Technology · Website Design

What to Check First When Your Business Website Goes Offline

Respond to a website outage with precise observations, limited scope checks, clear escalation, a customer fallback and verified recovery.

Published

What to Check First When Your Business Website Goes Offline

When a business website appears to be offline, the first useful step is to establish what has failed. A homepage that will not load, a broken enquiry form and an unavailable administration screen are different incidents. They can also have different causes. A calm initial check gives your website team evidence to investigate and helps the business decide what customers need while the problem is being resolved.

You do not need to diagnose every technical layer yourself. You do need to record the symptoms accurately, understand the affected tasks and contact the people responsible for the relevant services. Avoid making several unrelated changes before anyone has captured the original condition. That can make the incident harder to explain and recover from.

Record the exact symptom and its timing

Write down the full address you were trying to use, the time and timezone, and what appeared on the screen. Capture the visible error message or status code if one is shown. A screenshot can help preserve wording that is otherwise remembered only as an error.

Record whether the problem began during a specific task. Did a page fail when you followed a link, when you signed in or when you submitted a form? Include what happened immediately before the failure without recording passwords or unnecessary personal information.

Separate observation from interpretation. A browser reporting that it cannot reach a site does not by itself prove that the hosting server has failed. A page displaying a technical error does not automatically mean that all pages are affected. Give support the evidence before the conclusion.

Check the scope with a small set of tests

Try the homepage and a known important service page. If both work, test the affected journey carefully with suitable test information. A contact form might display normally while failing to submit or deliver its message. The visible page is only part of the process.

Where practical, compare another device or connection and ask a colleague to check the same address. This can help distinguish a wider incident from a local connection or device issue. Record the result instead of repeatedly refreshing the same page and assuming nothing has changed.

Keep the tests limited and purposeful. Repeated payments, bookings or form submissions can create duplicate activity if part of the system is still working. Use an agreed harmless test, and tell the receiving team what you are checking.

Use error details to guide the support conversation

HTTP response codes provide clues about an incident. A 404 response relates to a resource not being found, while 5xx responses describe server-side errors. The code is a clue about the response, rather than a complete diagnosis of the underlying cause. Our application monitoring and incident evidence guide provides the related checks.

When contacting the host or administrator, record the precise error, time and timezone, and affected URL. Add the scope of the interruption and any recent changes. These details help the responsible team investigate the incident without relying on an incomplete description. Our incident response and support guide provides the related checks.

If a browser shows a certificate or security warning, record it and ask the responsible technician to investigate. Do not treat bypassing the warning as a repair. The warning is part of the symptom that needs to be understood.

Identify which customer tasks are affected

List the journeys that matter immediately: viewing service information, sending an enquiry, making a purchase, signing in to an account or managing an existing booking. Mark each as working, failing or not yet checked. This gives the business a clearer view than a single statement that the website is down.

Consider connected tools as well. A booking service embedded in the site may be unavailable while ordinary pages work. A form may save an enquiry in the CMS while its notification email fails. These partial failures affect the response your team should give customers.

Ask staff handling enquiries whether they have observed missing messages or customer reports. Their evidence can help establish the start and extent of the problem. Avoid assuming that no complaints means no impact; some visitors will leave without reporting a failure.

Review recent changes and provider status

Check whether anyone recently released code, updated the CMS, changed a domain setting or moved a connected service. Record what changed and when. A recent change is a candidate for investigation, not proof that it caused the incident.

Look at the relevant provider's official status information where available. Match the affected service and region or platform to your arrangement. A general incident notice may be relevant, but your website team still needs to confirm whether it explains the symptoms.

Share this timeline with support. Include maintenance work that was planned as well as changes made in response to the failure. An accurate sequence helps the technician decide what evidence to inspect and whether a controlled rollback is appropriate.

Contact the responsible team with a clear report

Use your agreed support route and include the observations gathered so far. A useful report gives the affected addresses, timestamps, visible errors, scope tests, important business impact and relevant recent changes. Name someone who can answer follow-up questions.

Clarify who is coordinating the incident when several suppliers are involved. Your developer, host and an external integration provider may each own a different part of the journey. The business needs a point of coordination while the cause is still uncertain.

Ask how updates will be communicated and when you should expect the next status check. Distinguish confirmation that the report was received from evidence that the service has been restored. Keep the contact route available outside the affected website if necessary.

Preserve evidence before making changes

Ask the technician to retain relevant logs and configuration information before applying a fix where practical. Different layers can record different parts of an incident, so the website's visible error message may need to be connected to hosting or integration evidence.

Keep an incident note of actions taken and their results. Record who made each change and when. This does not need to be elaborate, but it should allow someone joining the investigation to understand what has already happened.

Do not remove security settings or change several services speculatively to see whether the website returns. A deliberate test needs an owner, a reason and a way to assess its effect. The investigation should reduce uncertainty rather than introduce additional unknowns.

Provide a practical customer fallback

Decide how customers can complete the affected task while work continues. You might direct enquiries to a working telephone number or explain an alternative booking route through an existing communication channel. Confirm that the team can handle the additional activity before promoting the route.

Keep wording accurate and limited to what you know. State which service is affected and how customers can get help. Avoid promising a restoration time unless the responsible team has provided a dependable basis for it.

Track requests handled through the fallback so that they can be reconciled afterwards. An incident can create duplicate enquiries or bookings when customers try several routes. Staff need enough context to recognise that activity and respond consistently.

Verify the recovery through real journeys

When support reports a fix, repeat the important checks that originally failed. Confirm that pages load, forms submit, relevant records appear and notifications reach their destinations. A restored homepage does not establish that every affected journey works.

Check any work created during the incident or recovery period. If data was restored, ask what happened to recent activity and whether reconciliation is needed. Give that task an owner rather than leaving staff to discover missing records later.

Record the final outcome and the follow-up work that would make another incident easier to detect or manage. Giraffe Digital's website design service can help connect the website's important journeys to practical technical requirements. A useful outage response leaves the business with evidence, clear responsibilities and confirmed recovery.

Frequently asked questions

How should we record the exact symptom and its timing?

Write down the full address you were trying to use, the time and timezone, and what appeared on the screen. Capture the visible error message or status code if one is shown. A screenshot can help preserve wording that is otherwise remembered only as an error. Record whether the problem began during a specific task. Did a page fail when you followed a link, when you signed in or when you submitted a form? Include what happened immediately before the failure without recording passwords or unnecessary personal information. Separate observation from interpretation. A browser reporting that it cannot reach a site does not by itself prove that the hosting server has failed. A page displaying a technical error does not automatically mean that all pages are affected.

How should we use error details to guide the support conversation?

HTTP response codes provide clues about an incident. A 404 response relates to a resource not being found, while 5xx responses describe server-side errors. The code is a clue about the response, rather than a complete diagnosis of the underlying cause. Our application monitoring and incident evidence guide provides the related checks. When contacting the host or administrator, record the precise error, time and timezone, and affected URL. Add the scope of the interruption and any recent changes. These details help the responsible team investigate the incident without relying on an incomplete description. Our incident response and support guide provides the related checks. If a browser shows a certificate or security warning, record it and ask the responsible technician to investigate.

How should we review recent changes and provider status?

Check whether anyone recently released code, updated the CMS, changed a domain setting or moved a connected service. Record what changed and when. A recent change is a candidate for investigation, not proof that it caused the incident. Look at the relevant provider's official status information where available. Match the affected service and region or platform to your arrangement. A general incident notice may be relevant, but your website team still needs to confirm whether it explains the symptoms. Share this timeline with support. Include maintenance work that was planned as well as changes made in response to the failure. An accurate sequence helps the technician decide what evidence to inspect and whether a controlled rollback is appropriate.

How should we preserve evidence before making changes?

Ask the technician to retain relevant logs and configuration information before applying a fix where practical. Different layers can record different parts of an incident, so the website's visible error message may need to be connected to hosting or integration evidence. Keep an incident note of actions taken and their results. Record who made each change and when. This does not need to be elaborate, but it should allow someone joining the investigation to understand what has already happened. Do not remove security settings or change several services speculatively to see whether the website returns. A deliberate test needs an owner, a reason and a way to assess its effect. The investigation should reduce uncertainty rather than introduce additional unknowns.

How should we verify the recovery through real journeys?

When support reports a fix, repeat the important checks that originally failed. Confirm that pages load, forms submit, relevant records appear and notifications reach their destinations. A restored homepage does not establish that every affected journey works. Check any work created during the incident or recovery period. If data was restored, ask what happened to recent activity and whether reconciliation is needed. Give that task an owner rather than leaving staff to discover missing records later. Record the final outcome and the follow-up work that would make another incident easier to detect or manage. Giraffe Digital's website design service can help connect the website's important journeys to practical technical requirements. A useful outage response leaves the business with evidence, clear responsibilities and confirmed recovery.

TechnologyWebsite Design