Standard PDF text requires absolute X/Y grid coordinates for every word. When you turn on Reflow View (often called "Mobile View" or "Liquid Mode" in Adobe apps), the software extracts the text strings out of their rigid boxes and displays them like a standard HTML website. The viewer can instantly increase the text size to 24pt, and the words will simply wrap elegantly to the next line instead of overflowing off the edge of the monitor.
The Dependency on Tagged PDF
Reflow is incredibly difficult for software to pull off if the document is poorly built. To reflow effectively, the PDF must be "Tagged" (Logical Structure).
- The Problem: A PDF doesn't naturally know what a 'paragraph' is. It just knows "Draw the word 'apple' at X:100, Y:500". If a page has two columns of text side-by-side, basic software might read straight across both columns, combining two completely different sentences.
- The Solution: A Tagged PDF contains a hidden XML-like tree dictating the precise Reading Order: `
Title
`, `Left Column
`, `Right Column
`. - The Result: When Reflow is activated, the software ignores the X/Y coordinates entirely and simply renders the Tag Tree sequentially, ensuring the reading order is flawless.
Standard View vs. Reflow View
| Feature | Standard Fixed Layout | Reflow Mode |
|---|---|---|
| Mobile Reading | Requires constant pinch-to-zoom | Vertical scrolling only (like a website) |
| Visual Integrity | Perfect (Print Ready) | Poor. Charts/Graphs may overlap or break |
| Screen Readers (Blind UI) | Often reads out of order | Flawless sequential reading |
The Logical Structure Architecture
% To enable Reflow, the document must contain a Structure Tree. 12 0 obj << /Type /StructTreeRoot /K [ 13 0 R % Points to an Article (Art) tag 14 0 R % Points to a Section (Sect) tag ] >> endobj % Inside the Page Content Stream, text is wrapped in Marked Content (BDC) /P << /MCID 1 >> BDC BT /F1 12 Tf (This sentence will reflow perfectly.) Tj ET EMC
The BDC (Begin Dictionary Content) tells the viewer exactly which block of text belongs to which Tag, allowing the viewer to safely extract it for Reflow rendering.
Common Authoring Mistakes
- "Print to PDF" Destroys Tags. If you write a beautifully formatted Word Document and use the Windows "Print to PDF" driver, the resulting file will have zero tags. Attempting to Reflow it will result in chaotic text wrapping. You must use "Save As > PDF" to preserve the logical structure.
- Complex Tables. Financial spreadsheets exported to PDF almost never reflow correctly. Because the rigid grid structure is lost during reflow, a 10-column table will often degenerate into a massive unreadable vertical list of raw numbers.
Frequently Asked Questions
Because your PDF is 'Untagged'. Reflow relies on a hidden logical structure tree that tells the software 'This is a Header, this is Paragraph 1, this is an Image caption'. Without tags, the software just guesses the reading order based on physical proximity, which usually fails spectacularly on multi-column layouts.
Not unless they have been processed with OCR (Optical Character Recognition) first. Reflow only works on vector text. A scan is just a flat photograph of a page; you cannot reflow a photograph.
In Acrobat Reader, go to View > Zoom > Reflow (or press Ctrl+4). As you resize your application window horizontally, the text will actively wrap to fit the window width.
If the creator tagged an image as an "Artifact" (background decoration), the Reflow engine intentionally hides it to keep the reading experience clean. It assumes the image contains no important data.
They share the same goal (responsive reading), but EPUB is natively built from the ground up to be liquid HTML. PDF is natively built to be rigid paper. PDF Reflow is a 'hack' layered on top to emulate EPUB behavior.
Make Scans Reflowable
Have a flat scanned document that won't reflow on your phone? Run it through our OCR tool to extract the vector text, making it fully liquid and accessible.
OCR PDF Editor