I Stopped Using Desktop PDF Apps for a Month — Here's What Happened
The Dare I Made to Myself
It started with a licensing renewal notice. My PDF editor wanted $179 for another year, and I sat there staring at the invoice wondering when I'd last used more than three of its features. Merge, compress, and maybe — maybe — annotate. That's it. So instead of clicking "renew," I closed the tab and made myself a deal: one month, browser tools only. No Adobe Acrobat. No Foxit. No installed PDF software of any kind.
I work as a freelance proposal writer, so PDFs aren't optional for me. They're basically the product. I was terrified, honestly. But also a little curious. And cheap enough to try.
Week One: Surprisingly Fine (Mostly)
The first thing I needed to do was merge eight separate quote PDFs from subcontractors into one clean deliverable. I'd done this in Acrobat for years, dragging thumbnails around until the order looked right. I pulled up Smallpdf's merge tool instead, dropped the files in, shuffled the pages in the browser, and downloaded the result in about four minutes. It looked perfect. Same thing, different window.
Compressing files was equally painless. I'd been compressing PDFs before emailing them for so long it was muscle memory — Acrobat's "Reduce File Size," same result every time. ILovePDF's compressor gave me comparable results, and for files under 15MB, I genuinely couldn't tell the difference in the output. The quality wasn't degraded in any way that mattered for a business proposal.
The first frustration hit around day five. I needed to extract specific pages from a 62-page contract — pages 4, 7, 11-13, and 44 — and every browser tool I tried either made me extract a range or download each page individually. One tool let me pick non-contiguous pages but then buried that feature behind a paywall. I ended up doing two separate extractions and merging the results. Solved, but annoying. With Acrobat I'd have clicked four checkboxes and been done.
The Annotation Situation
Here's where things got genuinely interesting, not because it was bad, but because it forced me to rethink my workflow.
I review contracts with a lot of tracked comments — "clarify this clause," "increase this number," that kind of thing. In Acrobat, I'd highlight, add a sticky note, initial things with a text box. The experience felt document-native. In the browser, I tried several tools. Adobe Acrobat Online (yes, the free browser version of the thing I'd paid for) was the most capable. PDF.js Express was a decent runner-up. Both let me highlight and add comment bubbles, but saving felt one extra step removed from reality — download, reupload, download again when done.
What surprised me: my clients actually received these annotated PDFs fine. The annotations rendered perfectly in their readers. The friction was all on my end, in the process, not in the output quality. That's a meaningful distinction.
I also discovered that for simple read-only markup — just highlighting passages to reference later — my browser's built-in PDF viewer with a basic highlight extension was completely sufficient. I'd been overpaying for infrastructure I only half-needed.
Where It Actually Broke Down
Week two had a rough patch. A client sent me a scanned PDF — physical pages photographed and bundled, no selectable text — and asked me to copy-paste several paragraphs into a different document. This is an OCR (optical character recognition) job. Most browser tools either don't do it or do it badly behind a paywall.
I tried three free browser tools. The first just returned the original unprocessed file. The second produced text that looked like someone had sneezed on a keyboard — "Th3 p4rty of th£ firft p4rt" — useless. The third, a tool called PDF2Doc, gave me surprisingly readable output but scrambled the table formatting completely.
I ended up using Google Drive's built-in trick: upload the PDF, right-click, open with Google Docs. It's not perfect OCR, but for clean scans it does a respectable job. I fixed maybe thirty small errors manually and it took fifteen minutes total. Not ideal. But not impossible either.
The other genuine breaking point was form creation. I had to build a fillable intake form for a new client — text fields, dropdowns, checkboxes. Browser tools for this are, to put it generously, limited. The free options are almost non-functional. The paid options are basically just subscription versions of the desktop apps. I ended up creating the form in Google Docs, exporting it, and calling it a day with non-fillable fields. The client didn't complain, but I knew it was a workaround, not a solution.
The Privacy Question Nobody Talks About
Around week three, I started thinking about something I'd ignored in my initial excitement: these browser tools are uploading my documents to someone's server. Sometimes sensitive ones. A contract with payment terms. A proposal with a client's business figures.
Most reputable tools — Smallpdf, ILovePDF — claim files are deleted within a few hours and use encryption in transit. That's probably fine for most documents. But I found myself hesitating before uploading anything that felt genuinely confidential. That hesitation cost me time and mental energy. With local software, I never had to think about it.
I'm not saying browser tools are insecure. I'm saying the question exists in a way it simply doesn't with local apps, and that cognitive overhead is real.
The Unexpected Wins
Here's what I didn't expect: browser tools pushed me toward better habits in a few ways.
Cross-device freedom. I split my time between a laptop and a desktop depending on where I'm working. Without desktop software to install and license on both machines, I just worked. Open Chrome, go to the site, do the thing. No activation headaches, no "you've reached your device limit" messages.
No bloat creep. Desktop PDF apps — especially the big ones — have a habit of installing update daemons, adding themselves to startup, and generally making your computer heavier over time. Uninstalling Acrobat felt like evicting a hoarder. My machine ran noticeably cleaner.
Forced simplicity. When a browser tool can't do something complex, it made me ask whether I actually needed that complexity. Half the time, I didn't. I was using features because they were there, not because the job required them.
Where I Landed After Thirty Days
At the end of the month, I added up what I'd actually needed a desktop app for:
- OCR on scanned documents (more than once)
- Building fillable PDF forms from scratch
- Batch processing — watermarking or compressing twenty files at once without uploading each one individually
- Editing the actual text inside an existing PDF (not adding annotations, but genuinely changing what it says)
That's a shorter list than I expected. For everything else — merging, splitting, basic annotation, compression, format conversion — browser tools handled it with maybe 10-15% more friction than desktop software, mostly in the form of extra clicks and file downloads.
I didn't renew my Acrobat subscription. But I also didn't stay 100% browser-based. I installed PDF24, which is free, open-source, runs locally, and covers almost everything I missed. Lighter than Acrobat, no subscription, and it handles batch processing and OCR competently enough for my volume of work.
If you work with PDFs occasionally — printing, reading, occasionally merging a few files before sending — you genuinely don't need a desktop app in 2024. Browser tools have gotten good enough that the subscription cost is hard to justify. But if PDFs are core to how you work, not something you touch twice a week but something you're in and out of all day, the desktop experience still has real advantages: speed, privacy, batch workflows, and the ability to work offline without thinking about it.
The month changed how I think about software defaults. I'd been paying for Acrobat out of habit, not need. That's a bad reason to spend money on anything. The experiment cost me a few frustrating afternoons. It also saved me $179 and made me realize how much of my daily PDF work could run fine in a browser tab while I kept my coffee warm.
Sometimes the best software audit is the one where you just quietly stop using something and see what you miss.