Your Own Cloud and Server Monitoring: When a Business Should Stop Hoping for the Best

A company's own cloud and server monitoring usually appear after something unpleasant happens. An employee leaves — and the contracts stored in their personal Google Drive leave with them. The server running the website goes down on Saturday morning — and nobody finds out until Monday, when customers start calling.

Let's put numbers on it with an example. An online store takes about $1,000 in orders a day. The site is down from Saturday morning to Monday — that's roughly $2,000 in orders that never happened. And that doesn't count customers who went to a competitor for good, or ad spend that kept sending traffic to a dead site for two days. Setting up monitoring that would have flagged the problem within minutes costs a fraction of that.

Let's look at the typical gaps in a small company's infrastructure and how to close them without a huge investment.

Why this happens

Infrastructure in small businesses grows organically. Someone once created a shared folder in a personal account, the website was set up by a freelancer who's long gone, and backups are "probably running". While everything works, nobody thinks about it. The problem is that you learn about a failure at the worst possible moment — usually from your customers.

The risks I see most often

• Work files in employees' personal clouds. Price lists, contracts, site photos — in personal Google Drive or Dropbox accounts. The company has no control over who can access what.
• Lost access when someone leaves. The person is gone, only they knew the password, and the files are still in their account. Sometimes the departure isn't friendly.
• No backups. Or there are backups, but they sit on the same server as the data. A disk fails or ransomware encrypts the server — everything is gone at once.
• Customers notice downtime first. Nobody checks whether the website, CRM or phone system is up, whether the disk is filling up, or whether the SSL certificate has expired.

A corporate cloud on Nextcloud + S3

Nextcloud is an alternative to Google Drive or Dropbox that runs on your own server. Files are kept in S3 storage, which is easy to expand as your data grows.

What the company gets:

• Roles and access by department. Accounting sees its folders, sales sees theirs, management sees everything. A new employee gets access in a minute.
• Control when people leave. The account is disabled, and all files stay with the company.
• Two-factor authentication (2FA). A stolen password isn't enough to log in.
• SSL encryption for all connections.
• Sync with desktops and phones — employees work with files just as they would in Dropbox.
• Scheduled backups to a separate location.

How the cost compares with subscriptions

Below is an illustrative example only. Subscription and server prices vary between providers and change over time, so run the numbers for your own case. As an example, take a cloud service at about $12 per user per month versus your own cloud on a rented server.

Users	Subscription ~$12/user/month	Own cloud: server + S3 (~$50/month) + support ($100/month)
10	~$120/month	~$150/month
25	~$300/month	~$150/month
50	~$600/month	~$170–200/month (more storage)

Plus a one-off setup for your own cloud. What the example shows:

• With 10 users, a subscription is usually cheaper — and that's fine.
• From around 20–25 people, your own cloud starts to win on cost, and the gap grows with every new hire.
• Keep in mind: subscription suites often include email and office apps. Compare what you actually need.

But the main argument is often control rather than price: the data is on your server, access is managed centrally, and backups are set up the way you need.

Monitoring: Grafana + Prometheus + Telegram alerts

Monitoring is a system that watches your servers and services around the clock and messages you as soon as something goes wrong. I usually use Prometheus (collecting metrics), Grafana (clear dashboards) and alerts sent to Telegram.

What to monitor

• Availability of the website, CRM, APIs and phone system — checked every minute.
• Server resources — CPU, memory, disk space.
• SSL certificate expiry — a warning well in advance, not when browsers start showing customers an error.
• Databases — whether they're running and not growing abnormally.
• Backups — whether last night's backup actually completed.

What it looks like

A message arrives in your team's Telegram group: "Server disk is 90% full" or "Website not responding for 3 minutes". You or your technician react before customers notice. And Grafana shows how your servers behaved over a week or a month, so it's easy to see when it's time to scale up.

The 3-2-1 backup rule

The simplest and most reliable rule for backups:

• 3 copies of your data — the live one and two backups.
• 2 different media or storage types — for example, a server and S3.
• 1 copy offsite — in another data center or with another provider.

And a mandatory fourth point that people forget: test restores regularly. A backup you've never tried restoring from is a hope, not a backup.

Checklist: is your infrastructure OK?

Tick what you already have:

• Work files are stored in company storage, not in personal accounts.
• When an employee leaves, their access is revoked the same day using a clear procedure.
• Two-factor authentication is enabled for important systems.
• Backups run automatically and are stored in a different location.
• A restore from backup has been tested in the last 3 months.
• You find out about a website or server outage within minutes, not from customers.
• Servers, domains and passwords are registered to the company, not a former contractor.
• There's a written overview: which servers exist, what runs on them, who is responsible.

If three or more boxes are unticked, it's worth dealing with this before an incident happens.

What it looks like in practice

A corporate cloud on Nextcloud with S3. I deployed a cloud with SSL and 2FA, set up roles and access by department, backups, and sync with employees' desktops and phones. Read the case study.

A platform for a company with distributed sites and field teams. The system for tracking equipment, warehouse and on-site work runs on private servers that are reachable only through a VPN. Backup and restore are in place. Read the case study.

When you don't need this yet

• You have 3–5 people and little sensitive data — company accounts in a cloud service with 2FA are enough.
• You have one simple website on good hosting that already includes backups and uptime monitoring.
• Nobody will maintain the server and you're not ready to pay for support. A self-hosted cloud without maintenance is worse than a subscription.

How much it costs

Setting up a server, cloud or monitoring starts at $150, depending on scope. Support with monitoring, updates and backup checks starts at $100 a month.

Let's check your infrastructure

Not sure what state your backups are in, or who would be the first to know if the server went down? Book a free 30-minute review — we'll go through the checklist and I'll tell you what to fix first. More about the service: servers and support.