PDF Accessibility Checklist

Note: PDF stands for Adobe’s Preferred Document Format file type

Accessible PDF documents are “tagged” PDF documents. The tags within a PDF document provide a textual representation of the information being visually presented. Properly tagged PDFs are essential for users of assistive technology such as screen readers or refreshable braille displays!

A properly tagged PDF document provides assistive technology software with the necessary semantic and structural elements needed to identify, interpret, and present the content to the user in a manner that allows them to fully comprehend the context, meaning, and relationships of the data being presented.

PDFs can be created from within Acrobat, however, more often than not they are created in a different application and then converted into a PDF document. The accessibility issues that may or may not arise will depend on how the source file was formatted as well as how the PDF was created.

For these reasons, it is always wise to have the source file on hand, if possible, should you need to recreate the PDF.

The Checklist

1. Was the PDF given a descriptive file name?
   • Descriptive file names help all users easily locate, open, and switch between documents.
   • File names need to be different from the title set within the document properties. Assistive technology will read the file name when the document is accessed, if the file name and the title are the same, it may lead to confusion for the user as they may think they did not open the file.
2. Is the PDF a tagged PDF?
   • Untagged PDFs do not provide the necessary semantic and structural elements assistive technology software needs to identify, interpret, and present the content to the user.
3. In the document properties, was the document given a contextual title in the Title Field, and was the Initial View set to “Document Title”?
   • Contextual document titles allow users of assistive technology to quickly determine whether the file contains the information they are seeking without having to read through the content.
   • Setting the initial view to document title will tell assistive technology software to read the document title when the document is opened rather than the file name.
   • The default view in the settings for a PDF is set to File Name, if this is not changed, users of assistive technology will hear the file name twice. Once when the file is accessed, and again when it is opened. This can confuse the user as it may lead them to open the file more than one time.
4. If the document content is not written in the English Language, has the appropriate Language been set within the document properties?
   • The default language within Acrobat is English. If the appropriate language for the document is not set, words may be mispronounced by screen readers, which can lead to misinterpretation of both the meaning and overall concept of the information being conveyed.
5. In the document security properties, is “Content Copying for Assistive Technology” allowed?
   • If “content copying for assistive technology” is not allowed it may block assistive technology from accessing the content.
6. Does the PDF contain a Table of Contents or TOC? If so, is it properly tagged as a Table of Contents? (in the tags pane is there a ‘TOC’ tag and are there one or more ‘TOC’ tags nested under it?)
   • Properly tagging the table of contents provides users of assistive technology with the structure and overview of all the topics and subtopics within the document.
   • Users of assistive technology can use the table of contents links to easily locate and jump to specific sections of the document without needing to skim through all of the content to locate the information they are seeking.
7. For documents longer than 9 pages, were accurate bookmarks provided?
   • Users of assistive technology can use the bookmarks to easily locate and jump to specific sections of the document without needing to skim through all of the content to locate the information they are seeking.
8. Are all visual headings tagged with Heading Tags, and do they follow a logical heading structure (e.g., H1, H2, H3, etc)?
   • Assistive technology does not recognize visual text formatting (bold, underline, etc.), therefore it cannot infer and translate the meaning behind these types of visual cues to the user.
   • Headings are used to break up sections of content, and they provide a logical structure for the user to navigate through PDF documents to quickly locate specific sections of content.
9. Is each visual paragraph of text tagged with individual paragraph tags?
   • If text containing meaningful information is not tagged as text it cannot be identified as such by assistive technology and therefore will not be conveyed to the user.
   • Improper tagging of content, such as tagging paragraphs of text as list items; or vice versa can lead users of assistive technology to misunderstand the meaning of the content being presented.
   • Visual separation of paragraphs cannot be identified by assistive technology. Tagging paragraphs of text as they are visually presented provides users of assistive technology with the structure needed to determine the meaningful associations between distinct pieces of content.
10. Are all visual lists tagged as lists? If so, are the lists tagged correctly? (in the tag pane, is there a list tag ‘L’ and is there one or more nested list item tags ‘LI’ under it)?
    • Assistive technology cannot infer meaning from symbols and indentations. Properly tagged lists provide the structure that assistive technology software needs to identify, interpret and present the list items to the user in a manner that allows them to fully comprehend the context and meaning of the data being presented.
11. If color/text formatting (bold, font, etc.) was used to convey important information, was the importance of the information conveyed in another manner?
    • The importance of information conveyed by using color/text formatting will be unavailable to users of assistive technology.
    • The importance of information indicated by color alone will be missed by those using non-color/non-visual displays.
    • In addition, those who experience color-blindness may not be able to discern the colors used to indicate the importance of the information being conveyed.
12. Are there paragraphs of text presented in a different language than the document language set in document properties? If so, is the appropriate language attribute selected within the paragraph tag properties?
    • When the language of a paragraph differs from the document language, assistive technology uses the language attribute in the tag properties to provide the appropriate pronunciation for the section.
    • If the appropriate language attribute is not set, words may be mispronounced which can lead to misinterpretation of both the meaning and overall concept of the information being conveyed.
13. Do all images, images of text, charts, and graphs that provide meaningful information not conveyed within the surrounding text have descriptive alternative text?
    • Those with visual impairments or cognitive/learning disabilities often rely on the text-based speech output of assistive technology to describe what is visually being presented.
    • Alternative text should convey the meaning and concept of the information being presented without the need for visual cues.
    • Low-vision users may use screen magnifiers and may have the magnification set to a level which makes it difficult to see an entire image and surrounding content at once.
14. Are hyperlinks present within the PDF? If so, are they tagged as actionable links (in the tags pane, is there a ‘link’ tag with an ‘OBJR’ tag nested under it)? Does hyperlink text convey the destination?
    • Links that do not contain the link OBLR tag are not seen as actionable links to assistive technology, therefore the user will not be able to use the link to navigate to the intended destination.
    • Screen Readers contain a feature called a “Links List” which users can use to pull up and navigate to all the links contained within a document. When generic hyperlink text such as “click here” is used multiple times, users are unable to discern one link from another within the links list.
    • In addition, generic hyperlink text will not provide users of assistive technology with the intended purpose or destination of the link.
15. Does the PDF contain data tables? If so, are they tagged as tables with proper structural markup (in the tags pane is there a ‘Table’ tag with the appropriate ‘TR’, ‘TH’ and ‘TD’ tags nested under it)? Does the table markup match the visual representation of the table within the document?
    • Tables that do not have proper structural markup will not be fully accessible to users of assistive technology as the relationships between the intended header and data cells is not accurately represented.
    • If the structural markup of a table does not match the visual representation of that table, the information will not be accurately identified and rendered by assistive technology in the manner intended.
    • Improper table markup can lead to misinterpretation of the data being presented and may lead users to misunderstand the overall concept of the data.
16. Does the PDF contain form fields? If so, does each form field have a unique label and tooltip associated with it? Are the form fields properly tagged (for each form field: in the tags pane, are there ‘Form’ tags with an associated ‘OBJR’ tag nested under it?
    • Each form field must have a unique label and tooltip that allows assistive technology users to correctly identify what the form field’s purpose is and enter the correct information being requested.
    • The use of generic or repetitive form labels and tooltips can lead to confusion and frustration for users of assistive technology, as they will not be able to properly identify the intended purpose of each form field.
    • Additionally, should changes need to be made to the form fields, the use of generic form fields and labels will make this task more difficult as the editor will need to manually scroll through the form field list while looking at the PDF form itself to determine if they are editing the correct field.
17. If any audio or video content was embedded into the PDF, were text alternatives (e.g., captions/transcripts) provided and are the audio and video controls accessible?
    • Without captions or transcripts for video with audio, users with hearing impairments may misinterpret the meaning and context of the visual content being presented. Additionally, if there is an audio file with no corresponding video, they will miss the content entirely.
    • Screen reader users rely on text based speech output to describe what is visually being presented; therefore, although they may be able to hear the corresponding audio, the meaning and context of the information being presented may not be clear to them.
    • If the video/audio controls are not accessible (e.g. labeled), screen reader users will find it difficult to access and navigate through them.
18. Is the document free of flashing or flickering content?
    • Content that flashes or flickers can trigger seizures for some users.
    • In addition, flashing or flickering content can be distracting for users with cognitive/learning disabilities and all users in general.
19. Are all non-standard tags appropriately mapped to standard Adobe tags?
    • Standard tags in Adobe are universal semantic structure tags such as <H1>, <H2>, etc., for headings, <P> for paragraph text, and <Figure> for images.
    • PDF documents created from source files such as InDesign or Word may contain “non-standard tags” such as “<Heading>”, or “<Text>”, these tags need to be “role-mapped” to standard Adobe tags for screen readers to appropriately identify and render the content to end-users.
      • For Example: <Heading> would need to be mapped to the appropriate heading tag such as <H1>, <H2>, etc., and<Text> would need to be mapped to <p>.
20. Is the order in the tag structure accurate and logical? Do the tags match the reading order?
    • Screen readers will use the tag structure to read the document, and you want to make sure that the screen reader follows the reading order you set when creating your tags.