Change Detection Software: Types, How It Works and How to Choose

"Change detection software" sounds like one product category. It isn't. Search the phrase and you'll find tools that watch competitor pricing pages, tools that compare screenshots in a CI pipeline, security agents that watch system files, database connectors and satellite imagery platforms. They share one idea and almost nothing else.

This guide sorts that out. It explains what change detection software does, the five types you'll run into, how website change detection works under the hood, and the features that decide whether a tool is useful or just noisy. If you already know you want to watch websites, skip to how website change detection software works.

What is change detection software?#

Change detection software saves a known state of something, checks it again later, and reports the differences. That "something" can be a web page, a screenshot, a folder of system files, a database table or a satellite image.

Every tool in the category does the same four things:

1. Capture the current state.
2. Compare it with a reference state.
3. Decide which differences matter.
4. Report them, usually as an alert or a report.

Steps 1, 2 and 4 are the easy part. Step 3 is where tools differ, and it's the step that decides whether people keep reading the alerts or quietly filter them into a folder.

The 5 types of change detection software#

Before comparing products, make sure you're comparing the right kind.

The rest of this guide is about the first two, because they're about websites, and they're often confused with each other.

Website change detection vs visual regression testing#

They both compare "before" and "after", but they run at different points:

• Visual regression testing runs before release, in your build pipeline, on code you control. It answers: "Did my pull request change how the app looks?" See our visual regression testing guide.
• Website change detection runs after release, on a schedule, against the live site. It answers: "Did anything change on the live site since I last approved it?"

The second catches what the first can't: a plugin auto-update, a CMS edit by a client, a tag manager change, a CDN rule, a hosting migration, an expired script. None of those go through your test suite.

How website change detection software works#

Good website change detection follows the same pipeline, whatever the brand:

1. Fetch the page. Simple tools download the raw HTML. Better ones load the page in a real browser (headless Chrome) so JavaScript-rendered content, scripts and layout are captured as visitors see them.
2. Stabilize it. Wait for the page to settle, pause animations, wait for fonts and stop waiting after a sensible limit. Skipping this step is the main source of false alerts.
3. Extract what matters. Visible text, title, meta description, canonical, robots meta, headings, links, scripts, HTTP status, page weight and a screenshot.
4. Fingerprint it. Hash each part so unchanged parts can be skipped quickly.
5. Compare with the baseline. Text against text, tags against tags, screenshot against screenshot.
6. Filter noise. Drop differences inside ignored areas, below thresholds or caused by known volatile content.
7. Rate severity and alert. A page returning 404 is not the same as a changed footer year.

We go deeper into each step, including how text and pixel diffs are calculated, in web page change detection: how it works.

Baseline vs previous check#

This one design choice matters more than most feature lists.

Many tools compare each check with the previous check. If something changes on Monday and you miss the email, Tuesday's check compares against Monday's already changed page and reports nothing. The change is gone from view.

Better tools compare against an approved baseline: the version you said was correct. A change keeps showing up until you either fix it or approve it as the new normal. That's how you get a monitoring history you can trust, not a stream of one-off emails.

What to look for in change detection software#

Use this as a checklist when you trial a tool.

1. What it actually compares#

Ask what each check captures. Some tools only diff visible text. Some only compare screenshots. For websites you usually want several layers, because each catches different problems:

• Text catches content edits and removed sections.
• Screenshots catch layout breaks, missing images and CSS failures that leave the text intact.
• SEO tags catch title, canonical and noindex changes that are invisible on screen.
• HTTP status and redirects catch pages that broke or moved.
• Scripts catch removed analytics or an unknown third-party script.
• Specific elements catch a missing "Add to cart" or signup button.

2. Noise control#

Ask how the tool handles dates, rotating banners, cookie notices, live chat widgets and ads. Look for:

• Ignored selectors or areas so you can exclude a carousel or a timestamp.
• Visual thresholds so a 0.3% pixel shift doesn't page anyone.
• Stabilization (waiting for the page to finish loading, disabling animations).
• Grouped alerts, one email per scan, not one per change.

If a trial produces more than a handful of alerts a week on a quiet site, that's a sign it will be ignored within a month.

3. Severity, not just "changed"#

"Something changed on /pricing" doesn't help much. "The pricing page returned 500" or "the page changed from index to noindex" tells you what to do. Prefer tools that rate changes and let you choose which levels send email.

4. Before and after you can read#

When an alert arrives, you should be able to see the old value next to the new one, and the old screenshot next to the new one, without digging. Diff views are the product.

5. JavaScript rendering#

If your site uses React, Vue, Shopify themes, Webflow interactions or a tag manager, a tool that only reads raw HTML will miss changes or report confusing ones. Check whether it uses a real browser.

6. Frequency and limits#

Check how often pages are checked on each plan, how many pages are included, and how long history is kept. Hourly checks on thousands of pages are expensive to run, so the cheapest plans usually trade frequency for page count.

7. Where alerts go#

Email is standard. Slack, webhooks and a dashboard for many sites matter more once you manage client websites.

Which kind of tool do you need?#

Pick by the question you're trying to answer:

For a side-by-side of the website tools, read our list of the best website change detection tools.

Change detection for compliance: payment page scripts#

There's one place where change detection is not optional. PCI DSS version 4.0 added requirements for e-commerce payment pages: keep an inventory of the scripts that run on them and justify each one (requirement 6.4.3), and run a change and tamper detection mechanism that alerts on unauthorized changes to those pages' scripts and security headers (requirement 11.6.1). Both became mandatory on March 31, 2025.

The reason is skimming attacks, where a malicious script is added to a checkout page to copy card numbers as they're typed. The page looks and works exactly the same. Only a script inventory compared over time shows it.

Website change detection software that tracks scripts per page, flags new third-party domains and alerts on them covers the monitoring side of this. You can see which scripts a page loads right now with our free script detector.

Free vs paid change detection software#

Free options are good for a few pages:

• changedetection.io is open source and can be self-hosted, which suits developers happy to run a server.
• Browser extensions such as Distill.io's run checks from your own browser while it's open.
• Free plans of hosted tools usually cover a handful of pages checked daily or weekly.
• MyKavo's free plan covers 1 website with 5 monitored pages checked weekly, and our website change detector runs a one-off check with no account.

Paid plans make sense when you need more pages, daily checks, history, or checks that run in a real browser at scale. Rendering a page, taking a full-page screenshot and storing it costs real computing time, which is why browser-based checks are rarely unlimited.

How to get started without drowning in alerts#

1. Start with the pages that make money or traffic. Home, pricing, signup, checkout, top landing pages. Five well chosen pages beat five hundred random ones.
2. Run a baseline and look at it. Make sure the screenshot and extracted values are right before you trust comparisons against it.
3. Exclude the noisy bits on day one. Cookie banners, rotating testimonials, "last updated" dates.
4. Only email yourself high and critical changes at first. Review the rest in the dashboard. Loosen later.
5. Approve expected changes promptly. After a planned release, approve the new baseline so the next scan compares against the right version.

For a step by step walkthrough of the methods, from manual checks to automated monitoring, see how to monitor website changes.

Summary#

Change detection software is a family, not a single product. Website change detection watches live pages after release; visual regression testing checks code before release; file integrity monitoring, CDC and remote sensing tools solve unrelated problems with the same idea. For websites, judge tools on what they compare, how they control noise, whether they compare against an approved baseline and how clearly they show before and after. Those decide whether you'll still trust the alerts in three months.