# How StatusTick works

> The core ideas in StatusTick: monitors, regions, checks, incidents, alert channels and status pages, and how they connect.

StatusTick watches your websites, APIs, servers and scheduled jobs, opens an incident when one of them really fails, and tells the right people. This page explains the building blocks.

## Organizations and members

Everything in StatusTick belongs to an **organization**. People join an organization by invite and have one of three roles:

| Role | Can do |
| --- | --- |
| Owner | Everything, including member management and deleting the organization |
| Admin | Create and change monitors, alert channels, incident settings and status pages |
| Member | See monitors, incidents and reports, and work on incidents |

Changes to monitors, channels and members are recorded in the organization's audit log.

## Monitors

A **monitor** is one thing StatusTick checks: a URL, a host and port, a DNS name, or a scheduled job that reports in. Each monitor has a type, a target, an interval (every 30 seconds to every 24 hours) and the regions to check from. See [Monitor types](https://statustick.com/docs/monitor-types).

## Regions, private agents and checks

On every interval, StatusTick runs a **check** from each region you chose, at the same time. Regions are public StatusTick locations or **private locations**: agents you run inside your own network to check internal services. Each region reports its own result: up, degraded, blocked or down, with the response time and the error if there was one.

The regions are combined into one monitor status. By default, a monitor is only **down** when at least two regions fail. A single failing region is re-checked at once. See [Avoiding false alarms](https://statustick.com/docs/avoiding-false-alarms).

Every check result is kept for 14 days. Each hour, StatusTick rolls the results up into hourly uptime and average, p95 and p99 response times, and keeps that history for 13 months.

## Incidents

When a monitor goes down and stays down for the number of failed checks you set, StatusTick opens an **incident**. When the monitor recovers for the number of good checks you set, the incident resolves on its own.

On an incident, your team can acknowledge it, assign it, change its severity (Critical, High, Medium, Low) and add comments. You can also open an incident by hand for problems a monitor cannot see.

## Alert channels

An **alert channel** is where StatusTick sends incident alerts: email, Slack, Microsoft Teams, Discord, Telegram, PagerDuty or your own webhook. See [Alert channels](https://statustick.com/docs/alert-channels).

## Status pages

A **status page** shows your users the state of your services. It is made of components, and each component is linked to one or more monitors. When the monitors disagree with a component, StatusTick suggests the change and a person publishes it. Visitors see the overall status, 90 days of uptime per component and your incident posts. StatusTick drafts each post from the incident, and pages can use your own domain and email subscribers.