Monitors

A monitor runs a job on a schedule, compares each check with the last one, and tells you what changed: new jobs on a board, a price drop, new places on Google Maps, a change on a page. When nothing changed, you hear nothing. Open Monitors in the dashboard sidebar, or use the API.

Start a monitor

From a template: fill in the form and press Watch for changes.
From a preset: on the Presets page, open a preset’s menu and choose Watch for changes.
For pages: on the Monitors page, press New monitor and paste up to 50 URLs.

One panel opens with the schedule, what to alert on, and where to send alerts. It shows the cost per month before you press Start watching. The first check runs straight away. It records a starting point and doesn’t alert. Every check after that is compared with the one before.

What you can watch

Source	Best for
A template with its inputs	Lists with a stable key per row: places, products, jobs, listings, posts. New, changed and removed items are exact.
A preset	Any saved scrape, batch, map, search or template preset.
Pages (up to 50 URLs)	“Tell me when this page changes.”

A monitor can watch up to 10 sources and sends one alert per check for all of them.

What counts as a change

Alert me when	For	How
New items appear	New jobs, listings, places, posts	Rows that weren’t in the last check
A field changes	Prices, stock, ratings, follower counts	The same item, with a field you choose changed
Something is removed	Sold homes, closed jobs	Items that are gone
The page changes	Any page	Lines added to or removed from the page’s text
Every run	A daily export	The results, every time

Two optional filters cut the noise:

A rule on a field: is below, is at most, is above, is at least, equals, contains, drops by % or more, rises by % or more. For example “price drops by 10% or more”, or “rating is at least 4”.
Only if it matters: a goal in plain words, such as “only dental clinics with 4 or more stars”. When something changed, AI reads the changes and decides whether they match, with a one-line reason. It runs only on checks where something changed. If it can’t decide, the alert goes out anyway.

Schedule

Every 5, 15 or 30 minutes, every hour, every 6 or 12 hours, every day at a time, or every week on a day at a time, in your time zone. Over the API you can also give a cron expression. Check now runs one straight away. Your plan sets the shortest schedule and how many monitors you can have:

Plan	Shortest schedule	Monitors
Free	Daily	2
Starter	Hourly	10
Builder	Every 15 minutes	50
Pro	Every 5 minutes	200

Paused monitors count toward the limit.

Where alerts go

Email to your account’s address, on by default. One email per check, with the monitor’s name and a summary in the subject (“3 new”) and a table of what changed. You can add up to 10 more addresses. Each one gets a confirmation email first and receives nothing until it confirms, and every alert has a link to stop them.
Webhook: a POST to your https URL. See below.
Slack: an incoming webhook URL (https://hooks.slack.com/...).
Discord: a webhook URL (https://discord.com/api/webhooks/...).

Send a test sends a sample alert to the channels you picked before you save.

Webhooks

Each check that alerts sends a JSON body with the header X-Spidra-Event: monitor.check.completed. This is the one Send a test sends:

{
  "event": "monitor.check.completed",
  "monitor": { "id": "test", "name": "Test monitor", "alertOn": "new" },
  "check": { "id": "test", "summary": "Test notification", "url": "https://app.spidra.io/dashboard/monitors/test" },
  "judgment": null,
  "sources": [
    {
      "id": "test",
      "label": "Example",
      "added": [{ "title": "This is what a new item looks like", "url": "https://spidra.io" }],
      "changed": [{ "key": "x", "item": { "title": "And a changed one" }, "fields": [{ "field": "price", "previous": 30, "current": 25 }] }],
      "removed": [],
      "pages": []
    }
  ],
  "sentAt": "2026-10-24T18:50:30.112Z"
}

In a real alert, monitor and check hold your monitor’s and check’s IDs, check.url opens the check in the dashboard, and each source can also have rows (with “every run”) or error. added, changed, removed and pages have the same shape as in Get a Monitor Check. Every webhook is signed. The monitor has a secret (set your own, or read the generated one from Get a Monitor under notify.webhook.secret), and X-Spidra-Signature is sha256= followed by the HMAC-SHA256 of the raw request body with that secret. Check it before trusting the payload:

import crypto from "node:crypto";

function isFromSpidra(rawBody, signature, secret) {
  const expected = "sha256=" + crypto.createHmac("sha256", secret).update(rawBody).digest("hex");
  return signature && crypto.timingSafeEqual(Buffer.from(signature), Buffer.from(expected));
}

import hashlib, hmac

def is_from_spidra(raw_body: bytes, signature: str, secret: str) -> bool:
    expected = "sha256=" + hmac.new(secret.encode(), raw_body, hashlib.sha256).hexdigest()
    return hmac.compare_digest(signature or "", expected)

Use the raw body as received. Parsing and re-serializing the JSON changes it and the signature won’t match.

The monitor page

Each monitor has a timeline of checks, newest first: “3 new places”, “No change”, “Price dropped on 1 item”. Open a check to see the changed rows, with old and new values side by side. Check now, Pause, Edit and Delete are at the top.

What it costs

A check costs exactly what the same run costs: a template check costs what the template costs, a page costs 1 credit. The panel shows the credits per month before you save, and Estimate a Monitor gives it over the API. A Google Maps leads search checked daily is about 30 credits a month. The “only if it matters” goal adds a little AI on checks where something changed.

When you’re out of credits, checks are skipped, not failed, and the monitor stays on.
A check that is still running when the next one is due makes the next one skip.
After 5 failed checks in a row, the monitor is paused and you get an email, so a broken source doesn’t keep using credits.
Checks are kept for 30 days.

Good to know

A monitor sees what its source returns. If a site can’t be read in a check, that check reports the error and the next one tries again.
Page watching compares the page’s text. Busy pages with counters or rotating content change often; a goal helps filter that out.
Monitors aren’t real time. The fastest schedule depends on your plan.

Get started

Tools

Ready-made data

Getting the data you want

Supported sites

Guides

Dashboard

Account and billing

Start a monitor

What you can watch

What counts as a change

Schedule

Where alerts go

Webhooks

The monitor page

What it costs

Good to know

​Start a monitor

​What you can watch

​What counts as a change

​Schedule

​Where alerts go

​Webhooks

​The monitor page

​What it costs

​Good to know

Start a monitor

What you can watch

What counts as a change

Schedule

Where alerts go

Webhooks

The monitor page

What it costs

Good to know