What converting to Word can and cannot do
It is worth being honest about this one, because it is the conversion most likely to disappoint. A PDF describes where marks sit on a page. A Word document describes a flowing structure of paragraphs, styles and headings. Going from the first to the second means inferring structure that the PDF does not record — and inference is sometimes wrong.
For a straightforward text document the result is usually good: paragraphs come across, the text is editable, and you can get on with your edit. For a document with a complex multi-column layout, tables, sidebars or precise typography, expect to spend time fixing the result. That is not a defect in any particular converter; it is the nature of the problem.
There is one hard limit worth knowing before you start. If the PDF is a scan — a photograph of a page rather than text — there is no text in the file to extract, and a converter will produce an empty or nearly empty document. That is what OCR is for, and it is a separate step.
When to convert and when to do something else
Convert when you genuinely need to edit the words: revising a contract you no longer have the source of, updating a CV whose original is long gone, reusing text from a report. In those cases some layout cleanup is a fair price for getting editable text.
Do not convert when you only need to read the document, when you only need a few sentences, or when you need the layout preserved exactly. If you want to quote from it, extracting the text is faster and cleaner. If you want to fill in a form, filling the PDF form directly keeps the layout intact and produces something that looks like the original rather than an approximation of it.
And if the aim is simply to make small changes to how the document looks — adding a page number, stamping a watermark, signing it — do that to the PDF. Round-tripping through Word to make a cosmetic change tends to cost you more in layout damage than the change was worth.
Getting the cleanest possible result
Expect to fix things in a predictable order. Check the paragraph breaks first, because line-by-line breaks instead of proper paragraphs are the most common artefact and the most annoying to live with. Then look at tables, which frequently arrive as text laid out to look like a table rather than as a real one. Then fonts, which will substitute if the original was not a font you have installed.
If the layout comes out badly enough to be unusable, extracting the plain text and repasting it into a clean document is often faster than repairing the conversion. You lose the formatting, but you were going to lose most of it anyway, and you start from something predictable.
Finally, proofread rather than skim. Converters occasionally merge or drop a character in unusual fonts, and the errors they produce tend to be plausible-looking rather than obvious — exactly the kind that survive a quick read and embarrass you later.
Why the free converters that want your email are worth avoiding
Search for a PDF to Word converter and most of the results will ask you to upload your file and then hand over an email address to receive it. That pattern exists because the file and the address together are worth more than the conversion costs to run. The documents people convert to Word are contracts, CVs, reports and letters — among the most revealing things on anyone's hard drive.
There is a simpler objection too: it is slow. Upload, wait in a queue, check your inbox, download. Converting in your browser skips all of it, works without a connection once the page has loaded, and does not leave a copy of your contract on a server whose retention policy you have not read.