Think about what people actually merge and split. Signed contracts and their appendices. Bank statements for a mortgage application. Payslips, medical letters, tax returns, scans of a passport. PDF operations cluster around exactly the documents you would least like to leave on someone else's server, which is why the processing location is not an abstract concern here.
Splitting is the clearest case. The usual reason to split a document is that someone asked for one page and the other pages contain things they have no business seeing. Uploading the whole file to extract that page defeats the purpose of extracting it. Doing it locally does not.
One caution regardless of where processing happens: removing a page is not redaction. Text hidden behind a black box drawn in another application is still text in the file, and metadata can travel with the pages you keep. If something genuinely must be unreadable, use a proper PDF editor's redaction feature and check the output before sending it.
Browser-side processing is not strictly better, and the last row is the honest cost of it. Here is the trade in full.
| Typical server-based tool | Fileraz | |
|---|---|---|
| Where your PDFs are processed | Uploaded to the operator's servers | In your browser tab, on your own device |
| What the operator can see | The file contents, while processing and in any retention window | Nothing — the file is never transmitted |
| Usage limits | Commonly a daily cap or a size cap, with a paid tier above it | No cap, because your conversions cost us nothing to run |
| Account required | Often, for anything beyond the basics | Never — there is no public sign-up at all |
| Speed | Upload time, queue time, then download time | Starts immediately; bounded by your own hardware |
| Large or complex files | Handled well — server hardware does the heavy lifting | Bounded by your device memory, and may fail where a server would not |
Worth knowing before you start, rather than after.
