Diana Kosovacova
Digital Marketing Lead & PDF Accessibility Evangelist
Quick Overview: Broken table tags usually come down to three problems: cells tagged as the wrong type (TH vs. TD), TH cells missing a scope attribute (row, column, or both), or tables that shouldn’t be data tables at all. In PDFix Desktop, run Validation to find every table error, fix TH/TD structure and scope in the Tags panel, then remove or flatten layout-only tables and revalidate to confirm compliance. Tagging a table from scratch instead? See our companion guide How to make PDF tables accessible.
Key Takeaways:
- Screen readers depend on correct TR / TH / TD structure and the scope attribute – fixing one without the other still fails accessibility checks.
- Not every table tag in a PDF represents real data – some are used purely for visual layout and should have their table structure removed, not “fixed” as a table.
- Fix Table Tag → Flatten Table Structure only runs when PDFix detects a structural table error (e.g. span/grid problems). A perfectly regular layout table is left unchanged – use Delete Tags instead (and set Content to Move nested tags to the parent tag).
- Actions run from the toolbar or Actions tab apply document-wide – always use right-click → Run Action on Selection when you want to fix just one table without touching real data tables elsewhere in the file.
- Fix Table Cells can batch-apply the scope attribute across many tables at once, using rules based on font, color, or cell position, and the rule set can be saved as a reusable template.
ON THIS PAGE
Why Broken Table Tags Break Accessibility
Correct table tagging depends on structure and attributes working together. Each table needs TR tags for rows, containing TH cells for headers and TD cells for data — and each TH needs a scope attribute (row, column, or both) telling assistive technology which data cells that header describes. Get the structure right but skip scope, and a screen reader still can’t announce the correct header for a given cell.
PDFix Desktop lets you diagnose and fix both problems, plus a third one that’s easy to overlook: tables that were never meant to hold data at all.
❶ Run Validation to Find Table Tag Errors
Run Validation from the Tags panel to get a complete list of every table tag error before you fix anything by hand.
Open your PDF in PDFix Desktop, go to the Tags panel, and run Validation. Choose your target profile:
- PDF/UA-1 – checked via the veraPDF engine and the Matterhorn Protocol
- Custom WCAG 2.1/2.2 profile – for organization-specific rules
- Arlington PDF Model – checks underlying PDF syntax issues
Turn on Show Errors to jump straight to the problem element in the page view. Try AutoFix first, then Revalidate – fix anything that remains using the steps below.
❷ Fix TH/TD Structure Errors
Review the tag tree and correct any cell that’s tagged as the wrong type before touching scope or automation.
- Header tagged as data (or vice versa)? Double-click the tag and choose the correct type from the dropdown.
- Rows or cells out of order? Rearrange them with drag-and-drop.
- Whole table tagged as a paragraph? Delete the incorrect tag and retag it as a table with the Table Tool.
❸ Add Missing Scope Attributes
A TH cell with no scope attribute is the single most common table validation failure – here’s the fix.
- Open the Tag Attributes panel (Properties button in the Tags panel).
- Right-click the
THtag → Add Attribute → select Scope from the Standard Table Attributes list. - Set the value to
row,column, orboth. - Select multiple
THtags at once to apply the same scope value to all of them in one step.
❹ Fix Tables That Are Used Only for Page Layout
Not every Table tag holds real data – layout-only tables should become plain text, not an accessible table. PDF/UA validators often flag these as the table is not regular (or “tables are not regular”) error, meaning the rows and columns don’t form a consistent grid.
NOTE
Running an action from the toolbar buttons or the Actions tab applies it to the entire document — including any real data tables you want to keep intact. To fix only the layout table, right-click that specific table tag and choose Run Action on Selection, which scopes the fix to just that tag.
| Method | What it does | Works on |
|---|---|---|
| Delete Tags Run Action on Selection → Delete Tags → pick TD, TH, TR, Table | Removes the table structure entirely; moves the nested paragraph tags up to the parent so the text stays readable. | Any table shape – regular or irregular. |
| Flatten Table Structure Run Action on Selection → Fix Table Tag → check Flatten Table Structure | Converts Table/TR/TH/TD (and related row section tags) to Div. Runs only when PDFix already sees a structural table error. On a clean regular table it does nothing – use Delete Tags. Do not invent an irregular span just to force flatten; prefer Delete Tags for layout tables. | Irregularly shaped tables only – e.g. a row where one cell spans two columns while others don’t. On a perfectly regular table it does nothing until you make one cell’s span irregular, or you use Delete Tags instead. |
Either way: after removing the table structure, check the reading order of the resulting paragraph tags – content that used to read across columns may need reordering to flow correctly.
❯❯❯ Watch Paul Rayius’s PDF Layout & Data Table: How to Fix Table Tagging for Accessibility | PDFix Desktop Tutorial
❺ Link Data Cells to the Right Headers with Associated Header
For tables with row, column, and section-level headers, link each data cell to its header explicitly – scope alone isn’t always enough.
Select the data cells, right-click the header cell they belong to, and choose Associate Header. Turn on Highlight Associated Headers and Highlight Children of Associated TH (Tags panel display options) to visually confirm every data cell has both a row header and a column header linked to it.
Seeing the exact PDF/UA error “table header cell has no associated subcells”? For a focused, error-specific walkthrough, see Table Header Cell Has No Associated Subcells (TH/TD Fix)
❻ Automate Scope and Header Fixes Across Many Tables
Fixing the same missing-scope error one cell at a time doesn’t scale past a handful of tables – automate it instead.
PDFix Desktop’s Set Table Cells function applies scope to every matching TH cell in one action, based on a rule you define:
- By font/color – e.g. “any
THwith white text is a column header; anyTHwith black text on the left is a row header” - By position – e.g. “any header left of point X on the page is a row header”
Save the rule set as a template and reuse it across a whole folder of documents that share the same layout. For the complete click-by-click walkthrough and a ready-to-use JSON template, see Automate PDF Table Cell Scope: Fixing TH and Row Headers at Scale.
❼ Revalidate and Confirm PDF/UA Compliance
Run Validation one more time – a clean pass confirms the fix actually holds up under the accessibility profile you’re targeting.
It’s still worth spot-checking the tag tree directly: a table can pass automated checks while still being tagged in a way that doesn’t make logical sense – for example, correctly structured TR/TH/TD cells that are tagged as a paragraph at the parent level.
What’s New in PDFix Desktop (2026)
Batch Scope Automation with Set Table Cells Attributes + Templates
Instead of fixing missing scope attributes cell-by-cell, define the rule once (by font, color, or position) and apply it across every matching TH cell in a document – or save it as a template and re-run it across a whole batch of similarly formatted files. Full guide: Automate PDF Table Cell Scope.
Layout Template for Repeated, Complex Tables
For document sets where the same table structure repeats – including tables that span multiple pages – PDFix’s Layout Template feature auto-tags the recurring structure from a single saved JSON template, dramatically cutting manual remediation time.
AI-Generated Table Summaries
Once a table’s tags are fixed, PDFix Desktop can generate a text summary for the table using OpenAI – install the Table Summary action from the Action Marketplace, add your OpenAI API key, and run it against your fixed table tags to produce a screen-reader-friendly description automatically.
PDF TABLES LEARNING HUB
Every guide, webinar and reference
All PDF Table accessibility guides organized by what you are trying to do.
PDFix Table Tool Tutorial
How to Use the Table Tool in PDFix Desktop
PDFix Webinar
How to Tag Complex PDF Tables for Accessibility
PDF Layout & Data Table Tutorial
How to Fix Table Tagging for Accessibility | PDFix Desktop Tutorial
PDF Table Tutorial
Tagging Complex Tables with Row Headers | PDFix Desktop Tutorial
Frequently Asked Questions
What’s the difference between TH and TD in a PDF table?
TH marks a header cell and TD marks a data cell. Screen readers use this distinction, combined with the scope attribute on TH cells, to announce the correct header when reading a data cell.
What does the scope attribute do?
scope tells assistive technology whether a header cell applies to the rest of its row (row), the rest of its column (column), or both. Without it, a screen reader may correctly identify a cell as a header but still won’t know which data cells it describes.
How do I fix a table that’s only used for page layout, not real data?
Select that specific table tag, use Run Action on Selection (not the toolbar button, which applies document-wide) with either Delete Tags (removes TD/TH/TR/Table and keeps the text as paragraphs) or Fix Table Tag → Flatten Table Structure (converts the structure to div tags). Then check the reading order of the resulting text.
Can I fix missing scope attributes across many tables at once?
Yes. PDFix Desktop’s Fix Table Cells function applies scope based on rules you define – such as font color or cell position – across every matching header cell in a document, and the rule set can be saved as a template for reuse on similar files. See Automate PDF Table Cell Scope for the full walkthrough.
What does “table header cell has no associated subcells” mean?
It means a TH header cell hasn’t been explicitly linked to the TD data cells it describes, which fails PDF/UA, WCAG, and ISO 14289 validation. Fix it with the Associate Header action described in Step 5, or see our dedicated guide, How to Add Associated Headers in PDF Tables.
What does “tables are not regular” mean in PDF/UA validation?
It means a table’s rows and columns don’t form a consistent grid – for example, rows with a different number of cells or inconsistent column spans. This affects which fix method works: Fix Table Tag → Flatten Table Structure only works on irregular tables, while Delete Tags works on any shape (see Step 4).
