email and messaging apps as a fallback: workarounds are kept as secure as normal working RW-RES-09
  • A way for the response team to talk that doesn't depend on the provider that's down RW-DET-28
  • If it never comes back: an exit plan for getting your data and moving on RW-SUP-09
  • Known and owned
  • Planned for in advance
  • Ruled out
  • Single point of failure
What depends on a large provider, and where each requirement applies. Leadership agrees which activities matter most and what each relies on (RW-RES-18). The services behind them often come from the same provider, so each critical supplier's risks are in the risk register (RW-SUP-07). For the day the provider stops, each critical activity has a continuity plan (RW-RES-08), its workarounds are as secure as normal working (RW-RES-09), the response team has a way to talk that doesn't rely on the provider (RW-DET-28), and an exit plan covers the case where the service never comes back (RW-SUP-09).

About the RiskWitness Blueprint

The RiskWitness Blueprint is a free information security framework that we publish for organisations of every size. It sets out 249 requirements across a core and eleven areas, from identity and access to suppliers and artificial intelligence. Each requirement says what to do and why it matters, and leaves the choice of how to the organisation.

Requirements sit in three tiers, and each tier includes everything below it:

  • Foundation: what every organisation should have, whatever its size, including the controls that stop the most common attacks.
  • Established: a managed programme, with deeper risk work, supplier assurance, testing, monitoring and regular review.
  • Advanced: for organisations with more at stake, with round-the-clock monitoring and deeper, independent testing.

The Blueprint is also proportionate. It says what to achieve, not how far to take it, so a small firm and a large one can both meet the same requirement in ways that suit their size and risks. The full framework can be downloaded free.

What the requirements ask

RW-RES-18Foundation

Critical services and what they rely on

Identify the services your organisation and your customers depend on most, record which systems, information, suppliers and people each one relies on, and have senior leadership agree their order of importance.

RW-SUP-07Established

Supplier risks in your risk register

Record the security risks that each critical supplier brings in your risk register, and assess and treat them alongside your other risks.

RW-RES-08Established

Continuity plans for critical activities

Keep a continuity plan for each critical activity that sets out how it would carry on if its systems, its premises or its key people were unavailable.

RW-RES-09Established

Workarounds that stay secure

Make sure the workarounds in your continuity plans keep information as well protected as your normal ways of working do.

RW-DET-28Established

A backup way to communicate in an emergency

Set up a secure way for the people who respond to incidents to reach each other that doesn't depend on your normal email, chat or phone systems, and check from time to time that it works.

RW-SUP-09Established

An exit plan for each critical supplier

Keep an exit plan for each critical supplier, including cloud services, that sets out how you would get your data back and move to an alternative.

The first is a Foundation requirement, because everything else depends on it: until leadership has agreed which activities matter most and what each relies on, nobody can say which outage would hurt. The other five are Established, because each needs a running process. Supplier risks are assessed and reviewed alongside every other risk. Continuity plans are written, kept secure and exercised. A second way to communicate is set up and checked. An exit plan is kept current for each critical supplier. Together they turn “we use Microsoft” or “we’re on AWS” from a fact nobody thinks about into a dependency that is known, owned and planned for.

Why they matter

One fault, many organisations

The November outage followed a familiar pattern. It began at about 1pm US Eastern time on 12 November, and MacRumors, citing the Associated Press, reported nearly 130,000 reports of Outlook problems on Downdetector shortly afterwards. Microsoft first saw servers become unavailable and the rest overloaded. It later said the issue had been introduced during an earlier network capacity upgrade and surfaced after a more recent network change. Microsoft said it was not related to a cyberattack. It restored service in stages, and extended the time it would keep trying to deliver delayed email to 48 hours so that messages weren’t lost. It has said a full post-incident report will follow.

The causes vary, but the shape repeats. In October 2025, a latent fault in the automation that manages DNS records for one Amazon Web Services database service left an empty record in its Northern Virginia region. The core problem lasted under three hours, but its knock-on effects on other services stretched the event to about 14.5 hours. In July 2024, a faulty update to CrowdStrike’s security software stopped an estimated 8.5 million Windows devices. Microsoft said the incident demonstrated “the interconnected nature of our broad ecosystem”. None of these was an attack. Each was an ordinary change or fault inside one provider, felt by its customers all at once.

Regulators now treat this as a risk in its own right

Financial regulators have started overseeing the largest providers directly. Under the EU’s Digital Operational Resilience Act (DORA), the European Supervisory Authorities designated their first critical ICT third-party providers on 18 November 2025. FStech reported that the 19 providers included Amazon Web Services, Google Cloud and Microsoft. In the UK, the Bank of England, the Prudential Regulation Authority and the Financial Conduct Authority began overseeing four critical third parties on 13 July 2026. Nikhil Rathi, the FCA’s chief executive, said that “when the same providers serve thousands of firms, a single failure can reverberate across the financial system”.

That oversight doesn’t move the responsibility. The UK regulators were clear that firms “remain responsible for managing their own third-party arrangements”, including due diligence, risk management and contingency planning. In Australia, APRA’s operational risk standard, CPS 230, has applied to banks, insurers and superannuation trustees since 1 July 2025. It requires them to manage the risks from material service providers and to plan for severe disruptions. These rules apply to financial services, but the expectation behind them applies to any organisation: know what you depend on, and be ready to carry on without it.

Plan for the abrupt loss of a supplier

National cyber security agencies give the same advice. The UK’s National Cyber Security Centre, in guidance updated on 1 October 2026, says a clear understanding of business-critical services and their dependencies helps organisations find workarounds and decide what to restore first. It advises planning for the “abrupt loss of supplier services”, understanding suppliers’ own resilience arrangements, and testing continuity plans regularly. It also notes that manual workarounds may provide short-term mitigation, but they are not a long-term answer.

How to meet them

How far an organisation takes these requirements depends on its size, how much it relies on one provider and how long its critical activities could stop.

A small organisation

Typical set-up

Around twenty people, with email, files, calendars and sign-in all from one cloud productivity suite, and an IT provider who manages it.

What it could look like

  • List your three to five most important activities, such as answering customers, invoicing and paying staff, and note which services each one needs. For most, the answer will be the same suite.
  • Record the suite’s provider as a critical supplier, with the risk of it being unavailable for a day in your risk register, an owner and a decision.
  • Write a one-page plan for a day without it: who calls whom, where the key phone numbers are kept, and how urgent customer requests reach someone.
  • Agree which workarounds are allowed, such as a shared mobile phone or a printed contact list, and which are not, such as forwarding work to personal email.
  • Know where you would get your files back from if you ever had to leave, and keep a recent copy somewhere else.

A larger organisation

Typical set-up

Hundreds of people across several sites, with one cloud provider for productivity, another for hosting, and many software services that depend on both.

What it could look like

  • Have leadership agree the critical services in order of importance, and map each to the systems, providers and people it relies on, including hidden dependencies such as single sign-on.
  • Assess each critical provider’s risks alongside your others, look for activities that share a single provider, and decide which to reduce, accept or plan around.
  • Give each critical activity a continuity plan that covers the loss of its main provider, with how long it can stop, and exercise the plans each year.
  • Set up a way for the incident team to reach each other that runs on a different provider from your everyday tools, and check that it still works.
  • Keep an exit plan for each critical provider, including how to get your data back, and review it when the contract or the service changes.

How to show it’s done

Leadership, auditors and customers will want to see more than a list of suppliers. Useful evidence includes:

  • A list of critical services agreed by leadership, in order of importance, showing the systems, suppliers and people each relies on.
  • Each critical provider’s risks in your risk register, with an owner, a score and a decision.
  • A continuity plan for each critical activity, with the date of the last exercise and what was changed as a result.
  • The workarounds your plans allow, and confirmation that each keeps information as secure as normal working.
  • Details of the backup way to communicate, and when it was last checked.
  • An exit plan for each critical provider, reviewed within the last year.

Where it usually falls short

One supplier counted as one risk

An organisation lists “Microsoft 365” once, but email, files, calendars, chat and sign-in all stop together. The dependency is far bigger than the line in the register suggests.

Hidden dependencies

Other software often relies on the same provider, through single sign-on or hosting, so a service that seems separate stops as well.

Plans kept in the system that's down

A continuity plan saved only in cloud files, or an incident call arranged over the chat tool that has stopped, can’t be reached when it’s needed.

Unsafe workarounds

When the normal tools stop, people reach for personal email and messaging apps. Information leaves the organisation’s control, and attackers know crises are a good time to send convincing messages.

Exit plans nobody believes

A plan to move everything to another provider in a week isn’t realistic. An exit plan is useful when it is honest about how long a move would take and what you’d do meanwhile.

How RiskWitness helps

RiskWitness is a hosted information security risk register. It ships with the Blueprint built in, with starter controls already mapped to its requirements, and it can be used to manage these six requirements from start to finish. Suppliers and assets are on the Team plan and above, which also includes all three tiers of the Blueprint, as the Established requirements here need.

  1. Record the provider as a supplier

    Add the provider and answer four questions about the information it holds, the access it has, how much you depend on it and how hard it would be to replace. If key operations would stop within hours, it is Critical, and RiskWitness suggests reviewing it every six months. Record whether the contract covers business continuity and service levels, and the return or deletion of your data at the end.
  2. Record what depends on it

    Add each service you use from the provider, such as email, file storage and sign-in, as an asset with an owner, where it’s hosted and how much its availability matters. Link each one to the supplier, so you can see everything that relies on it in one place.
  3. Record the risk

    Start from risks in the library such as “Outage of a cloud service we rely on” or “Over-reliance on a single supplier”. Give the risk an owner, score it before and after controls, link it to the supplier and assets, and decide how to treat it.
  4. Accept what you can't remove, with an end date

    Some dependence on a large provider is a choice worth making. If you accept the risk, record the reason and an expiry date. If it’s outside your tolerance, a named approver signs it off and confirms it’s them.
  5. Make the plans controls, and test them

    Use starter controls such as “Business continuity plans”, “Continuity exercises” and “Exit plans for critical suppliers”, map each to its Blueprint requirement, and record every test with its result. RiskWitness shows how well each requirement is met.
  6. Give each gap an owner

    If an exercise shows the plan was stored in the system that went down, or nobody knew the backup phone numbers, create an action with an owner and a due date. Overdue actions show on the dashboard.
  7. Record the outage when it happens

    Record a supplier incident linked to the assets and the risk, with who is handling it and which regulators may need telling. Afterwards, review the risk’s score with what you learned, and produce a PDF report or slide deck for leadership.

Every step is kept in a history that shows who did what, when, and the old and new values, and nobody can edit or delete it. When leadership, an auditor, a customer or a regulator asks what would happen if your main provider stopped, the answer is already written down.

Sources