WCAG 2.2 PDF Techniques
Web Content Accessibility Guidance (WCAG) offers PDF Techniques, which are best practices for keeping PDF documents accessible. Bloom’s PDF Accessibility Scanner can verify or partially verify whether many of these techniques have been correctly applied.
-
Strong coverage. Distinguishes missing /Alt, empty /Alt, non-empty /Alt, and ActualText-only. Detects image-backed figures through marked-content MCIDs. Cannot judge whether alt text is semantically meaningful.
View PDF1 technique ↗ -
Verifies bookmark presence for documents over 20 pages. Does not verify that bookmarks link to the correct sections.
View PDF2 technique ↗ -
Recognizes artifact scope for text-showing content and avoids failing artifacted content. Does not fully verify decorative image artifacting.
View PDF4 technique ↗ -
Validates table hierarchy, row/cell parentage, headers, and row regularity warnings.
View PDF6 technique ↗ -
Detects likely image-only PDFs by checking for fonts and text objects. Does not perform OCR or verify OCR accuracy.
View PDF7 technique ↗ -
Catches skipped heading levels, plain H tags, first heading not H1, and missing headings in tagged documents.
View PDF9 technique ↗ -
Verifies programmatic field descriptions such as /TU tooltip metadata. Does not check visual label positioning.
View PDF10 technique ↗ -
Identifies annotation-to-structure linkage via /StructParent, /ParentTree, and matching /OBJR. Does not verify link text quality or keyboard activation.
View PDF11 technique ↗ -
Detects field type, descriptions, widgets, and page association. Does not verify full accessibility API value/state.
View PDF12 technique ↗ -
Can catch header/footer-like text that is untagged or not artifacted. Does not verify consistency.
View PDF14 technique ↗ -
Verifies that /Lang exists and is a valid language tag.
View PDF16 technique ↗ -
Strong direct coverage. Verifies title presence and DisplayDocTitle flag.
View PDF18 technique ↗ -
Covers the resulting structural condition: correct table cells, headers, and spans.TablesTest table structure model headers parentageView PDF20 technique ↗
-
Validates L, LI, Lbl, LBody relationships. Intentionally allows some Adobe-compatible patterns.
View PDF21 technique ↗ -
Detects interactive form fields, widgets, and descriptions. Does not test keyboard operation.
View PDF23 technique ↗
Adobe Acrobat Pro comparison
Adobe’s PDF accessibility scan is the industry standard but requires a paid license or a paid API key. The table below shows every automatic check Adobe performs and which ones the PDF Accessibility Scanner covers.
| Check | Adobe Acrobat | PDF Accessibility Scanner |
|---|---|---|
| Document | ||
| Accessibility permission flag | ✓ Yes | ✓ Yes |
| Image-only PDF detection | ✓ Yes | ✓ Yes |
| Tagged PDF | ✓ Yes | ✓ Yes |
| Logical reading order | Manual only | Manual only |
| Primary language | ✓ Yes | ✓ Yes |
| Document title | ✓ Yes | ✓ Yes |
| Bookmarks | ✓ Yes | ✓ Yes |
| Color contrast | Manual only | Manual only |
| Page Content | ||
| Tagged content | ✓ Yes | ✓ Yes |
| Tagged annotations | ✓ Yes | ~ Partial |
| Tab order | ✓ Yes | ✗ No |
| Character encoding | ✓ Yes | ✗ No |
| Tagged multimedia | ✓ Yes | ✗ No |
| Screen flicker | ✓ Yes | ✗ No |
| Scripts | ✓ Yes | ✗ No |
| Timed responses | ✓ Yes | ✗ No |
| Navigation links | ✓ Yes | ~ Partial |
| Forms | ||
| Tagged form fields | ✓ Yes | ~ Partial |
| Field descriptions | ✓ Yes | ✓ Yes |
| Alternate Text | ||
| Figures alternate text | ✓ Yes | ✓ Yes |
| Nested alternate text | ✓ Yes | ✓ Yes |
| Associated with content | ✓ Yes | ~ Partial |
| Hides annotation | ✓ Yes | ~ Partial |
| Other elements alternate text | ✓ Yes | ✗ No |
| Tables | ||
| Row structure (TR) | ✓ Yes | ✓ Yes |
| TH and TD cells | ✓ Yes | ✓ Yes |
| Table headers | ✓ Yes | ✓ Yes |
| Row/column regularity | ✓ Yes | ✓ Yes |
| Lists | ||
| List items (LI parentage) | ✓ Yes | ✓ Yes |
| Lbl and LBody structure | ✓ Yes | ✓ Yes |
| Headings | ||
| Appropriate nesting | ✓ Yes | ✓ Yes |