Company System Down: Immediate Action Guide for Managers

By Pedro Corgnati, Founder of SystemForge

When a company's system goes down, the sequence of actions in the first 60 minutes determines whether the problem lasts hours or days. The first priority is not to fix the system — it's to isolate the scope of the problem, reach the right support, and communicate with your team to prevent parallel actions that make things worse. Companies that recover quickly are the ones with this protocol defined before the crisis happens.

A system outage during business hours is one of the most stressful situations for any manager. Losses compound every minute: unprocessed orders, blocked teams, customers without support. The difference between a 2-hour and a 2-day outage is usually not in the complexity of the problem — it's in the speed and quality of the response.

This guide is a practical action plan for this crisis moment.

---

The First 10 Minutes: What to Do Right Now

In the first 10 minutes after identifying the outage, execute in this order:

1. Confirm whether it's a total or partial outage Ask 2-3 people in different departments if they have the same problem. A single user issue could be their device. Three users in different departments with the same symptom indicates a system-wide outage.

2. Document the exact time the outage started Note (or ask someone to note) the time the problem began and the first reported symptoms. This significantly accelerates technical diagnosis.

3. Activate your responsible technical support If you have an active support contract, call now. Don't send an email. Call. If you don't have a support contract, this is the time to find emergency support — see the section below.

4. Communicate with your team with a clear message Send a message to everyone: "The system is down. Do not try to restart anything, do not make any changes. Wait for further communication." This prevents people from taking random actions that complicate diagnosis.

5. Activate temporary manual mode For critical operations that cannot stop (invoicing, payments), activate your manual contingency process. If one doesn't exist, document today what you'd do manually — that becomes your plan B.

---

What Might Have Caused the Outage

Understanding the category of the problem helps identify who to call and a realistic recovery timeline:

Problem type	Common symptoms	Average resolution time
Server/hosting down	All users locked out, no specific error	1-4 hours (depends on provider)
Corrupted database	Connection errors, incomplete data on access	2-8 hours
Update that broke the system	Problem started right after an update	1-3 hours (rollback)
Security attack or breach	Unusual behavior, extreme slowness, missing files	4-24 hours
Local network problem	Only on-site staff affected	1-2 hours
Third-party integration error	One specific module stopped, rest works	2-6 hours

Share this with whoever handles your support — it helps direct the diagnosis.

---

How to Find Emergency Technical Support

If you don't have an active support contract, you need to act quickly:

Option 1: The company that built the system If the system is custom-built, the first attempt should always be whoever built it. They know the codebase and can identify the issue in minutes. If they're unresponsive or no longer in business, move to option 2.

Option 2: A specialized IT emergency support company Firms that work specifically with emergency IT interventions exist. They charge more than standard support (expect US$ 150-500/hour for emergency rates), but they resolve issues fast. When hiring, require: diagnosis within 30 minutes, estimated resolution time, and an incident report at the end.

Option 3: Your hosting provider If the problem is in the infrastructure, your hosting provider has a contractual obligation to resolve it. Open an urgent ticket. If it's not resolved within 2 hours, escalate to your account manager or level 2 support.

What NOT to do: Don't hire the first freelancer you find online without checking references. In a crisis, someone unfamiliar with your system can make the situation worse.

---

Post-Recovery: How to Prevent It From Happening Again

When the system comes back online, you have a window of motivation to fix the structural problem. Use it.

Require a complete incident report:

• Root cause identified
• What was done to resolve it
• What can be done to prevent recurrence
• Whether there's a risk of it happening again

Evaluate your support contract: Companies without active support contracts suffer the most from outages. A good contract includes:

• Defined SLA (maximum response and resolution time)
• Proactive monitoring (you're alerted before you feel the problem)
• Regular, verified backups
• Monthly preventive maintenance window

The cost of a monthly support contract (typically US$ 300-2,000/month for mid-size systems) is a fraction of the cost of 4 hours of downtime during business hours.

Implement basic monitoring: Tools like UptimeRobot (free tier available) monitor your system every 5 minutes and send alerts via email or messaging when downtime is detected. Setup takes 10 minutes and can save hours of undetected problems.

---

Frequently Asked Questions

What if technical support takes more than 2 hours to respond?

Escalate to the manager or director of the support company. If you don't have a contract, search for a second company in parallel. In emergencies, having two teams investigating is better than waiting passively.

How much does emergency technical support cost per hour?

Emergency IT support rates range from US$ 150 to US$ 500/hour depending on system complexity and the professional's expertise. For critical systems (ERP, e-commerce, healthcare), expect to pay toward the top of that range. Compare with the cost of downtime: a team of 20 people blocked for 4 hours at US$ 30/hour average cost = US$ 2,400 in direct losses before factoring in lost customers or contracts.

Is cloud-based software more reliable than on-premise systems?

For most small and mid-size businesses, yes. Cloud platforms like AWS, Azure, or SaaS solutions offer 99.9%+ guaranteed uptime and automatic failover that on-premise physical servers rarely match — at comparable costs once on-premise maintenance is factored in.

How do I know if the system is at risk before it goes down?

Warning signs to monitor: progressive slowness over the past weeks, sporadic errors that "fix themselves," growing error logs, backup not tested in more than 30 days, system without security updates for more than 6 months. If any of these are present, schedule a preventive technical review.

Who is legally responsible for losses caused by the outage?

It depends on your contract with the software company or hosting provider. Most contracts define SLAs and penalties for non-compliance. If your contract has no SLA, negotiate one at the next renewal.

---

Need Support Right Now?

If your company's system is down or you want to set up preventive support before the crisis happens, contact us via WhatsApp. We respond within 30 minutes during business hours for emergencies and provide a free initial diagnosis.

We work with web systems maintenance, APIs, databases, and integrations — with defined SLAs and a report for every intervention.