PDF to Word: How to Keep More of the Formatting

Converting a Word document to PDF is a one-way trip that usually goes smoothly. Going back is harder, and the reason is not that converters are badly written. It is that most of what you think of as formatting is not stored in a PDF at all.

What a PDF actually remembers

A PDF records where each piece of text is drawn on the page: these characters, in this font, at these coordinates. It does not have to record that a group of lines is a paragraph, that a bold line is a heading, or that a grid of numbers is a table. Those structures existed in the document that produced the PDF, and most of them were discarded when it was exported.

So every PDF to Word converter is guessing. It looks at positions, sizes and gaps and infers where paragraphs start, which lines belong together and which text sits in a column. When the guesses are good the result looks effortless. When the layout is unusual, the guesses are what break.

Two kinds of converter, two kinds of result

Converters broadly take one of two approaches, and knowing which one you are using explains most surprises.

• Layout reconstruction tries to make the Word file look like the PDF. It often achieves this with positioned text boxes, frames and fixed line breaks. It looks right on screen, but editing is awkward: add a sentence and nothing reflows, because each block is pinned in place.
• Text reconstruction extracts the words in reading order and rebuilds them as ordinary paragraphs. It will not look like the original, but it behaves like a normal document. Type into it and the text flows the way you expect.

Neither is better in general. If you need to change a date on a one-page letter and send it back looking identical, layout reconstruction is closer to what you want. If you need to reuse the text of a report in a new document, text reconstruction saves you from fighting dozens of text boxes.

Decide what you actually need

Before converting anything, it is worth asking whether conversion is the right route at all.

1. 1If someone else created the PDF, ask them for the original Word file. It is the only version that still has the real styles, tables and tracked history, and asking is faster than any clean-up.
2. 2If you only need to reuse passages of text, extraction is enough, and plain text is easier to paste into your own template than a converted layout.
3. 3If you need to change a few words and keep the look, consider editing the PDF directly rather than converting it and converting it back.
4. 4If you need a fully editable document that looks like the original, expect to rebuild parts of it by hand, whichever converter you use.

Prepare the PDF before converting

• Check it has real text. Try to select a single word. If your selection draws a box over the whole page instead, it is a scanned image and there is nothing to convert until OCR has been run. OCR explained covers what that involves.
• Watch for mixed documents. A file can have typed pages and scanned pages side by side, such as a report with a signed page scanned in at the end. Scanned pages will come out empty even when the rest converts well.
• Convert only the pages you need. Extract Pages lets you pull out a section first, which means less to check and less to clean up.
• Remove the password. An encrypted PDF generally has to be unlocked before its text can be read.

What usually needs fixing afterwards

• Multi-column layouts. Newsletters, academic papers and brochures are the hardest case. If a converter reads straight across the page, lines from neighbouring columns end up joined together.
• Tables. Without a table structure in the PDF, the cells on one row collapse into a single line of text. Check every figure you intend to reuse, and see getting a table out of a PDF if the numbers are the reason you are converting at all.
• Headers, footers and page numbers. These are ordinary text on every page, so they tend to reappear as stray lines in the middle of your document.
• Hyphenated line endings. A word split across two lines in the PDF can come through as two fragments with a hyphen between them.
• Fonts. If the PDF used a typeface you do not have installed, Word substitutes another, and spacing changes with it.
• Images and links. Depending on the converter these may be dropped, and hyperlinks often arrive as plain text.

A quick way to check a conversion is to search the Word file for two or three distinctive phrases from different parts of the PDF, including one from a table and one from the last page. If any are missing or scrambled, you know where to look.

How PDF to Word on Formynex handles this

The PDF to Word tool on Formynex is a text reconstruction converter, and it runs in your browser. It reads the text layer already inside the PDF and writes a real .docx: text in reading order, lines grouped into paragraphs, larger lines detected as headings, and a page break at the end of each PDF page.

It does not attempt to reproduce the visual layout. Fonts, bold and italic, colours, images and table grids are not carried across, and because it reads each line across the full width of the page, a two-column layout comes out with the columns interleaved. For single-column reports, letters and articles, that trade usually works well: you get clean, editable paragraphs rather than a page of pinned text boxes.

If a PDF has almost no text layer, the tool stops and says so rather than handing you an empty document. OCR is not currently available on Formynex, so a scanned PDF needs to go through OCR elsewhere before it can be converted here.

If you plan to edit the Word file and turn it back into a PDF, remember that the return trip has its own pitfalls. Converting Word to PDF without losing formatting covers what to check on the way back.