Deep Scan explained: why WCAGScan tests your pages in a real browser
A plain-English guide to Deep Scan, the Chromium engine behind it, and the accessibility problems it finds that a simple code check cannot.
The short version
| Deep Scan | WCAGScan’s scanning method. It opens each page in a real web browser, waits for the page to build itself, and then tests what a visitor would actually get |
|---|---|
| Chromium | The open-source browser engine that Google Chrome and Microsoft Edge are built on. Deep Scan runs Chromium on our servers with no screen attached, which is called a headless browser |
| Why it matters | Most modern pages are assembled by JavaScript and styled by CSS after they load. A scanner that only reads the raw file misses much of what is on the page |
| What it adds | Content built by JavaScript, real color contrast, visible keyboard focus, and layout at phone width |
| What it does not do | Replace testing by a person. No automated scan can judge everything in WCAG |
Two ways to scan a web page
Every web page starts as a file of HTML code sent by the server. The simplest kind of scanner downloads that file and reads it, the way you might proofread a script without watching the play. This is fast, and it catches real problems such as images with no text alternative or a page with no title.
But the file is only the starting point. Your browser then runs the page’s JavaScript, applies its style sheets, loads fonts, and builds the finished page you see. On many sites the menus, calendars, staff directories, product lists and forms do not exist in the original file at all. They are added afterwards by JavaScript.
Deep Scan watches the play. It loads each page in a real browser, lets it finish building, and tests the finished result.
What is Chromium?
Chromium is a free, open-source web browser project. Google Chrome and Microsoft Edge are both built on it, along with several other browsers. When Deep Scan renders your page in Chromium, it is using the same engine a large share of your visitors use, so the page it tests is very close to the page they see.
Our servers run Chromium in “headless” mode. That simply means the browser runs without a window or a monitor. It still loads the page, runs the scripts and calculates every color and position exactly as it would on a desktop.
What Deep Scan finds that a code check cannot
1. Content that JavaScript builds
Sites built with tools such as React, Vue or Angular, and many school, government and shop platforms, send a nearly empty file and fill it in with JavaScript. A scanner that reads only the file reports a clean result because it sees almost nothing. Deep Scan sees the menus, buttons, forms and text after they appear, and tests those.
2. Real color contrast
WCAG requires enough contrast between text and its background (success criterion 1.4.3). The colors are usually set in style sheets, inherited from other elements, or layered on top of each other. You cannot work them out reliably from the HTML file. Deep Scan asks the browser what color each piece of text really is and what is really painted behind it, then calculates the contrast ratio from those values. The report shows both colors and the ratio.
3. Visible keyboard focus
People who cannot use a mouse move through a page with the Tab key and need to see where they are (success criterion 2.4.7). Deep Scan presses Tab through the page with real keyboard events and checks whether each element shows a visible focus indicator when it receives focus. This is impossible to test without a working browser.
4. Layout at phone width
People with low vision often zoom in until the page is as narrow as a phone screen. WCAG asks that content still fits without scrolling sideways (success criterion 1.4.10, Reflow). Deep Scan resizes the browser to 320 pixels wide and checks whether the page overflows.
5. Pages behind a login
Because Deep Scan uses a real browser, it can sign in to a members-only or staff-only area with a test account you provide and scan the pages behind it. See pricing for which plans include this.
What this means for your report
- Fewer missed problems. Issues inside JavaScript-built content are tested instead of skipped.
- Fewer false alarms. Contrast is measured from the colors actually displayed, not guessed from the code.
- Evidence you can act on. Each finding names the page, the element and the measured values, so a developer can find and fix it.
- A score that reflects the real page. Your score is calculated from what visitors receive.
What Deep Scan cannot do
Automated testing finds only part of the accessibility problems on a site. A browser can measure contrast and detect a missing label. It cannot decide whether alt text describes the picture well, whether captions are accurate, or whether instructions make sense. WCAGScan marks those items for manual checking instead of guessing.
- Deep Scan tests the page as it loads. Content that appears only after a specific click or a long series of steps may not be reached.
- Some sites block automated visitors. If a page refuses to load for the scanner, the report lists it as not scanned.
- A real browser takes longer than reading a file, so a large Deep Scan can take several minutes.
Questions people ask
Do I have to turn Deep Scan on?
No. Scans on wcagscan.net use the Deep Scan engine. There is nothing to install and nothing to configure.
Does Deep Scan change my website?
No. It visits your pages the way a visitor’s browser does and reads them. It does not change anything on your site.
My visitors use Safari or Firefox. Does that matter?
The accessibility rules are the same in every browser, and most problems Deep Scan finds exist in all of them. For anything critical, also check the page yourself in the browsers your visitors use.
Where WCAGScan helps
Every WCAGScan report starts with Deep Scan, then lists each issue with its WCAG success criterion, level, page and element. AI-written fix suggestions explain how to correct the most common problems; see the guide to AI enhancements. Run a free scan or see pricing.