Every web PDF editor I tried, Smallpdf, iLovePDF, Acrobat online, does the same thing: drops an invisible text box on top of the page. You're not editing the document, you're sticking a note over it. The second your text runs longer than the original it overflows or gets cut. Tables are the worst, add a row and everything below just breaks.
So I rebuilt the import so a PDF comes back as actual semantic HTML. Real tables, rows, cells, not absolutely-positioned divs. Then a rendering engine (PHP, 3300+ tests, most of CSS including flexbox and grid) lays it back to PDF. Because it's real HTML/CSS underneath, content reflows: edit a cell, the table restructures, and if it runs long it flows onto the next page like a normal document.
The bit I'm oddly proud of: the editable HTML source gets AES-encrypted and embedded into the exported PDF's metadata. So the file is its own source. Reopen it later and you're editing the original structure losslessly, no account, no cloud, no reconversion.
The closest thing that does real reflow is Apryse, but it's a $10k+/yr server-side SDK you drive with code, built for enterprise batch pipelines. There's no "open it in a browser and edit a table live" version. That's the gap I'm filling, cross-page structural reflow in a consumer tool you just open and use.
There's a no-signup demo if you just want a quick look (watermarked, capped at 3 pages), and a genuinely generous free tier without those limits. Open a PDF, edit a table, watch it reflow across pages, export it, reopen it, keep editing. There's a bug report button right in the tool too: if something breaks and you want to send it over, I'll look into your specific case myself and fix it, it's just me on the other end. Docs are at reflowpdf.com/docs if you want to go deep.
What I'm actually trying to figure out: if you work with docs that have to restructure, translators dealing with text expansion, anyone regenerating invoices or reports, what would make a tool like this worth paying for over just rebuilding it in Google Docs?
The interesting part to me isn't the PDF editing itself.
It's that you've changed the cost of modifying a document after it's already become a PDF.
I suspect the biggest opportunities might come from people who currently accept painful workflows because rebuilding the document feels even worse.
yeah, that's exactly it. nobody wakes up wanting to "edit a PDF", they want the thing the PDF represents to be different, and right now the only options are recreate it from scratch or fight with an overlay tool that breaks the layout.
the painful-workflow angle is spot on. the people I keep running into are translators (a doc gets 30% longer in German and the whole table layout dies) and anyone dealing with invoices/reports they didn't author. they've all just accepted that "received PDF = locked PDF" and budget hours for rebuilding.
curious if you've seen this somewhere specific — you described the problem like someone who's lived it.
Possibly, but that's exactly why I stopped short earlier.
I don't think the interesting part is the examples themselves.
I think it's the decision that follows from them.
I'd be careful unpacking that casually in a thread.
If you'd like the tighter version, drop your email and I'll put it together properly.