WCAG Violations: The Data on What Breaks, What Gets Sued Over, and What 35M Fixes Show

Last updated: September 8, 2026

A WCAG violation is a specific, testable failure of one of the Web Content Accessibility Guidelines' success criteria — a single element, on a single page, that does not meet a defined requirement. This analysis separates the violations that are most common from the violations that most often appear in legal complaints, using WebAIM's 2026 crawl of one million home pages, US Department of Justice guidance, and TestParty's remediation of 35 million-plus issues.

Key numbers: Six error categories — low-contrast text, missing alt text, missing form labels, empty links, empty buttons, and missing document language — account for roughly 96% of all detected WCAG errors across the top one million home pages (WebAIM Million 2026). 95.9% of those home pages had detectable WCAG 2 failures in 2026, up from 94.8% in 2025, averaging 56.1 errors per page (WebAIM). Federal website accessibility lawsuit filings reached 3,117 in 2025, a 27% increase over 2024 and 36% of all ADA Title III federal filings (Seyfarth Shaw). Of the six barriers the DOJ names in its web accessibility guidance, three cannot be reliably detected by any automated scanner (ADA.gov, March 2022). As of August 2026, TestParty has remediated 35 million-plus accessibility issues across 100+ brands, with detection running 60–70% automated and 30% expert manual.

What exactly is a WCAG violation?

A WCAG violation is a single instance of content failing a success criterion: the criterion is the rule, the violation is the countable breach of it, and one failed criterion can generate hundreds of violations on a single page — or none.

That distinction matters more than it sounds. WCAG 2.2 contains 86 success criteria, 55 of them at Levels A and AA (W3C), and Level AA is the benchmark US courts, the DOJ's Title II rule, and EN 301 549 all point to. A site with "1,200 violations" has not failed 1,200 rules; it has probably failed twelve, repeated across a product grid. This is why raw violation counts are useful for tracking work and useless for judging risk. Two stores with identical error totals can carry completely different legal exposure depending on which criteria failed and where. Counting tells you the size of the job. Only classification tells you the size of the problem.

Which WCAG violations are most common?

Low-contrast text is the most common WCAG violation on the web, detected on 83.9% of the top one million home pages in 2026 — followed by missing alternative text (53.1%) and missing form input labels (51.0%), according to the WebAIM Million.

The distribution has been remarkably stable for years. What changed in 2026 is direction: after a modest improvement in 2025, four of the six categories got worse.

+-------------------------------+----------------------------+--------------------+----------------------------------+
|         Violation type        |   % of home pages (2026)   |   Change vs 2025   |      Primary WCAG criterion      |
+-------------------------------+----------------------------+--------------------+----------------------------------+
|       Low-contrast text       |           83.9%            |      +4.8 pts      |   1.4.3 Contrast (Minimum), AA   |
+-------------------------------+----------------------------+--------------------+----------------------------------+
|     Missing image alt text    |           53.1%            |      −2.4 pts      |    1.1.1 Non-text Content, A     |
+-------------------------------+----------------------------+--------------------+----------------------------------+
|   Missing form input labels   |           51.0%            |      +2.8 pts      |       1.3.1, 3.3.2, 4.1.2        |
+-------------------------------+----------------------------+--------------------+----------------------------------+
|          Empty links          |           46.3%            |      +0.9 pts      |      2.4.4 Link Purpose, A       |
+-------------------------------+----------------------------+--------------------+----------------------------------+
|         Empty buttons         |           30.6%            |      +1.0 pts      |    4.1.2 Name, Role, Value, A    |
+-------------------------------+----------------------------+--------------------+----------------------------------+
|   Missing document language   |           13.5%            |      −2.3 pts      |    3.1.1 Language of Page, A     |
+-------------------------------+----------------------------+--------------------+----------------------------------+

Source: WebAIM Million 2026, analysis of 1,000,000 home pages.

The likely driver is complexity, not carelessness: the average home page grew to 1,437 elements in 2026, a 22.5% year-over-year increase (WebAIM). More components means more surface area, and error counts rose 10.1% to 56.1 per page.

Which violations actually appear in lawsuits?

The violations that appear in accessibility complaints are not simply the most frequent ones. Plaintiffs describe barriers that stopped them from completing a task — buying, booking, contacting — which is a different list from what scanners surface.

The DOJ's web accessibility guidance (ADA.gov, March 2022) names six example barriers. Mapping them against what automated tooling can find produces the central finding of this analysis.

+------------------------------------------------+----------------------------------+--------------------------+
|         Barrier named in DOJ guidance          |   Found by automated scanners?   |   In WebAIM's top six?   |
+------------------------------------------------+----------------------------------+--------------------------+
|              Poor color contrast               |               Yes                |       Yes (83.9%)        |
+------------------------------------------------+----------------------------------+--------------------------+
|           Lack of alt text on images           |    Presence only, not quality    |       Yes (53.1%)        |
+------------------------------------------------+----------------------------------+--------------------------+
|           Inaccessible online forms            |       Missing labels only        |   Labels only (51.0%)    |
+------------------------------------------------+----------------------------------+--------------------------+
|      Use of color alone to convey meaning      |                No                |            No            |
+------------------------------------------------+----------------------------------+--------------------------+
|             No captions on videos              |                No                |            No            |
+------------------------------------------------+----------------------------------+--------------------------+
|   Mouse-only navigation (no keyboard access)   |                No                |            No            |
+------------------------------------------------+----------------------------------+--------------------------+

Three of the six are effectively invisible to a scan. A tool can confirm an image has an `alt` attribute; it cannot confirm the text describes the product. It can flag an unlabeled input; it cannot tell you the error message after a failed checkout is conveyed by red border alone. Keyboard access — the barrier most likely to make a transaction impossible — requires a human with a keyboard. This is precisely why TestParty's detection stack is 60–70% automated and 30% expert manual review: the automated share finds the volume, the manual share finds the barriers that end up quoted in a complaint.

The volume is not trivial. Federal website accessibility filings hit 3,117 in 2025, up 27% from 2,452 in 2024, concentrated in New York (1,021), Florida (961), and Illinois (585), per Seyfarth Shaw's tracking. Ecommerce accounts for 69–77% of digital accessibility suits (Seyfarth Shaw).

How should you rank violations by severity?

Rank violations by what they do to a user, not by how many there are. A single keyboard trap in a checkout modal outweighs four hundred low-contrast footer links, because one ends the transaction and the other slows it down.

We use a three-tier framework in remediation planning:

+--------------+----------------------------------------------------+----------------------------------------------------+
|     Tier     |                    User impact                     |             Representative violations              |
+--------------+----------------------------------------------------+----------------------------------------------------+
|   Blocker    |          Task cannot be completed at all           | Keyboard trap in a modal; unlabeled required checkout field; custom dropdown with no keyboard operation (2.1.1, 2.1.2, 3.3.2) |
+--------------+----------------------------------------------------+----------------------------------------------------+
|   Barrier    |         Task completable only by guessing          | Empty link or button announced as just "link"; image-only "Add to cart"; filters that update results silently (2.4.4, 4.1.2, 4.1.3) |
+--------------+----------------------------------------------------+----------------------------------------------------+
|   Friction   | Task completable, but slower and more error-prone  | Low-contrast body text; vague link text; missing document language (1.4.3, 2.4.4, 3.1.1) |
+--------------+----------------------------------------------------+----------------------------------------------------+

In our assessment, blockers carry disproportionate legal exposure relative to their share of the error count — they are the failures a tester can narrate as a lived experience, and narration is what fills a complaint. Friction dominates the volume charts. Blockers dominate the filings. Sorting a scan report by criterion severity rather than instance count reverses the work order most teams start with, and it is the single highest-leverage change to a remediation plan. Our WCAG remediation walkthrough covers the code-level fix for each tier.

Where do ecommerce sites break most often?

In TestParty's audits of Shopify and ecommerce storefronts, the most persistent violation categories cluster in five places — and none of them is the home page that scanners crawl.

The recurring hot spots, qualitatively, are:

• Variant swatches. Color and size selectors built from `div` or styled `label` elements with no accessible name, no group semantics, and selected state conveyed by border color alone (1.4.1, 4.1.2).
• Faceted filters. Results update via JavaScript with no status announcement, so a screen reader user has no signal that the grid changed (4.1.3 Status Messages).
• Promotional popups. Email-capture and discount modals that do not move focus in, do not trap it, and do not close on Escape (2.1.2, 2.4.3).
• Checkout fields and errors. Placeholder text used instead of labels; validation errors indicated by red outline and position rather than programmatic association (3.3.1, 3.3.2, 1.4.1). We cover this surface in depth in our Shopify checkout accessibility guide.
• Product media. Carousels that auto-advance with no pause control and gallery thumbnails carrying filename alt text (2.2.2, 1.1.1).

Industry patterns for other verticals are broken out in our analysis of common WCAG failures by industry.

Why do violations keep coming back after you fix them?

Violations accumulate because the storefront changes faster than the audit cycle. A theme update, a new app, a seasonal landing page, or a merchandiser adding two hundred products can reintroduce a fixed criterion within a single sprint.

The platform baseline sets the starting point. Shopify's own theme store requirements cover roughly 16–22% of WCAG success criteria — a floor, not compliance. In TestParty's assessments, the Dawn reference theme ships with 30–100 violations out of the box, and premium themes commonly carry 100–350 before a merchant adds anything. Third-party apps are not reviewed for accessibility at all, and each one injects markup nobody audited. Multiply that across 2 million-plus Shopify merchants and 60,000-plus Plus stores and the drift is structural rather than negligent. A point-in-time audit produces a snapshot with a short shelf life; the audit process guide explains where that snapshot does and does not help.

What does fixing violations at scale look like?

At scale, fixing WCAG violations means treating them as a continuous code-quality problem rather than a project. As of August 2026, TestParty has remediated 35 million-plus accessibility issues across 100+ brands, including Dickies, Eddie Bauer, and Billabong.

The operating pattern that produces those numbers is unglamorous: daily automated scans catch regressions within a release cycle, monthly expert manual audits cover the 30% of issues no scanner reaches, and every fix ships as a pull request against the theme's actual source — HTML, Liquid, CSS, JavaScript — so the correction lives in the markup assistive technology parses rather than in a script layered on top. Initial remediation runs about 14 days, and merchants spend roughly 15–30 minutes a month reviewing PRs. Verification is threshold-based rather than vibes-based: a Lighthouse accessibility score of 90 or above, five or fewer WAVE errors, three or fewer axe errors. If you want your own baseline distribution before deciding what to prioritize, start with a free Shopify accessibility scan.

Frequently Asked Questions

What is the difference between a WCAG violation and a WCAG success criterion? A success criterion is a testable rule in the WCAG specification — there are 86 in WCAG 2.2, with 55 at Levels A and AA (W3C). A violation is a single instance of content failing one of those rules. One failed criterion applied to a product grid template can produce hundreds of violations at once, which is why instance counts overstate how many distinct problems exist.

What is the most common WCAG violation? Low-contrast text, detected on 83.9% of the top one million home pages in 2026 (WebAIM Million) and mapping to success criterion 1.4.3 Contrast (Minimum) at Level AA. It rose from 79.1% in 2025. It is also among the cheapest categories to fix, since most instances trace back to a small number of color tokens in a theme's stylesheet rather than to individual elements.

Do automated scanners catch every WCAG violation? No. Automated tooling reliably detects 60–70% of issues; the remaining 30% require expert manual testing. Three of the six barriers the DOJ names in its web accessibility guidance — use of color alone, missing video captions, and mouse-only navigation — cannot be reliably detected by any scanner, and those are the barrier types most likely to make a transaction impossible.

Which WCAG violations are most likely to trigger a lawsuit? Barriers that stop a transaction: keyboard-inaccessible navigation and controls, unlabeled or unannounced checkout form errors, screen-reader-blocking modals, and missing alt text on functional images. Federal website accessibility filings reached 3,117 in 2025, up 27% year over year, with ecommerce named in 69–77% of digital accessibility suits (Seyfarth Shaw).

How many WCAG violations does a typical Shopify store have? It depends on the theme and app stack. In TestParty's assessments, Shopify's Dawn theme ships with 30–100 violations out of the box and premium themes commonly carry 100–350 before a merchant adds products or apps. Shopify's theme store requirements cover roughly 16–22% of WCAG success criteria, so passing them is a floor rather than compliance.

Can you ever reach zero WCAG violations? Zero on an automated scan is achievable and worth targeting, but it is not the same as conformance, because scanners cannot evaluate the criteria that require judgment. Practical verified thresholds are a Lighthouse accessibility score of 90 or above, five or fewer WAVE errors, and three or fewer axe errors, paired with periodic manual testing of keyboard paths and screen reader flows.

How fast do fixed violations come back? Fast enough that monitoring matters more than any single audit. Theme updates, new apps, seasonal landing pages, and bulk product uploads all reintroduce previously fixed criteria. Average home page complexity rose 22.5% year over year in 2026 and average errors per page rose 10.1% to 56.1 (WebAIM Million), which is drift showing up in industry-wide data.

Humans + AI = this article. TestParty uses a cyborg approach to content — combining human accessibility expertise with AI capabilities to produce accurate, comprehensive guides. This content is for educational purposes and reflects our analysis of publicly available information as of the publication date. TestParty competes in the digital accessibility market, and we encourage readers to evaluate all solutions independently based on their specific needs.