How to Convert a PDF to Word and Keep the Formatting

By the SAA Tool editor, an IT support engineer · Last reviewed 9 October 2026 · 7 min read

A PDF to Word converter promises an editable copy that looks like the original. Sometimes you get one. Other times you get columns in the wrong order, text locked in invisible tables and a gap where the logo was. This guide explains what a converter is really doing, exactly what our PDF to Word tool keeps and drops, and how to tidy the result.

Why PDFs are hard to edit

PDF was designed so a document looks the same everywhere. The standard behind it, ISO 32000, describes a format for viewing documents "independent of the environment in which they were created". To manage that, a PDF page is a list of drawing instructions: these characters at this position in this font and size, a filled rectangle here, a picture there.

What most PDFs don't store is the structure a word processor works with. There is usually no paragraph, column or table in the file, only pieces of text at fixed coordinates, and the PDF Association points out that the order of those pieces in the file can differ from the order you see on the page. Some PDFs carry tags that mark headings, lists and tables, mainly for accessibility. Many don't, and software that works without tags relies on "its own interpretation of the page layout, fonts and other characteristics". Ours ignores tags and works from positions alone.

So every converter rebuilds the document by inference. In a two-column newsletter the top line of each column sits at the same height, so software that reads straight across would produce nonsense like "The council voted on Tuesday to Prices rose 4% in March after". Our tool looks for empty space instead: blank horizontal strips split the page into bands, and a clear vertical gutter splits a band into columns, which Keep the page layout sets side by side. Tables are found the same way, from text that lines up, not from ruled lines. Microsoft says Word's own PDF import works best with files that are mostly text.

First, check whether it is a text PDF or a scan

Open the PDF and try to select a sentence, or press Ctrl+F (⌘F on a Mac) and search for a word you can see. If nothing highlights, each page is a picture from a scanner or phone app. Our tool stops with "This PDF doesn't contain any text we can read" when it finds fewer than 20 characters in the whole file.

For a scan, run PDF OCR first: choose the language, click Make searchable, then Download searchable PDF and convert that file. For a single-column scan, pick Text only: the recognised text has no real font, and the scan's lines and logos won't come across anyway. See how to OCR a scanned PDF for scan quality tips.

Two layouts: look-alike or easy to edit

The Layout box offers two results:

In both, the page size comes from the PDF's first page, so an A4 PDF (210 × 297 mm) gives an A4 document and a US Letter PDF gives 8.5 × 11 in. Each PDF page starts a new Word page, and Text only raises anything smaller than 9 pt to 9 pt.

What survives conversion, and what doesn't

Kept: all horizontal text, bold and italic, font sizes, and 15 common fonts matched by name, such as Arial (Helvetica becomes Arial), Times New Roman, Calibri and Georgia. Web addresses written out in full, starting http:// or https://, become clickable links. Keep the page layout also keeps alignment, indents and filled colour boxes.

Not kept:

• Pictures: logos, photos, signatures and stamps.
• Drawn lines: table borders, rules and box outlines.
• Text colour and underlining, so a red heading arrives in the default colour.
• Links behind words: "click here" loses its link, and so does an address with no http:// in front.
• Sideways or angled text, such as a vertical margin label.
• Word's own structure: headings aren't Heading styles, bullets are typed characters rather than Word lists, and headers, footers and page numbers become ordinary text on each page.
• Other fonts, which fall back to the default font.

Take an invoice with a logo, a Bill to block beside a Ship to block, a ruled table of items and a shaded total box. With Keep the page layout, the page becomes an invisible table: the address blocks sit side by side, each item gets a row, and the total keeps its fill. Columns that sit close together can share a cell, split by a tab. The logo and ruled lines need adding back.

How to convert a PDF to Word with our tool

1. Open PDF to Word and click Browse, or drop the PDF on the page.
2. If the file is protected, a PDF password box appears. Type the password and click Convert to Word again.
3. Choose a Layout: Keep the page layout (columns, tables, shading) or Text only (easiest to edit).
4. Click Convert to Word. A summary then shows the page count, the style and the file size.
5. Click Download Word file and open the .docx in your word processor.
6. Put it next to the PDF and check every page, especially figures and names.

The conversion runs in your browser, so the PDF and its password never leave your device.

Tidy the Word file

• Text cut off after you enlarge it: Keep the page layout fixes line spacing to match the PDF. Select the text and set line spacing to Single.
• Can't see what holds the layout: click in a table and choose View Gridlines on the table's Layout tab to show the invisible tables. To get ruled lines back, add borders to the table.
• A block you want as ordinary paragraphs: select the table and choose, on the same tab, Convert to Text.
• A table that came out as lines (Text only): select the rows, then Insert, Table, Convert Text to Table, and separate text at Tabs.
• Split words such as "docu- ment": words hyphenated at a line end keep the hyphen. Search for a hyphen followed by a space and fix each by hand.
• Paragraphs broken at page ends: turn on Show/Hide on the Home tab, double-click an unwanted page break, press Delete and join the two halves.
• No navigation pane: apply Heading 1 and Heading 2 styles to the headings.
• Missing logo: insert the original, or turn the page into an image with PDF to JPG and crop it.

When Word isn't the right target

• Tables of figures such as bank statements and price lists: use PDF to Excel. Amounts become real numbers you can add up, on one sheet per page or all in one. Numbers written with a decimal comma, such as 1.250,50, stay as text.
• Just the words, to quote, translate or count: PDF to Text is quicker, though it too can mix lines on two-column pages.
• A small fix to a file you must send back as a PDF: convert with Keep the page layout, edit, then save as PDF from your word processor. Our Word to PDF works without Word installed, but uses simple default fonts and layouts, so it suits plain documents.

Troubleshooting

• "This PDF doesn't contain any text we can read": it's a scan. Run PDF OCR first.
• "That password is wrong": check Caps Lock, or ask the sender. Neither this tool nor our password remover can get round a password you don't know.
• "This file isn't a valid PDF, or it's damaged": try PDF Repair & Checker, then convert the repaired copy.
• Strings of odd symbols instead of words: the PDF's fonts don't record which letter each shape stands for, so no converter can read them. Treat it like a scan: in PDF OCR, untick Skip pages that already have real text and use Copy text.
• A whole page is missing: pages with no text, such as one scanned page in a typed document, are skipped. Run the file through PDF OCR first.
• Lines from two columns mixed together: convert with Keep the page layout, then select and copy one column of the invisible table at a time.
• A single-column document full of awkward tables: convert again with Text only.

Spotted a mistake or something out of date? Tell us and we'll fix it.