PDF problems, explained properly
Each of these was written because people keep asking it and the answers are scattered across fifteen-year-old forum threads. Every post is meant to be useful whether or not you use our tools — and each says plainly what the tool will not do.
Which merge keeps your PDF bookmarks?
The outline is a tree of destinations pointing at page objects. Copying pages copies neither, so the navigation panel comes back empty on the far side.
Split a PDF that is too big to even open
The usual advice is to compress it first, but nothing can compress a document it cannot render. Here is the order to use when nothing will load it at all.
White bars round a PDF page on your slide
A4 is 1:1.41 and a widescreen deck is 16:9, so the shapes cannot agree. Trimming the margins first closes the gap; rescaling afterwards does not.
Is flattening the same as printing to PDF?
One rewrites fields and annotations into page content; the other re-renders everything as a picture. What each one costs you: text, links, file size.
Does renaming a PDF change what is inside?
No — unless the tool also writes the new name into the document title. That one difference decides whether a signature still verifies afterwards.
Put four photos on one PDF page
Most image converters give every picture a sheet of its own, which is the opposite of a contact sheet. Two steps get you a 2x2 grid instead.
Can a screen reader read a PDF turned into slides?
Each page arrives as a flat picture with nothing behind it, so assistive software finds no words at all. Where the text has to come from instead.
Does flattening a PDF kill its hyperlinks?
The blue underline survives; the click usually does not. Link annotations and field widgets are different objects, and tools treat them differently.
Strip the repeated header from extracted text
PDFs carry no idea of a running title unless the producer marked that region as an artifact. Filter by position, or by lines common to every sheet.
You closed the tab before saving your edits
A browser tool holds the document in memory, so there is nothing on disk to recover. What is worth trying, and how to make it impossible next time.
Which pages of your PDF actually need OCR?
A born-digital report with three scanned exhibits stapled in only wants work on three. How to find them without reading two hundred sheets yourself.
Does renaming a signed PDF break the signature?
The file name sits outside the byte range the certificate covers, so an operating system rename is harmless. Writing that name into metadata is not.
Get the text of just pages 10 to 20
Two jobs are easily confused: lifting a range out as its own document, and lifting the words out as a .txt file. Which one you want, and in what order.
What reordering PDF pages quietly breaks
Outline destinations, form fields, annotations and named links all point at objects that moved. Some tools re-point them; most never say either way.
Split a payslip print run into one file each
A 400-page export where an employee spans two sheets defeats any tool cutting every N pages. Match on the words instead, and name the outputs as you go.
Why your photo lands sideways in the PDF
Phone cameras store the picture as the sensor saw it and add an orientation tag. Browsers honour that tag, most converters ignore it, and the tilt sticks.
Why typing in a PDF changes the font mid-line
An embedded face is usually cut down to the glyphs the document already used. Ask for a character it never carried and the editor has to substitute.
Deskew before OCR, or after?
Straighten first: a recognition engine works along lines of type, and a tilted line costs it accuracy. The whole scan pipeline, in the order that works.
How to edit a PDF someone emailed you
It opened in a preview pane that cannot save, and it may be a scan with no words in it, or permission-locked. Work through the four checks in order.
Your HTML-to-PDF file came out blank
Three usual causes: a page drawn by JavaScript after the capture, a login wall returning a sign-in screen, and a print stylesheet hiding the body.
Why an edit made your scanned PDF bigger
Straightening, redacting and sanitising all re-render the page, so a tightly packed JPEG comes back as a fresh, fatter image. What to run afterwards.
Opens in Chrome, refused by Acrobat
Lenient viewers rebuild a broken cross-reference table on the fly; strict ones reject the file outright. The damage is real whichever one you believe.
Scan both sides on a single-sided scanner
Two passes, and the second one comes back reversed. The interleave that puts fronts and backs in order, plus what to do about an odd sheet count.
Extract pages without losing the form fields
Widgets live on the page, but the tree naming them and holding their values lives on the document, so half the structure is left behind. What survives.
Why did compressing my PDF make it bigger?
Text streams are already deflated, re-encoding a JPEG adds bytes, tagging gets rebuilt, and an incremental save appends rather than replaces.
Can you compress a password-protected PDF?
Which of the two kinds you have decides it. An owner restriction lifts off and the file re-saves cleanly; a user one blocks every tool until it is typed.
Why does your scan keep coming out crooked?
Dirty feed rollers, edge guides left wide, a bound spine curving at the gutter — and phone perspective, which no amount of straightening will correct.
Is it legal to remove a PDF password?
Knowing the owner code and lifting restrictions is a different act from circumventing one you were never given. Where the line usually sits, and why.
Why your repaired PDF is a different size
A rebuild keeps only what is still reachable, so stale revisions and orphaned objects go. Smaller is normal, and here is when growth is normal too.
Why won’t this tool accept my .docx file?
Converters that refuse a .docx are usually not broken. Here are the reasons a file gets rejected, in the order worth checking, and what to do about each.
Why does my PDF ask for a password every time I open it?
There are two passwords in a PDF and they do opposite things. Which one you have decides whether this takes ten seconds or is impossible.
The tool deleted the wrong page. Here’s why.
A PDF has three different page numbers and they rarely agree: the position in the file, the label the document shows, and the number printed on the paper.
Why doesn’t my converted PDF have page numbers?
Word shows page numbers on screen that are not in the document at all. Whether they survive a conversion depends on where they were put in the first place.
Why does my old PDF title still show up after I changed it?
You edited the title, saved, and the old one is still in the browser tab. A PDF stores its title in two places, and most tools only change one.
Why flattening broke my PDF’s digital signature
A signature is a promise that the bytes have not changed. Flattening changes them, which is not a bug in either feature — but the order you do things in matters.
Why is my JPG-to-PDF file so much bigger than the photos?
Twelve photos of 400 KB became a 30 MB PDF. The photos did not grow — something in the conversion decided to store them differently.
Locking a filled-in PDF form so the answers can’t be changed
Flattening makes the answers part of the page, so nobody can retype them. It is not encryption, and knowing the difference matters before you send it.
Why does my PDF show “Microsoft Word - Draft3” as its title?
The browser tab shows a filename you thought you had left behind years ago. It is stored inside the file, and it is visible to everyone you send it to.
Fewer files than bookmarks? Here’s what happened.
You split by bookmarks and got six files from twenty bookmarks. The tool did what you asked — the outline just does not mean what it looks like it means.
You probably don’t need Word: edit the PDF itself
Converting to Word to change three words is a round trip that ruins the layout. Editing the PDF directly keeps everything else exactly where it was.
My PDF file is 0 bytes — what now?
A zero-byte file contains nothing at all, so nothing can repair it. Here is what actually happened, and the places the real file may still exist.
Put a PDF page on a slide without it going blurry
PowerPoint stores what you paste at the size you pasted it. Getting a crisp page onto a slide is about resolution before the paste, not after.
What “created with” and “produced by” reveal about your PDF
Two fields nobody sets on purpose, both visible to anyone who opens the properties. They say which program made the file, and often which version.
Does straightening a scanned PDF make the text unselectable?
Sometimes, and it depends entirely on the order you do things in. Straighten before you OCR, and you keep both the level page and the searchable text.
Does clearing a PDF’s metadata make it anonymous?
No — and the gap between what metadata removal does and what people expect it to do is where real disclosures happen. Here is what stays in the file.
Why don’t the images show up in my HTML-to-PDF file?
The converter fetched your HTML and nothing else. Whether an image arrives depends on how it was referenced and whether the renderer could reach it.
Split a scanned book’s two-page spreads into single pages
A book scanned two pages to a sheet will not read on a tablet. Here is the whole recipe: crop twice, interleave, check the order, and when to use a desktop tool.
“Split a PDF in half” means two different things
Half the document, or half of every sheet? They need opposite tools, and picking the wrong one wastes an afternoon. Here is how to tell which one you meant.
Why the font changes when you edit text in a PDF
A PDF carries a cut-down copy of its fonts, so most editors cannot type in them and substitute instead. Here is what actually happens, and how to avoid it.
Why is my PDF text not selectable?
Three different causes, and they need three different fixes. Work out which one you have in about thirty seconds, before you pay for anything.
You chose greyscale and it still printed in colour
Choosing greyscale in the print dialog does not change the file, and some printers bill by what the page contains. Here is how to convert the document itself.
Converting a Markdown README to a PDF
What survives the trip and what does not: headings and tables come through, code blocks keep their line breaks, and badges and diagrams do not travel at all.
Merge odd and even page PDFs back into one document
A single-sided scanner leaves you two files: fronts and backs. Interleaving them is one pass — including when the backs came out in reverse.
Your bookmarks vanished when you merged the PDFs
Most merge tools drop the outline because keeping it means rewriting every destination. Here is why, and how to end up with a navigable document anyway.
Get a table out of a PDF and into Excel
There are four real routes and they are not equally good. Which one to use depends on how the table was made — and whether the PDF has text in it at all.
Get rid of the black edges on a scanned PDF
The black border is the scanner lid, not damage. Cropping removes it from view in seconds — and there is one thing to know before you send the file on.
Converting a scanned PDF to Word needs OCR first
A scan is a picture of words, not words. Any converter you point at it will produce a document containing one large image — here is the order that actually works.
Why the images you pull out of a PDF come out blurry
Usually they are not blurry at all — you are looking at what was actually stored. Here is how to tell the difference, and what to do in each case.
Delete the blank pages a duplex scan left behind
Every second page is the back of a single-sided sheet. Finding them is the real problem, because a “blank” scan is never actually blank.
Remove the form fields but keep what was typed in
Flattening turns the answers into part of the page, so nobody can change them and every reader sees the same thing. It is not the same as deleting the fields.
Keep one Bates sequence running across several PDFs
A production numbered per file is not a production. Here is how to run one continuous sequence across a set, and what to do when a document is added late.
Removing a password from a PDF you can already open
If it opens when you type the password, the password can come off in one pass. If it does not open, no tool on the web can help — and here is why.
Split a PDF into single pages, one file per page
One file per page, named so they sort correctly — including the padding detail that stops page 10 from filing itself between page 1 and page 2.
Add page numbers to a PDF but leave the cover unnumbered
Start the numbering on page 2 while page 2 still says “1” — the distinction between which pages are stamped and what number they are given.
Split a PDF into chapters using its own bookmarks
If the document has an outline, it already knows where the chapters start. No page numbers to work out, and the file names come from the headings.
Why won’t my PDF get any smaller when I compress it?
Compression works on images. A text-only PDF is already small, and squeezing it returns the original — here is how to tell what your file is made of.