Cloud Outage: How to Keep Your Small Business Running

A booking is due, a customer is waiting and your team is ready. Then the system holding the appointment details stops loading. Another customer wants to pay, but the payment service is unavailable too.

That is when a software problem becomes an operating problem. Waiting for the provider to fix it may be unavoidable. Leaving your team without a clear next step is something you can prepare for.

The Google Cloud disruption on 20 October 2026 provides a practical reason to review your arrangements. A small business cloud outage plan should explain how you will serve customers safely while essential services are unavailable, and how you will put the records right afterwards.

What happened during the Google Cloud outage?

According to Google’s official incident record, a power event caused an outage in its London zone, europe-west2-b, with some services in the wider region experiencing errors and intermittent network issues.

The disruption began at 07:08 BST on 20 October. Google reported recovery for most workloads from 13:10 BST. Its 15:03 BST update said the issue had been mitigated for most customers, while a small number remained affected. These times have been converted from the Pacific times used on Google’s dashboard.

At the time of checking on 21 October, Google said it would publish an analysis after completing its investigation. This was not evidence that every Google service or every customer was affected.

Sky News reported disruption to banking and delivery services, including accounts from people struggling to take payments and access orders. Those accounts illustrate possible consequences; they do not establish the experience of every business.

Which parts of your business would stop first?

Start with the work customers are relying on you to complete today. List the information and systems needed to do it, then ask what happens if each becomes unavailable.

For an appointment business, the immediate problem might be finding the time, location, assigned person and suitable contact details. For a retailer, it might be identifying paid orders awaiting dispatch. For a consultant, it could be accessing the documents needed for a client meeting.

My starting point would be the customer commitment. Which commitments can still be delivered safely? Which can wait? Which must be postponed because essential information cannot be checked?

Give each activity a maximum tolerable interruption based on its real consequences. A weekly management report may wait until tomorrow. A customer expecting someone at their premises this morning needs a decision much sooner.

Name the person who can make that decision and who covers their absence. A plan that requires the owner to approve everything can become another source of delay.

Keep essential information accessible securely

A backup is useful only if the authorised person can access what they need when the main service fails. A file stored behind the same unavailable sign in may offer little immediate help.

Consider whether a controlled, regularly refreshed copy of essential operational information is appropriate. It should contain the minimum needed for current commitments, show when it was updated and be accessible only to the people who need it. Protect devices and files, set a retention period and remove obsolete copies.

Do not turn an availability problem into a confidentiality problem by copying complete client records onto personal devices. Where sensitive information is necessary for safe delivery and cannot be accessed appropriately, pausing the service may be the right decision.

The ICO’s data security guidance covers restoring access to personal data after physical or technical incidents and testing the effectiveness of security measures. It emphasises measures appropriate to the organisation and the risks involved. This is existing guidance, not a new rule introduced because of this outage.

Check whether your alternative would actually work

Buying another subscription is not the same as establishing a usable alternative. Two services with different names may depend on some of the same infrastructure. A separate payment route may still require the same device, connection or sign in.

Ask providers practical questions: what happens during an outage, where are updates published, what information can be exported and what does recovery require from you? Ask about shared dependencies where relevant, without assuming a small business can inspect every part of a supplier’s infrastructure.

Then test the part you control. Can the right employee access the alternative? Is it configured? Do they know when to use it? Can they record what happened without creating a second, conflicting set of records?

If you are reviewing contracts as part of this exercise, the questions in Microsoft 365 Renewal: What Small Businesses Should Check provide a useful starting point.

Agree what staff should do during an interruption

Write a short procedure covering the first decisions. Check the provider’s official status information, identify the affected activities and appoint one person to coordinate updates. Record when the disruption was noticed and which temporary arrangements were authorised.

Set clear rules for provisional bookings, delayed payments and services that must pause. Do not leave each member of staff to invent their own approach.

Where a payment appears uncertain, check its status before asking the customer to try again. An error message alone may not tell you whether a transaction completed. Follow the payment provider’s guidance and keep a record of the reference and outcome.

If you choose to allow payment later, decide who can approve it, what limit applies and how it will be followed up. Deferred payment creates a credit decision. Our guide to when to stop offering customers credit explains the wider considerations.

Prepare a brief customer message that states what is affected, what the customer should do and when you will update them. For example:

Our booking system is temporarily unavailable. We are checking today’s appointments using our agreed backup process. Please contact us by phone if your appointment is urgent. We will provide another update at 2pm.

Only use that wording if the backup process and phone route genuinely work. Avoid promising a restoration time that the provider has not confirmed.

Reconcile records before returning to normal

The return of the app is not the end of the job. Temporary bookings, customer messages and payment attempts need to be checked against the restored system.

Give one person responsibility for reconciliation. Compare records before entering changes. Check whether orders or payments already completed, resolve duplicated bookings and confirm any promises made to customers during the interruption.

Keep a record of unresolved items and who will follow them up. Remove temporary copies when they are no longer needed under your retention arrangements.

If the incident also raises concerns about the loss, exposure or availability of personal data, assess those separately. Our CRM data breach guide covers the response to data protection concerns. A supplier outage should not automatically be described as a cyberattack or data theft.

A simple worksheet for your outage plan

Complete one row for each critical activity. Choose timings that fit your business rather than copying a generic recovery target.

Test your plan before you need it

Run a short practice exercise without disrupting live customer work. Ask the team to explain how they would handle one appointment, order or payment if the main service were unavailable. Check access and permissions, then record the steps that were unclear or impossible.

A second service adds cost and administration. Manual processes introduce their own risks. The aim is to choose arrangements proportionate to what could go wrong, including a clear decision to pause where continuing would be unsafe.

Before spending money on another tool, I would want answers to four questions: can we find today’s commitments, contact the right customers, record changes and confirm what has already been paid?

A useful plan turns those answers into something your team can follow under pressure.

By Julian Frincu, founder of Skills 2 Grow. Incident information checked on 21 October 2026. Google is the primary source for the incident details; Sky News provides reporting on customer experiences; ICO guidance informs the discussion of data availability and security. The operational recommendations and worksheet are Skills 2 Grow’s practical analysis.