Incomplete tagging guide

Why a tagged PDF can still contain untagged content

Find content omitted from a PDF's structure, distinguish meaningful material from artifacts, and verify repairs after edits or mixed-file assembly.

Updated November 1, 2026. Reviewed by the DocAccessible team under our editorial policy.

A tagged PDF can still omit important text, figures, links, or annotations from its logical structure. Diagnose the problem by locating each reported object, deciding what role it serves, and checking how it is represented. Do not respond by marking everything as an artifact or by assuming that the existence of a Tags panel means the whole document is covered.

Understand what the tagged label actually tells you

Imagine an illustrative application guide exported with headings and paragraphs, then edited to add a deadline callout. The original tag tree may still exist while the new deadline is absent from it. A reader navigating the structure can miss the most important change even though the file is described as tagged. The issue is coverage and meaning, not simply the presence of a tree.

The PDF Association's accessibility-technique overview explains the distinction between real content and artifacts within accessible PDF structure. Apply that distinction to each finding. A paragraph, a qualification, and a decorative border should not receive the same treatment merely because a checker reports them together.

Locate the actual object behind the finding

Adobe's Content and Tags panel documentation describes finding unmarked content, comments, links, and annotations, and locating tags from a selection. Use the relevant inspection tools to connect the reported item with visible page content. Work on a copy and save meaningful revisions so an incorrect repair can be compared or reversed.

Check the page around the highlighted object. Several drawing operations may combine into one meaningful figure, while several nearby text runs may belong to different paragraphs. Do not infer the intended reading unit from the number of low-level objects. Read the surrounding content and consult the source author if the role is unclear.

Illustrative classification of untagged findings.
ObjectContent decisionReview target
New application deadlineSubstantive informationAppropriate structure in the intended reading position
Decorative section dividerPage decorationArtifact treatment without hiding nearby text
Link added after exportInteractive contentMeaningful link representation and destination
Unpublished reviewer annotationEditorial decision requiredRemove from the release or include and review intentionally

Repair meaning and position together

For the deadline callout, determine whether it is a paragraph, a heading, or another meaningful structure in context. Associate its content with that structure and place it where the reader needs it. Adding a generic tag at the end of the document may technically include the words while presenting the deadline too late to make sense.

For decoration, use deliberate artifact treatment. W3C technique PDF4 addresses decorative images, but it is not a license to hide difficult content. An arrow that explains process direction, a warning icon with no equivalent text, or a background map carrying information needs a content-aware decision.

After correcting an object, inspect what else changed. Confirm that meaningful text was not accidentally included in an artifact selection and that the same content has not been duplicated in multiple structures. Check the surrounding reading order, not just the newly created tag name. Where possible, repair the source and re-export to prevent the omission from returning in the next revision.

Investigate how the gap was introduced

Compare the reviewed export with the final distribution file. Look for page insertion, replacement, merging, added annotations, and later conversion steps. Adobe's documentation notes that comments added to an already tagged PDF require attention to comment tagging. Keep late editorial additions inside the release process rather than treating them as harmless visual changes.

For a combined packet, review each component and the assembled result. A tagged cover does not establish that every appendix is tagged. The merge workflow provides a separate assembly check, and the artifact guide helps with repeated page furniture.

Close the finding with evidence

  1. Record the page, object, and reason the content matters or qualifies as decoration.
  2. Check the repaired association and its position in the logical structure.
  3. Run the same validation profile again and investigate remaining coverage findings.
  4. Read the surrounding passage in the intended viewer and assistive setup.
  5. Compare the final saved file with the approved source to ensure no information was lost.

A useful closure says 'deadline callout is now read between the eligibility paragraph and application steps', supported by the relevant file revision and test. A smaller number of warnings alone does not explain whether the information is correct or usable. Keep manual review items visible even when the automated coverage report is clean.

Frequently asked questions

Can a PDF have tags and still fail content coverage checks?

A tag tree can cover only part of a document or omit objects added later. Inspect the reported content and its association with the structure instead of treating a tagged status as proof that every meaningful item is included.

Can I mark every untagged object as an artifact?

Artifact treatment is appropriate for genuine decoration or qualifying page furniture, not for meaningful content that is difficult to tag. Decide what each object communicates and preserve substantive information in the reading experience.

Should I rerun automatic tagging over the entire file?

Review the existing structure and scope of the omission first. A broad retagging operation can change work that was already correct. Prefer a source correction or a targeted repair when suitable, then validate and test the complete result.

Sources

The statements above rest on the primary sources below. Where a source is a law or standard, the version and date named in the text are the ones checked on the review date. This guide is general information, not legal advice.

  1. PDF Association: Understanding PDF accessibility techniques
  2. Adobe: Edit document structure with Content and Tags panels
  3. W3C PDF4: Decorative images and artifacts

Keep reading