WEBSITE
Is the site serving pages?
Records up or down, the response time in milliseconds, the server's answer code, and whether the text you expect actually appeared on the page.
200 + page text
Know the moment
your site
goes down
WatchIt checks your sites and services from the outside, the way a real visitor reaches them — then alerts Email, Slack, or Teams the moment one goes down, and again when it recovers. Nothing to install on your servers.
Running right now in the live demo
10 sites monitored
~2,800 checks per site per day
Checks every 10s
Probes from multiple regions
Response time tracked per check
Uptime over 24h, 7d and 30d
Downtime caught in one interval
Latency history kept per region
Status recomputed on every probe
4 check types
Website, ping, port and SSL
3 alert channels
Email, Slack and Teams
Incidents opened automatically
Maintenance windows mute alerts
Alert rules by monitor or tag
Keyword checks on page bodies
Port checks on any host
Recovery alerts
Certificate expiry warnings
189 automated tests
Incident timeline per monitor
API keys for automation
Paused monitors stay quiet
Every timestamp in UTC
Acknowledge incidents in a click
Tags group monitors in bulk
PIPELINE
From a failed check to the right person, without waking anyone unnecessarily
A scheduler ticks every 10 seconds and hands each due site to a checker. Every result is recorded, then run through the incident engine: one failed check only increments a counter, and it takes several consecutive failures before a site is declared down — so a single network hiccup never pages anyone at 3 a.m. A site that answers but answers too slowly is marked degraded instead: visible, not an outage. Recovery closes the incident and reports how long it lasted. Alerts are only sent once the result is safely stored, so what your team reads in Slack always matches the timeline.
001
0s
Between scheduler ticks — the fastest check interval
002
0
Check types: website, ping, port, SSL certificate
003
0
Checks per site per day in the live demo
004
0
Automated tests across the monitoring backend
About WatchIt
What is WatchIt?
An outside-in uptime monitor. You give us an address and how often to check it; we visit it on schedule from our own infrastructure, record how fast and healthy the answer was, and open an incident the moment it stops answering properly. Setup takes under a minute per site.
Why it matters?
There is no agent to deploy, no port to open, and no access to grant. WatchIt never installs anything on your servers and never sees your code, your databases, or your end users’ data — it visits your public address exactly like any other visitor. What it does keep is the record: every check, every incident, every alert that went out, so post-mortems start with facts instead of screenshots.
See it running on the demoEvery check records a small, focused set of health facts — up or down, how long the answer took, and what the answer said. From that raw data WatchIt derives the story: uptime percentages, speed trends, and incidents opened and closed on your behalf.
Discover checksPING
Is the host reachable?
The check to reach for when there is no web page to ask — records reachability and round-trip time to the host itself.
ICMP round-trip
PORT
Is the service accepting connections?
Points at a database, mail server, queue, or anything else behind a port, and records whether the connection succeeded and how long it took.
TCP connect
CERTIFICATE
Is the certificate still valid?
Records the issuer, the expiry date, and the days remaining — and warns you while there is still time to renew.
14d early warning
1
Give us the address
A website, a server, or a service — shop.example.com is enough. That is the only thing we strictly need from you.
2
Choose the check
Website, ping, port, or certificate — and how often to run it. Optionally define what healthy means: expected text, or a speed limit in milliseconds.
3
Point the alerts
Email, a Slack channel, or Microsoft Teams — for one site, a group of sites, or everything. Add planned maintenance windows if you have them.
4
Read the record
Uptime percentages, speed trends, and incidents — “down for 14 minutes on Tuesday, cause: timeout” — for every site you watch.
( The feature set )
Every capability here is built and running today — no add-ons, no per-feature tiers, nothing to bolt on later.
Checks
Website checks
Fetches the page on your interval and records the status code and how long it took.
Checks
Ping checks
For hosts with no web page to ask, reachability and round-trip time to the machine itself.
Checks
Port checks
Opens a TCP connection to a database, queue, or mail server and times the handshake.
Checks
Certificate checks
Tracks issuer, expiry, and days remaining, so renewal lands before the browser warning.
Checks
Expected page text
A 200 that renders an error page is still down. Name the text that proves the page is real.
Checks
Speed thresholds
Set the millisecond budget above which a response counts as degraded, not merely slow.
Alerts
Email alerts
The channel that needs no setup, delivered the moment a check confirms a change of state.
Alerts
Slack alerts
Posts into the channel the team already watches, with the failing target and the reason.
Alerts
Microsoft Teams
The same notification into Teams, for estates where that is where the on-call actually lives.
Alerts
Recovery alerts
The all-clear goes out on the same channels, so nobody chases something that is already back.
Incidents
Degraded state
Up but over its speed budget is its own state, visible before it becomes an outage.
Incidents
Maintenance windows
Planned work is declared up front, so a deliberate restart never pages anyone at 3am.
Incidents
Incident timeline
Every incident keeps its start, end, duration, and cause. Down 14 minutes on Tuesday, timeout.
Incidents
Acknowledge
One person takes an incident, and the rest of the team can see it is already being handled.
Reports
Uptime %
The figure over any window you pick, computed from the checks themselves rather than estimated.
Reports
Response trends
Response time plotted over time, so the slow drift that precedes an outage is visible early.
API
API keys
Long-lived keys for scripts and CI, separate from your login. Revoke one and it dies instantly.
API
REST API
Every dashboard action is an endpoint, with a live OpenAPI schema and a Postman collection.
( The API )
Everything the dashboard does, a script can do too — so monitoring lives in the same pipeline as the thing it watches.
01
Every dashboard action is an API call
Sites, incidents, alert channels, rules, maintenance windows, and tags — all scriptable, nothing dashboard-only.
02
Two ways to authenticate
A JWT bearer token for dashboards, or a long-lived X-API-Key for scripts and CI. Revoke a key and it stops working immediately.
03
Ownership is never leaked
Every resource is scoped to you. Someone else's monitor returns 404, not 403 — existence isn't revealed.
04
Specs, not screenshots
A live OpenAPI schema, Swagger UI, ReDoc, and an importable Postman collection ship with the service.
Overview
Every site at a glance — how many are up, down, or slow, plus the latest incidents. Refreshes every 30 seconds.
Site detail
Response times, an uptime timeline over 24h, 7d or 30d, the key numbers, and a Test now button.
Incidents
When it started, how long it lasted, what caused it — and who acknowledged it.
Swagger UI
ReDoc
openapi.json
Postman
curl
API guide
Set up your first site in under a minute — or log in to the live demo, where 10 real sites have been checked continuously for weeks. Login: demo@watchit.dev / demo1234