I now think twice before uploading contracts to a free PDF site

PDFs tend to become urgent all at once. A school needs a transcript, a client wants one combined quotation, or accounts says the attachment must be under 10 MB. Then someone asks for the scan as an editable Word file.

It is easy to open a familiar search result, select the document and wait. Only after downloading do I wonder where that stamped contract went, and whether a server still has a copy. I am not assuming every service misuses files. I just prefer not to hand over a whole document for a small edit when my browser can do it.

Start with the actual job

Usually it is compression, combining files, removing pages or fixing their order. DocCrunch has separate tools, including PDF Compression and PDF Merge, without requiring an account.

Compression accepts up to 10 PDFs, each under 500 MB, with Clear, Balanced and Small settings. Those are selection limits, not a promise that an ordinary laptop will handle ten near-limit files comfortably. Try one first.

A smaller file can behave differently

Text-heavy PDFs may have little room to shrink. Scans and image-heavy documents can offer more, but there is no guarantee of getting below an email limit.

The tool compares a structurally optimized PDF with a version rebuilt from page images and chooses the smaller result. If neither beats the original, it keeps the original file. If the image-based version wins, selectable and searchable text may be lost. Links and forms should not be assumed to survive either.

After compression I inspect small print and stamps, try selecting text, and check any links I need. If Small blurs the page, I step back. I keep the original and send a copy; print-ready or digitally signed originals should not simply be replaced by a processed version.

Put the right pages in the right order

PDF Merge follows the file list; use the up and down buttons to arrange it. PDF Split creates one PDF per page and packages them in a ZIP. To remove a few unwanted pages, Delete PDF Pages is often quicker. Use PDF Rotate for sideways scans and Reorder PDF Pages for a misplaced cover.

Deleting whole pages is different from redacting an account number on a page. Cropping something out of view or covering it with a shape is not proof that its underlying content has been removed.

Word output is not a finished document

For a PDF with an existing text layer, PDF to Word and PDF to Excel extract that text and arrange it by page and position. They do not faithfully reconstruct every layout. Columns and complicated tables need checking.

These PDF converters do not currently run OCR on image-only scans. Export the required pages with PDF to Images, then use Image to Word or Image to Excel for OCR. A scan that already has an OCR text layer can be tried directly.

Stamps over letters, glare and tilted pages make recognition harder. Compare amounts, account numbers, names and dates with the original. Avoiding retyping is useful enough; it does not remove proofreading.

PDF to PPT places a rendered image of each page on a slide. It is handy for presenting handouts, but headings and charts do not become individually editable objects.

Local has limits too

These operations run in your browser without uploading the source or result to a processing server. The page, scripts and OCR resources may still need downloading. Large files use memory, encrypted files may fail, and signatures, interactive forms or precise print colour need tools that can verify those requirements. Shared computers and browser extensions remain separate concerns.

I keep the original, fix only what the recipient needs, and extract text before compression if it matters. Phone photos of receipts can go through Image to PDF before merging with a statement. The previous article on local photo tools covers reducing large images; keep small figures readable.

For an oversized attachment, try compressing one PDF. Open the downloaded copy and check it before sending.