What kinds of sources can OnChange monitor?

OnChange supports web pages, selected HTML regions, browser-rendered visual pages, JSON APIs, sitemaps and robots.txt files. The monitor type determines whether the resulting evidence is a text or structural diff, screenshot comparison, JSON patch or purpose-built crawl-control change.

Can OnChange monitor websites that require JavaScript?

Yes. Pro, Business and Enterprise plans can render a page in Chromium before capture. You can use a bounded wait or wait for a known selector when the important content arrives after the initial HTML response.

Does every detected difference send a notification?

No. The change record is created first. Selectors, normalization patterns, thresholds, confidence, keyword or natural-language rules and optional semantic-only mode then decide whether the event should notify, remain quiet in history or trigger an automation.

Which OnChange feature should I start with?

Start from the consequence. Use HTML monitoring when wording or metadata matters, visual monitoring when appearance matters, API monitoring for structured responses, accessibility monitoring for WCAG regressions and a scoping report when the complete public-site inventory must be established first.

Website change monitoring features

Follow the change from source to decision.

Detect the surface that moved, preserve the evidence needed to review it, and route only the signal that deserves a person or automation.

Start with five monitors Explore the demo

Change evidence pipeline

One source · three decisions

Monitoring

Detect

HTML + text
Visual
API + SEO

Explain

Exact diff
Confidence
History

Route

Email + Slack
Discord
Webhook

Evidence is retained even when an alert rule keeps the event quiet.

Monitoring has three separate jobs.

Treating detection, evidence and notification as one switch is how harmless churn becomes alert fatigue and important context disappears.

Capture the right representation

Text, DOM structure, pixels, JSON fields, sitemap membership and WCAG findings answer different questions. The source should decide the evidence type.

Explain consequence without hiding the diff

Classification, confidence and AI context help triage, while the timestamped before-and-after record remains available for verification.

Notify according to ownership

The same platform event can stay quiet, reach a reviewer, enter an incident channel or trigger a webhook based on the policy attached to that monitor.

Detect the right surface

Choose the evidence type that matches the risk instead of forcing every source through one generic page hash.

HTML change detection

Compare visible text, structure, sections and head metadata, with optional JavaScript rendering.

Text + structure

Visual difference monitoring

Capture rendered screenshots and review the pixels that moved between page states.

Rendered pixels

API and JSON monitoring

Track status, response structure and RFC 6902-shaped field changes without reading raw payloads.

JSON operations

SEO change monitoring

Extract titles, canonicals, robots directives, sitemap membership and structured-data drift.

Search signals

Preserve what happened

Keep the state and provenance needed for a reviewer to understand a change after the live page has moved on.

Webpage version history

Return to 90 days of retained change states with HTML, CSS, screenshot and delivery context.

Time Travel

Website snapshot comparison

Compare two whole-site evidence sets across pages, templates, HTML, CSS, visuals and WCAG results.

Whole-site delta

Website scoping reports

Freeze a coverage-qualified public-page inventory with capture outcomes and template families.

Complete scope

Accessibility regression evidence

Separate new, persistent and resolved axe findings against an attested baseline.

WCAG delta

Route the useful signal

Attach consequence, ownership and delivery state so the detected change can enter a real operating workflow.

Choose the first monitor by consequence.

The most valuable starting page is rarely the one that changes most often. It is the one where an unnoticed change creates cost, risk or customer confusion.

RevenuePricing, checkout, stock and offer pages SearchTemplates, canonicals, sitemaps and robots.txt TrustTerms, privacy, accessibility and public guidance OperationsAPIs, status surfaces and production entry points

Product questions

Match the feature to the evidence you need

What kinds of sources can OnChange monitor?: OnChange supports web pages, selected HTML regions, browser-rendered visual pages, JSON APIs, sitemaps and robots.txt files. The monitor type determines whether the resulting evidence is a text or structural diff, screenshot comparison, JSON patch or purpose-built crawl-control change.
Can OnChange monitor websites that require JavaScript?: Yes. Pro, Business and Enterprise plans can render a page in Chromium before capture. You can use a bounded wait or wait for a known selector when the important content arrives after the initial HTML response.
Does every detected difference send a notification?: No. The change record is created first. Selectors, normalization patterns, thresholds, confidence, keyword or natural-language rules and optional semantic-only mode then decide whether the event should notify, remain quiet in history or trigger an automation.
Which OnChange feature should I start with?: Start from the consequence. Use HTML monitoring when wording or metadata matters, visual monitoring when appearance matters, API monitoring for structured responses, accessibility monitoring for WCAG regressions and a scoping report when the complete public-site inventory must be established first.

Start with one source and keep the evidence connected.

Five monitors are free. No credit card required.

Create the first monitor