How I built a monitoring SaaS for Joomla, WordPress & PrestaShop agencies

#laravel #php #saas #webdev

After years managing dozens of Joomla, WordPress and PrestaShop sites
for clients, I kept running into the same problem: I was always the
last one to know when something broke.

A client would call me to say their site was down. Or I'd discover
during a routine check that an SSL certificate had expired three days
ago. Or a backup hadn't run in two weeks.

So I built SiteOps — a professional monitoring SaaS specifically
designed for CMS sites and web agencies.

The problem with existing monitoring tools

Most monitoring tools (UptimeRobot, Pingdom, etc.) do one thing well:
uptime. You get a ping every 5 minutes and an email if the site goes
down.

But managing client sites means you need much more:

Is the SSL certificate about to expire?
Did the Akeeba or UpdraftPlus backup actually run last night?
Is the Joomla or WordPress version up to date?
Are there brute-force attacks being blocked?
What's the current PHP version on the server?
Is the domain registration about to expire?

None of the general-purpose tools covered all of this in one place.

What SiteOps monitors

Here's what the platform checks for each site:

Availability

Uptime every 30 seconds from 8 worldwide locations
Response time tracking
Multi-location availability map

Infrastructure

SSL certificate validity and expiry alerts
DNS records (A, MX, TXT, NS)
WHOIS / domain expiry
IP reputation and DNSBL checks
TCP port monitoring

Security

HTTP security headers (HSTS, CSP, X-Frame-Options...)
Google Safe Browsing status
Brute-force detection and blocking status
File integrity monitoring

Performance & SEO

Google PageSpeed score (mobile + desktop)
Core Web Vitals
SEO meta tags analysis

CMS-specific (via native agents)

CMS version (Joomla 4/5/6, WordPress, PrestaShop)
Plugin/extension/module updates available
Backup status (Akeeba for Joomla, UpdraftPlus for WordPress)
PHP and MySQL versions
Server load and memory usage

The native agent approach

The key differentiator is the native agent architecture. Instead of
just pinging a URL from outside, SiteOps installs a lightweight agent
directly on the CMS:

Joomla: system plugin (available on JED, GPL licensed)
WordPress: plugin (available on WordPress.org)
PrestaShop: module (available on PrestaShop Addons)

The agent runs on every page load and pushes metrics to SiteOps via
a secure API. This gives access to data that external monitoring simply
cannot see — backup status, plugin versions, server metrics.

The tech stack

Backend: Laravel 13 / PHP 8.2
Admin panel: Filament v4
Database: PostgreSQL 16
Infrastructure: European VM (GDPR compliant)
Uptime checks: Cloudflare Workers at 8 worldwide locations
Alerts: Email, SMS, Slack, Telegram, Discord, Teams, Webhook

Multi-tenant architecture for agencies

SiteOps is built from the ground up for agencies managing multiple
client sites. Each tenant has their own isolated dashboard with:

All their sites in one view
Per-site alert configuration
White-label option (Agency plan)
Client-facing status pages

What I learned building this

1. PostgreSQL strict GROUP BY is your friend.
Coming from MySQL, the strictness catches real bugs. Embrace it.

2. Observers over lifecycle hooks.
Laravel's afterCreate() in Filament isn't always reliable. Model
Observers are more robust for post-creation side effects.

3. Cloudflare Workers are excellent for distributed uptime checks.
8 worldwide locations, sub-second latency, virtually free at low
volume. Highly recommended for distributed monitoring.

4. CMS agents need to be lightweight.
The Joomla plugin runs on every page load — it needs to be fast.
Async API calls and aggressive caching were essential.