Standard

Common WCAG violations

A WCAG violation is a failure to meet one of the Web Content Accessibility Guidelines' success criteria. In documents specifically, the same handful of failures account for the large majority of problems, because they stem from how documents are typically authored — designed to look right on screen, with the underlying structure left incomplete. Knowing the common ones is the fastest route to fixing most of a document's accessibility gaps.

Missing alternative text (WCAG 1.1.1, Non-text Content) is the most frequent. Images, charts, logos, and other graphics that carry meaning need a text alternative; when they have none, a screen-reader user gets silence or a filename. The mirror-image failure is decorative images that are not marked as artifacts, so they get announced pointlessly.

Untagged or badly tagged content is the structural equivalent. If a document is not tagged, assistive technology has no map of what anything is, and the content is effectively invisible even though it displays normally. Related failures include headings that are only visually styled rather than marked as real headings (undermining WCAG 1.3.1, Info and Relationships, and heading-based navigation), and lists or tables that aren't marked up so their relationships are lost.

Incorrect reading order (WCAG 1.3.2, Meaningful Sequence) is common in multi-column layouts, sidebars, and text boxes: the visual order and the underlying order diverge, so the content is read aloud in a nonsensical sequence. Tables without identified header cells break the association between a data cell and its row and column headers, making data tables unreadable in speech.

Undeclared document language (WCAG 3.1.1, Language of Page) causes speech synthesis to mispronounce content, and unmarked language changes within a document compound it. Low colour contrast (WCAG 1.4.3) affects readers with low vision, and images of text — a scanned page with no real text layer — fail multiple criteria at once until the content is put through OCR and tagged.

The important practical distinction is that some of these are machine-detectable and some are not. A tool can reliably flag missing alt text, untagged content, an undeclared language, and unmarked table headers. It cannot judge whether existing alt text is meaningful or whether a defined reading order is the logical one — those need human review. An honest check reports both.

How Normfluss checks for it

Normfluss detects the machine-checkable violations directly: missing alternative text, untagged or incorrectly structured content, undeclared language, unmarked table headers, and unlabelled form fields, and it remediates the fixable ones automatically. The violations that need judgement — whether alt text is meaningful, whether the reading order makes sense — are routed to a guided human review rather than passed silently. Every finding and fix is recorded in a signed evidence record that states exactly what was tested, within a stated scope, rather than a single pass/fail badge.

Frequently asked questions

What are common WCAG violations?

In documents, the most common are: missing alternative text on meaningful images (1.1.1), untagged content and visually-styled-but-unmarked headings (1.3.1), incorrect reading order (1.3.2), tables without identified header cells, undeclared document language (3.1.1), low colour contrast (1.4.3), and images of text with no real text layer. Most stem from documents authored to look right without the underlying structure.

What is a WCAG violation?

It's a failure to meet one of WCAG's success criteria — the specific, testable conditions that make content accessible. For example, an image with no text alternative fails 1.1.1; content that can't be operated by keyboard fails 2.1.1. Laws that reference WCAG at Level AA treat these criteria as the conformance baseline.

Which WCAG violations can be detected automatically?

Automated tools reliably detect structural, objective failures: missing alt text, untagged content, undeclared language, unmarked table headers, unlabelled form fields, and low colour contrast. They cannot judge subjective quality — whether alt text is meaningful or whether the reading order is logical — which is why those are referred to human review.

How do I fix WCAG violations in a document?

Tag the document and correct its structure, add alternative text to meaningful images and mark decorative ones as artifacts, define the correct reading order, mark up headings, lists and table headers, and declare the language. Machine-fixable issues can be remediated automatically at scale; the judgement-dependent ones need a person. Re-check after fixing and keep evidence of what was tested.