Why We Built fastpdfnow in Lima (and How It Works Under the Hood)

The problem we kept running into

My mother trying to shrink a certificate for a government portal that kept rejecting the file. A university classmate with five PDF appendices that had to be submitted as one. An office assistant I know who printed contracts, signed them with a pen, scanned them and emailed them back, because there was no other way to "sign". People without a printer at home, doing paperwork from their phones, with no time to learn a new program.

They all did the same thing: search for "merge PDF free" and land on the first result. And that is where the second fight began: daily file limits, watermarks stamped on the output, "create an account to download", buttons that said free but led to a Pro plan, and the uncomfortable question of which server had just received a copy of their ID.

From my work and my studies I knew none of that was technically necessary. Merging, splitting, rotating or compressing a PDF are operations an ordinary laptop or phone can handle on its own. So the question became simple: why did nothing exist that worked that way, without the hurdles, in Spanish, and built around the kind of paperwork people actually deal with here?

The core decision: everything happens in your browser

The first decision, and the one that shapes everything else, was that fastpdfnow would have no processing server. When you open any tool on the site, your browser downloads the JavaScript it needs and from then on works alone. You load a file, the browser opens it in your computer's or phone's memory, performs the operation and hands you the result. Your file never travels across the internet.

To make that work we rely on open-source libraries the community has maintained for years. pdf-lib handles creating and modifying PDFs: merging pages, reordering them, rotating, placing a signature image, writing page numbers. pdf.js, the engine Firefox uses to display PDFs, lets us render pages on screen, generate the thumbnails you see when reordering, and convert pages to images when compression requires it. We did not reinvent the wheel; we assembled it.

This decision has knock-on effects. There is no server storing files, so nothing can leak and no employee can peek. There is no storage or processing cost per user, so there is no reason to charge or to cap how many files you process a day. And there is no need to identify you, so there are no accounts.

No accounts, no artificial limits, free

Many PDF tools impose limits not because the technology demands it but because the business model does: two free files a day, the third one costs. Since fastpdfnow spends no server resources per file, those limits make no sense. Merge fifty PDFs in a row if you want.

We do not ask for an account either. An account would only be useful to store your files on our side (which is precisely what we do not want) or to send you marketing emails (which we do not want either). You come in, do your thing, download and leave. And it is free across every tool: merge (/merge), split (/split), sign (/sign), compress to a target size (/compress), images to PDF (/images-to-pdf), watermark (/watermark), page numbers (/page-numbers) and rotate (/rotate). There is no hidden Pro tier.

How the site pays for itself: ads, no tricks

A website does not run itself. There is the domain, the hosting (lightweight, since it processes no files), and above all the development time, which comes out of my evenings and weekends. To cover that, fastpdfnow shows ads through Google AdSense.

We want to be upfront about what that means. The ads are the only part of the site that talks to external servers; your files have nothing to do with them and Google never sees them. Ads are placed where they do not get in the way: you will not find a fake "download" button that is actually an ad, no pop-ups, no countdown timers. If you ever see something like that, it is a bug and we want to hear about it.

We prefer this model to charging because someone who needs to compress a certificate for a portal needs it once every few months. Nobody should pay a subscription for that.

Honest limitations: what we do not do well (yet)

Processing everything in the browser has a cost, and we would rather tell you now than have you discover it with an important file.

First, very large files depend on your device's memory. A 400-page scanned PDF on a phone with little RAM can get sluggish or crash the tab. On an ordinary laptop it rarely happens, but it does happen.

Second, the signing tool (/sign) places an image of your signature, drawn or uploaded, onto the page. That is a simple electronic signature. It is not a certificate-based digital signature: in Peru, Law No. 27269 regulates certificate-based digital signatures, which are a different thing. For many everyday procedures, private contracts and internal documents an inserted signature is enough, but whether it is accepted depends on the institution or the other party. We will not promise legal validity we cannot guarantee.

Third, heavy compression turns pages into images. When you ask for a PDF to come down to a very small size, the way to get there is to render each page and save it as a lower-resolution image. The result weighs what you asked for, but the text is no longer selectable and looks less sharp. For a receipt you upload to a portal that is fine; for a thesis going through a plagiarism checker, it is not. That is why we recommend compressing only down to the size you actually need.

What to do if something goes wrong

Because the processing happens in your browser, most problems can be fixed on your end, and quickly.

• The tab crashes or freezes with a large file: close other tabs to free up memory, or split the file into parts with /split.
• The PDF will not load: it may be password-protected or corrupted. Open it first in a regular reader; if it fails there too, the file is broken at the source.
• The tool does not respond: fastpdfnow needs a reasonably current browser (recent Chrome, Edge, Firefox, Safari or Brave). It does not work in Internet Explorer.
• None of the above helps: write to fastpdfnow@gmail.com with the tool's name, your browser and what happened. Do not send the file if it is sensitive; a description is usually enough.

What comes next

The current list of tools came from the problems we saw most often. The next list is coming from you. People have asked for PDF to image conversion, password protection, a more visual way to reorder pages, and text extraction from scanned documents. Some of these are straightforward to do in the browser; others, like text recognition, are heavier and we want to get them right before releasing them. The rule for deciding is the same as always: if it solves a real piece of paperwork someone had to deal with this week, and it can be done without the file leaving your device, it goes on the list.

If you have a suggestion, a problem, or just want to tell us what you used the tool for, write to fastpdfnow@gmail.com. I read every email. And if fastpdfnow saved you a trip to the print shop, a queue or a night of frustration, that is exactly what we built it for.

Common questions