Do you have physical servers on your premises? Server racks humming in a corner? Local hardware?
After 30 years supporting SMEs across Dublin and Ireland, here are the 5 things I check systematically for IT management in 2026.
1. Test Your Backups. Really.
The classic problem:
“We have backups running every day.”
OK. But do they actually WORK?
What I see too often: backups running for 6 months, nobody has ever tested a restore, and the day you need one — it fails.
The simple test: take 30 minutes. Restore a file. A database. Anything. Just verify it works.
Common traps:
- OneDrive/Google Drive = 30-day retention maximum. File deleted 35 days ago? Gone forever.
- Drive ≠ Backup. Synchronisation is not a backup. Ransomware? Everything is encrypted everywhere.
The real question: “If my server burns tonight, how much data do I recover?” If you hesitate, test it.
2. Check Your Power Supply
The context: power outages are more frequent. Load-shedding, grid issues, local incidents.
What I saw last winter: an SME with a UPS (uninterruptible power supply) not tested in 3 years. Power cut. Dead battery → server shut down abruptly → database corruption.
What to check:
- Does the UPS actually work? Unplug the power. How long does the server hold? 5 minutes? 1 hour?
- Is automatic shutdown configured? When the battery is empty, does the server shut down cleanly — or crash?
- Who intervenes if the outage lasts? 24-hour outage on a Sunday. Who goes on-site to restart?
No miracle solution. Just know what can happen — and have a plan.
3. Clean Up Your Access Rights
The recurring scenario: a colleague leaves. Their access remains active for months. Their OneDrive data disappears after 30 days. You realise 6 months later: “Wait, what did they have on their account?”
What to do:
- Audit active accounts: who still has access? Who left 6 months ago but still has a live account?
- Formalise your offboarding process: when someone leaves — recover their data (OneDrive, emails, Teams), disable access on day one, transfer critical information
- Review admin rights: who has admin rights on your servers? Who knows the critical passwords?
Not complicated. Just a checklist to follow systematically.
4. Monitor Your Server Temperature
The silent problem: your servers are overheating. If ventilation is poor, you are preparing a hardware failure.
What I saw last summer: an SME with a server room at 32°C for 3 months. Then: a hard drive failure. A motherboard fried. “We don’t understand, the hardware isn’t that old.”
What to check:
- Room temperature: ideal 18–24°C. Above 27°C regularly? That’s a problem.
- Dust and filters: dusty servers = poor ventilation = overheating
- Air conditioning: if you have it — does it work? If it fails in summer, what do you do?
Nothing rocket science. Just a regular visual check.
5. Have Someone to Talk To
The real problem: it is not the technology. It is being alone with IT decisions.
What I often hear: “I don’t know whether to replace this server.” “I’m unsure about this software.” “I don’t know if my backups are correct.” And nobody to ask. So decisions get postponed. Until something breaks.
What actually helps:
- Someone who knows your infrastructure
- Someone you can call BEFORE making a decision
- Someone who anticipates with you
- Not someone you call in emergency when everything is already broken
Less about budget. More about mental load. Stop being alone with these questions.
Summary: 5 IT Essentials for a Smooth 2026
The 5 things to check in 2026:
- ✅ Test your backups (really)
- ✅ Check your power supply
- ✅ Clean up access rights
- ✅ Monitor server temperature
- ✅ Find someone to talk to regularly
Nothing revolutionary. Just common sense. And anticipation. That is what makes the difference between a quiet year and a year full of IT headaches.
Let’s discuss your situation.
30 minutes, no obligation.
Let’s take a look together at what it would take to ease your IT workload. No sales pitch. Just an honest assessment of the situation.
Your IT architect. Your trusted partner.