Free PDF tools that work on your device

Try a real example before trusting a result

A successful download is only the beginning of a useful PDF workflow. The saved document needs the right pages, readable detail and the information you intended to keep. Our worked examples let you compare an original five-page workshop handout with two outputs made by SandboxPDF: a four-up print layout and a copy with document properties removed. Download the files, repeat the settings and check the difference in your own PDF reader.

The four-up example reduces five source pages to two landscape sheets. It also demonstrates a trade-off: small type becomes harder to read when several pages share a sheet. The metadata example removes the fictional author property while leaving the same name visible on the page. That distinction matters when preparing a document for someone outside your organisation. Removing a property is not the same as redacting a name.

All sample content is synthetic and can be reused for testing. The examples include settings, inspection steps and a machine-readable measurement record. They are controlled demonstrations, not a promise that every document will convert identically. Our editorial approach explains how we separate measured results from recommendations and document limitations. Use the guide below to choose your workflow, then compare the downloaded result with your original.

Start with the document you actually have

A PDF preserves the appearance of pages, but two files that look identical can contain very different information. A report exported from a word processor often contains selectable text, fonts and vector graphics. A photographed receipt usually contains a single image. That difference explains why one document converts cleanly and another needs text recognition first.

Try selecting a sentence in your PDF reader. If only the whole page behaves like an image, you may be working with a scan. If selection works but the copied words appear out of order, the PDF's internal reading order may differ from its visual layout. Choose a workflow based on that source, then check the output against the original. A different file extension alone cannot repair missing detail or recreate the source application's structure.

Put paperwork in order before sharing

For an application, expense claim or project handover, assemble the right pages before worrying about file size. Use Merge PDF to combine documents, then Organize PDF to check the sequence and remove duplicates. If only part of a report is relevant, Split PDF can create a smaller packet without resampling the page images.

Keep a separate original and use a filename that identifies the purpose of the new copy. Review the first page, last page and every join between source files: these are common places for duplicate covers or missing attachments. Check rotated pages at a readable zoom. Rearranging a signed document can invalidate an existing certificate signature, so preserve the signed original when authenticity matters.

Make a smaller PDF without sacrificing the useful details

Large scans usually owe their size to page images, not to the number of visible words. Compress PDF offers different trade-offs depending on the document. Start with a lossless option when preserving the source content is the priority. A document that is already efficiently encoded may not become noticeably smaller.

Image-based compression can reduce size more aggressively, but inspect small print, chart labels, stamps and faint handwriting afterwards. Repeatedly compressing the same lossy output compounds damage. Return to the original when trying a different quality setting, and compare the actual downloaded files. Our guide to why scanned PDFs are so large explains how resolution, colour and compression affect the result, including when recapturing a scan is the better solution.

Turn photographs into a PDF that fits

Use Images to PDF for JPG, PNG and WebP files such as receipts, diagrams or screenshots. Check the file list before exporting: each image becomes a page in that order. Automatic orientation matches each image individually, giving a wide diagram a landscape A4 sheet and a tall receipt a portrait sheet. Square images use portrait. Choose Portrait or Landscape instead when a whole packet needs the same page orientation.

Changing the sheet orientation does not rotate the photograph or crop its edges. The image is scaled proportionally and centred, so white space is normal when its proportions differ from the paper. Transparent areas receive a white background. Use Scan to PDF for document photos with optional grayscale and contrast. Neither tool restores blurred characters or automatically creates searchable text; use OCR when recognition is needed.

Recover text and tables with a realistic quality check

OCR PDF adds recognised text to scanned pages. Start with the correct language and a clear, upright scan. Automatic enhancement can improve some weak inputs, but enlarging pixels cannot recover detail the camera never captured. Compare names, dates, totals and identifiers directly with the page image. A misplaced decimal point matters more than a generally convincing paragraph.

For spreadsheets, PDF to Excel recognises scanned pages automatically and attempts to extract ruled tables into cells. It also includes page-image references for checking. Borderless tables, merged headings and complicated layouts may become text rows rather than a faithful editable grid. Original formulas are not recoverable from their printed values. Read PDF tables to spreadsheets before relying on converted data for calculations, and verify row and column assignments as well as individual numbers.

Protect information by choosing the right operation

Cropping a page or covering a word with an ordinary shape is not secure redaction. Content may still exist outside the visible area or underneath the overlay. Redact PDF rebuilds pages as images after masking the selected areas, discarding the original document objects. This also removes selectable text and interactive features. Reopen the exported PDF, inspect every sensitive area and try searching for removed terms before sharing it.

Sanitize PDF serves a different purpose: inspecting and removing document properties and other hidden information. Metadata cleanup alone does not erase a name printed on a page. Use the appropriate mode and review the tool's explanation of what it removes. The guide redaction is not a black box walks through the distinction between visual concealment and a rebuilt document, and the checks to perform on your final copy.

Use local AI with the source beside you

Chat with PDF helps locate relevant passages and answer questions about a document. Page references are a starting point for verification, not proof that an answer is complete. Check the cited page and neighbouring context, especially for exceptions, dates, financial amounts or obligations. If the source is a scan, recognition quality affects what the chat can find.

Summarize PDF selects key sentences rather than replacing a careful reading of the original. Local AI models download when needed and run on the device; the first run may take longer than later uses. Performance depends on the browser, available memory and document length. Our guide to checking AI PDF summaries explains how to spot missing qualifications, repeated headings and gaps in coverage before using a summary in your work.

Check the saved result, not just the preview

1. Keep the original file and choose a clear name for the new result.
2. Open the downloaded copy in a PDF reader and confirm the page count and order.
3. Zoom in on the smallest text, table values and any area you changed.
4. Test the property you need: search recognised text, copy converted values or inspect redacted areas.
5. For printing, check paper size and print one proof before sending a full batch.

A document can look correct as a thumbnail and still fail at the task you need it for. For example, a scan can be visually sharp without searchable text, while a spreadsheet can contain the right numbers in the wrong columns. The PDF field guides explain these distinctions with practical workflows. For paper handouts, start with the print preflight checklist or the N-up study handout guide.

Questions before you start

Are my documents uploaded for processing?

Document processing runs in your browser. Selecting a file does not upload its contents to a SandboxPDF processing server. The site still needs network connections to load application files and, for some tools, language packs or models. Contact messages and advertising services are separate from document processing; the privacy policy explains those interactions.

Is there a payment or a watermark at download?

The tools are free to use without an account. SandboxPDF does not automatically stamp its brand onto exported pages. A watermark you deliberately add with the watermark tool is part of your own edit. Save the downloaded result yourself; the workspace is not a cloud backup of your files.

Can I use the tools on my phone?

Many document tasks work in a modern mobile browser, but large scans, Office conversions and local AI can require more memory than a phone has available. Start with a smaller batch and keep the tab open while it processes. Read the mobile PDF workflow for file-picker, download and output-checking advice.

Will conversion reproduce every layout perfectly?

No. PDF describes the final page appearance, not necessarily its original paragraphs, spreadsheet formulas or slide objects. Complex layouts and scanned text require checking after conversion. Read capabilities and limitations to choose an appropriate tool and understand the trade-offs before relying on a result.