Website Monitoring Tool
How we built an uptime and performance monitoring SaaS — monitoring 500+ websites, sending instant downtime alerts, and generating performance reports with historical trend charts.
01 — The Problem
A web agency with no way to monitor its clients' sites
A web agency managing 200+ client websites was finding out about downtime from their clients — not their own systems. They had no monitoring infrastructure and were paying for enterprise tools that were overkill for their needs and budget.
They wanted to build their own monitoring SaaS that they could offer to clients as an add-on service — a product that would generate recurring revenue while solving a real operational problem.
"We found out our biggest client's site had been down for 4 hours when they called us angry. That can't happen. We needed to know before they did."
— Agency Founder
- No automated uptime monitoring — downtime discovered by clients, not the agency
- Enterprise monitoring tools too expensive for small website portfolios
- No response time tracking — slow sites not detected until complaints
- No historical performance data — impossible to spot degrading trends
- No white-label reporting to share with clients
02 — The Solution
A purpose-built monitoring SaaS with multi-tenant support
- Uptime Monitoring Engine
Queue-based monitoring checks every registered site at configurable intervals (1–15 minutes) — recording response status, code, and latency on every check. - Instant Downtime Alerts
When a site fails two consecutive checks (within 60 seconds), an alert fires via email and SMS — with the downtime start time, error code, and a direct link to the status page. - Performance History Charts
Interactive response time charts (Chart.js) showing 24-hour, 7-day, and 30-day trends — making it easy to spot when a site started slowing down before it became a problem. - Multi-Site Dashboard
All monitored sites in one view — colour-coded status (up/down/degraded), current uptime percentage, last check time, and average response time. - White-Label Client Reports
Monthly PDF uptime reports branded with the agency logo — shareable with clients as a value-add that justifies retainer relationships.
03 — Tech Stack
Laravel + Queue Workers + Chart.js + MySQL
Architecture note: The monitoring engine runs as queue workers separate from the web application — ensuring checks continue even if the web layer receives heavy traffic. Workers are horizontally scalable.
04 — Timeline
8 weeks from concept to SaaS product
Week 1–2
Monitoring Architecture Design
Designed the queue-based check engine, data model for check history, and alert logic (consecutive failure threshold).
Week 3–5
Core Engine + Dashboard
Queue workers, monitoring engine, site registration, multi-site dashboard, and real-time status updates.
Week 6–7
Alerts, Charts & Reports
Alert delivery (email/SMS), Chart.js performance history, and PDF monthly report generation.
Week 8
Load Test & Launch
Simulated 1,000 simultaneous site checks. Queue scaling validated. Launched with first 50 client sites.
05 — Results