What the engine actually does

Pdfmira’s PDF to Word path is an office conversion, not a magazine engine. Text-heavy papers often survive well enough to re-edit two paragraphs. Illustrated covers, baseline grids, wrap-arounds and placed plates do not. Fonts on your machine are not the fonts in the lab.

This is not the suite that invented the PDF, and it does not pretend to be. Libre-style office conversion plus heuristics. If millimetre fidelity is the brief, you already have the wrong job.

When a creative should not press the button

A lookbook, a thesis with a designed cover, a poster PDF: export pages to JPG or PNG and rebuild, or hand in the PDF. Pull embedded photos with Extract images if you own them. Do not launder a grid through Word and hope the folio comes back.

Scans without a text layer land as pictures inside the DOCX. OCR can make words searchable first; it still will not restore columns. OCR and Office are capped on the free sheet because they are heavy. That is a limit, not a teaser.

The other direction is also a lab

Word to PDF closes a paper for a tutor who refuses DOCX. Embed fonts in the source if that serif on the title matters. Otherwise the lab substitutes. Images that were already weak in the DOCX stay weak. Photographic plates belong on JPG to PDF, then merge — not inside a word processor pretending to be a press.

Laboratory spreads are listed in the index so you can find them. They are not the cover story of Pdfmira. The cover story is plates, contact sheets, and previews you can judge with your eyes.