Technology & security · · 5 min read · By Alexander
Website unavailable? An emergency plan for SMEs
If your website is unavailable, a calm and clear procedure is more useful than random technical changes. First establish whether the problem affects only your device, record what you see and give the right contact precise information.
Is the website unavailable for everyone or only for you?
Open the address on a second device and through another connection, such as mobile data instead of the company Wi-Fi. Test a known subpage as well as the homepage. If the site works there, the cause is probably local: browser cache, DNS resolution, firewall, VPN or the internet connection. If it remains unavailable everywhere, treat it as a genuine website outage.
Record the full URL, time, error message and the last time the site was known to work. A screenshot is more useful than “the website is broken”. Check email, the customer area, shop, forms and booking separately so support can distinguish a domain, DNS, server, application or individual-function problem.
- Test another device and internet connection
- Open the homepage and a known subpage
- Record the error, URL and time
- Check email, shop, forms and booking separately
- Do not include passwords in screenshots or chats
What should you do in the first 15 minutes?
Do not change DNS records, plugins, passwords and server settings at the same time. Uncoordinated interventions hide the cause and may prolong recovery. Appoint one person to collect observations, document decisions and communicate with the host or agency while the business protects affected customer processes.
Unknown redirects, changed content, new administrators or suspicious hosting emails may indicate a security incident. The German BSI cyber incident card advises organisations to remain calm, involve their IT emergency contact and isolate affected systems without switching them off prematurely; source checked on 28 October 2026. Do not disconnect or restart a server for a simple availability problem unless the technical owner instructs you to do so.
Who is responsible for each emergency role?
A useful emergency plan can fit on one page. Name a business decision-maker, technical contact, hosting provider, domain or DNS administrator and a deputy. Include customer number, contract name and approved communication channels, but keep passwords in a password manager with controlled emergency access.
Decide in advance who may approve changes and who informs customers. One person can hold several roles in a small company, but several people should not issue conflicting instructions. Germany’s Federal Office of Civil Protection and Disaster Assistance includes responsibilities, deputies and communication paths in its business-continuity guidance; source checked on 28 October 2026.
- Decision: priorities, approvals and business impact
- Technical: diagnosis, evidence, recovery and log
- Communication: customer notice and internal status
- Deputy: takes over when the main contact is unavailable
What information does the host or web agency need?
Send a concise but complete incident report: domain, observed error, start time, affected functions, last known change and tests already performed. State whether the error is permanent or intermittent. Mention immediately if an update, DNS change, certificate replacement or publication happened shortly before the outage.
Ask for a traceable resolution rather than only “it works again”: What caused the outage? Which version or backup was restored? What was tested afterwards? Is there any indication of data loss or a security incident? A website subscription can combine hosting, maintenance and a fixed contact; its operational value is a clear escalation route, not a promise that outages can never occur.
How should you inform customers during the outage?
Communicate when an important customer path is blocked or the outage continues. Use a channel that does not depend on the same website: Google Business Profile, social media, a telephone message or an external status page. State the affected function, an alternative and the time of the next update. Do not speculate publicly about the cause before it is confirmed.
A practical notice is: “Our website is currently unavailable. You can still reach us by [phone/email]. We are investigating and will update this message at [time].” Announce recovery just as clearly. If forms or orders may have been lost, explicitly ask affected customers to submit them again.
How is the website restored safely?
The fastest recovery is not always the safest. Before restoring, establish when the fault began, whether the backup predates it and which newer data may be lost. Preserve recent orders and enquiries separately for shops, booking systems and portals. After restoration, test the homepage, important subpages, login, forms, outgoing email, payment or booking flow and SSL certificate.
A backup is dependable only after a restore has been tested. After the incident, record the cause, duration, affected processes, decisions and improvements. External availability monitoring, separate backups, current contacts and a six-monthly emergency exercise turn the outage into a measurable improvement.
Frequently asked questions about an unavailable website
Why is the website unavailable for me but not for others?
Browser cache, DNS resolution, a VPN, firewall or local connection may be responsible. Test another device over mobile data before changing the server.
Should I restart the server immediately?
Not without diagnosis or instructions from the technical owner. A restart may destroy evidence, interrupt work and hide the underlying cause.
How quickly must the website be restored?
That depends on the business impact. A shop or booking system is more critical than an information-only site, so define priorities and an escalation path in advance.
Does a small business really need an emergency plan?
Yes, but it can be short. One page covering roles, contacts, checks, customer alternatives and recovery is a solid start.
How can I prevent the next outage?
No outage can be ruled out completely. Monitoring, tested backups, controlled updates, current access and clear ownership reduce detection and recovery time.