Website Change Monitor – Page & Price Change Alerts

Website change monitor for prices, stock, policies and docs: watch any public web page, or just the part of it that matters (a price, a stock label, a changelog), and get only what changed since the last run: new, changed and removed pages with the added and removed lines. Plain HTTP, no browser, so checks are fast and cheap.

Schedule it hourly or daily, connect Slack or e-mail, and you have change alerts without writing code.

Who uses it

• E-commerce and pricing teams: watch competitors' price and stock elements (.price, #availability) and get the old and new value.
• Legal, compliance and procurement: track terms of service, privacy policies, supplier pages and public notices.
• Developers and product managers: follow changelogs, release pages, API docs and status pages.
• Marketers and SEO: see when a competitor changes a landing page, a plan or a headline.

How to use it

1. Put the pages in Page URLs, one per line. Optionally add a CSS selector to watch only part of each page (e.g. .price), or use Pages with their own selector for different selectors per page.
2. Run it once. The first run saves every page as new (your baseline) in a memory store in your Apify account.
3. Schedule it (hourly, daily, weekly). Every later run returns only pages that changed or disappeared, with the lines that were added and removed.

Input example

{
  "urls": ["https://www.python.org/downloads/", "https://example.com/terms"],
  "pages": [
    {"url": "https://shop.example.com/product/123", "selector": ".price", "label": "Competitor price"},
    {"url": "https://shop.example.com/product/123", "selector": "#stock", "label": "Competitor stock"}
  ],
  "ignorePatterns": ["Last updated .*", "\\d+ people are viewing"],
  "includeUnchanged": false,
  "stateStoreName": "my-competitor-watchlist"
}

Use one Memory store name per watchlist: each store remembers the last version of its pages, so two different schedules never mix their baselines.

Output example

{
  "url": "https://shop.example.com/product/123",
  "finalUrl": null,
  "label": "Competitor price",
  "selector": ".price",
  "changeType": "changed",
  "checkedAt": "2026-10-07T18:00:00Z",
  "previousCheckedAt": "2026-10-06T18:00:00Z",
  "httpStatus": 200,
  "title": "Acme Pro – Acme Store",
  "textLength": 11,
  "hash": "5f1c0d0e8a4b0c3f9e2d7a1b6c4e8f20",
  "previousHash": "a3b9e0f1c2d4e5f60718293a4b5c6d7e",
  "addedCount": 1,
  "removedCount": 1,
  "addedLines": ["$35 / month"],
  "removedLines": ["$29 / month"],
  "text": "$35 / month"
}

Field	Type	Description
url	string	The page you asked to check
finalUrl	string / null	Where the page ended up after redirects (null when it is the same URL)
label	string / null	Your label from Pages with their own selector
selector	string / null	CSS selector used (null = whole page)
changeType	string	new (first check in this memory store), changed, removed (page now 404/410), unchanged (only with Also output unchanged pages)
checkedAt, previousCheckedAt	string	This check and the version it was compared with (ISO 8601 UTC)
httpStatus	integer	HTTP status of the page
title	string / null	Page <title>
textLength	integer	Characters of normalised text compared
hash, previousHash	string / null	Fingerprints of the compared text
addedCount, removedCount	integer	Lines added and removed since the last check
addedLines, removedLines	array	The lines themselves (up to Max added/removed lines)
text	string / null	Current text, first 5,000 characters, e-mail addresses and phone numbers removed

Pages that could not be checked are listed in the run's key-value store as ERRORS (with the reason: ROBOTS_DISALLOWED, BLOCKED, FETCH_FAILED, NOT_FOUND, SELECTOR_NOT_FOUND, EMPTY_PAGE, UNSUPPORTED_CONTENT, REFUSED), and a run summary as SUMMARY.

Use it with

• Schedules: save your input as a Task and run it every hour, day or week; each run compares with the previous one.
• Alerts: the Integrations tab sends results to Slack, e-mail, Google Sheets or a webhook. With the default settings a run only outputs changed pages, so an empty run means "nothing changed".
• Make, Zapier and n8n: start a workflow when a run finishes and read the changed pages.
• API and code: the Apify API, the JavaScript and Python clients, or CSV/JSON/Excel downloads.
• AI agents: the Apify MCP server exposes this Actor as a tool; the input is a list of URLs.

Pricing

Pay per page check: $1.00 per 1,000 page checks on the Apify Free plan; on paid plans $1.00 (Starter), $0.90 (Scale) and $0.80 (Business and higher) (the price for your plan is shown on the Pricing tab). The examples below use the Free-plan price. One check = one page (or one page + selector) fetched and compared, whether or not it changed. Pages that are blocked, disallowed by robots.txt, not found on the first check or fail are never charged.

• 20 competitor pages checked daily → 600 checks/month → about $1.00/month or less.
• 200 pages checked hourly → about 144,000 checks/month → about $144.00/month.
• 2,000 pages checked once a week → about 8,600 checks/month → about $9.00/month.

A small start fee of $0.00005 per run applies. Set a maximum cost per run in the run options and the Actor stops cleanly when it is reached; pages it could not get to keep their old version and are compared next run.

Limits and notes

• Plain HTTP only, no browser. Pages that build their content with JavaScript may show little or no text (reported as EMPTY_PAGE, not charged). Many shops and docs render server-side and work fine.
• robots.txt is respected. Before checking a page the Actor reads the site's robots.txt (RFC 9309, including * and $ rules) and skips pages it disallows for crawlers (ROBOTS_DISALLOWED, not charged). If robots.txt itself can't be read (server error), the site is skipped to be safe.
• No bot-check bypass, no proxies, no logins. A Cloudflare/DataDome-style challenge, a 401/403 or a persistent 429 is reported as BLOCKED and not charged. There are no cookie, login or header inputs, and private or internal addresses are refused.
• Politeness: at most one request per second per website, five websites in parallel, pages up to 5 MB.
• Up to 1,000 pages per run. Text is compared line by line after whitespace is normalised; use Ignore text matching for dates, counters or rotating banners so they don't trigger false changes.
• E-mail addresses and phone numbers are removed from all output text.

FAQ

Is it legal to monitor websites? The Actor only reads public pages that the site's robots.txt allows for crawlers, at a polite rate, without logging in or getting around any protection, and it removes contact details from its output. Whether you may use a site's content depends on that site's terms and your purpose; check them for the pages you add. If in doubt, ask a lawyer.

Why is a page changed when nothing visible changed? Something in its text changed: often a date, a counter or a rotating promo. Add an ignore pattern for it, or watch only the relevant part with a CSS selector.

Why did a page show EMPTY_PAGE or SELECTOR_NOT_FOUND? The content is probably loaded by JavaScript after the page opens. Try a selector on server-rendered parts, or the site's RSS feed or sitemap.

Can I monitor a page behind a login? No. The Actor only checks public pages and never takes passwords or cookies.

How do I get alerts only when something changes? Keep Also output unchanged pages off and add a Slack or e-mail integration to your scheduled Task; runs with no changes produce no rows.

Changelog

See CHANGELOG.md.

Support

Found a problem or need a feature? Open an issue on the Actor's Issues tab. We read every one.