IsonAIMost everyday PDF needs can be done for free—as long as you use the right method for each task. Filling forms and signing need only a free PDF reader; changing original text needs an editor with object editing; a scan must be OCR’d first before it can be searched or edited; and hiding data requires true redaction—not a black box painted over the text. This guide maps tasks to tools, gives desktop and mobile steps, then shows how to verify the result.
Reviewed as of July 2026. Each app’s capabilities and free-tier terms change often; this guide is deliberately tool-agnostic and does not state a free limit without a source—always check the current terms in the app you choose.
Match the task to the method
| Your task | The right method |
|---|---|
| Fill a form / add text on top of a document | Fill & Sign or annotation in a PDF reader |
| Change existing original text | A PDF editor with object/text editing (not just annotation) |
| A scanned PDF (an image; text can’t be selected) | OCR first, then edit/search |
| Merge, split, reorder pages | A page organiser |
| Hide sensitive data | True redaction that removes the text—not a black box |
| Sign | Distinguish a signature image (approval) vs a certificate-based digital signature |
Fill a form
Desktop: open the PDF in a free reader (e.g. your system’s built-in reader or Acrobat Reader). If there are interactive fields, click and type directly. If there are no fields, use Fill & Sign to add text boxes on top of the document. Save as a new copy.
Mobile: open via the Files app (iOS: the Markup feature) or a free PDF app. Tap a field to type, or add text via annotation. Correct output: your text sits neatly on the field lines. Failure sign: text drifts off the field lines—resize/reposition the text box.
Add a signature
Desktop/mobile: use Fill & Sign, then draw a signature, type a name, or paste a photo of your signature. This is a signature image (a visual approval), not a certificate-based digital signature.
Need stronger legal weight? A digital signature uses a certificate from a trusted provider and can verify that the document hasn’t changed since signing. For official documents, consider a certified electronic signature from a recognised provider. Important: editing a PDF after it has been digitally signed will invalidate that signature.
Edit original text
Desktop: you need an editor with object editing (e.g. paid Acrobat Pro, or the free, offline LibreOffice Draw). Open the file, click a text block, and edit. Failure sign: the text can’t be selected at all → likely a scanned PDF (needs OCR) or the font isn’t embedded, so the layout shifts when edited.
Mobile: editing original text on a phone is limited; for larger changes, work on desktop or via an assistant that can edit document contents. Always check that fonts and alignment stay consistent after editing.
OCR a scan
A scanned PDF is really an image; its text can’t be selected or searched until it’s OCR’d. Desktop: use the editor’s “Recognize Text / OCR” function and pick the right language (e.g. Indonesian + English for a mixed document). Mobile: a scanner app (e.g. your built-in scan feature or Microsoft Lens) can photograph and OCR into a PDF. How to confirm: after OCR, try to select and search a word—if it appears, OCR worked. Accuracy drops on blurry, skewed, or handwritten scans.
Merge, split, and reorder pages
Desktop: the built-in macOS reader (Preview) can reorder, delete, and merge pages offline; on other systems, use a page-organiser app or a web tool. Mobile: free Files/PDF apps usually support merge/split. Failure sign: page order scrambled or orientation flipped—fix rotation/order before saving.
Permanently remove sensitive data (redaction)
This is the most commonly botched task. Painting a black box over text does NOT remove the text beneath it. PDFs render content in layers, so a black box only covers the display; the original text remains and can be copy-pasted, searched, or extracted—it has happened to real “redacted” court filings that were read back out with a simple copy-paste.
The right way: use a Redaction feature that actually removes the text from the file’s content stream, then save. Afterwards, test it: try selecting the redacted area and pasting into a text editor—if no text appears, the redaction worked. For ID cards, contracts, or tax documents, never rely on a black box.
Capability matrix of common tools
Grouped by capability, not brand hype. “Uploads to server” means the file is sent to an online service (a privacy consideration); “offline” means it’s processed on your device.
| Tool category | Platform | Account needed? | Server/offline | OCR | Edit original text? |
|---|---|---|---|---|---|
| Free PDF readers (e.g. Acrobat Reader, built-in readers) | Web/Win/macOS/Android/iOS | Usually no | Offline | No | No (adds a layer/annotation only) |
| Preview (macOS) | macOS/iOS (Markup) | No | Offline | No | No (fill & sign, organise pages) |
| Offline desktop editors (e.g. LibreOffice Draw) | Win/macOS/Linux | No | Offline | Limited/not built-in | Yes (layout may shift) |
| Full paid PDF editors (e.g. Acrobat Pro) | Win/macOS | Yes | Offline | Yes | Yes + true redaction |
| All-purpose web apps (merge/split/compress/convert) | Web | Sometimes | Uploads to server | Some | Limited |
| Offline/self-host tools (e.g. PDF24 desktop app, Stirling-PDF) | Win/macOS/Linux | No | Offline/self-host | Some | Limited–yes |
| Mobile scanner apps (e.g. Lens) | Android/iOS | Sometimes | Depends on the app | Yes | No (produces a new PDF) |
Free limits/watermarks, export quality, and paid features vary and change often—check directly in the app. If privacy matters, prefer an offline option.
Verify the result before sending
- Reopen the output in a different reader to confirm it displays correctly.
- Search/select text to confirm OCR worked (for scans).
- Check fonts and layout—text edits can shift lines or swap fonts if the original font isn’t embedded.
- Check page size and orientation (A4 vs Letter, portrait vs landscape).
- Check a digital signature: is it still valid after editing (it often becomes invalid).
- Test redaction with copy-paste/search.
- Keep the original version separately before editing, so you can retry if the result is wrong.
Sensitive documents (ID cards, contracts, tax, health)
For documents with personal data, prefer an offline workflow or a service with a verifiable privacy policy—don’t upload carelessly to sites that don’t clearly state whether they store or share your data.
IsonAI can help edit and fill PDFs right in the conversation. Honestly: documents are processed to fulfil your request and not kept longer than needed, and result files are downloadable for a limited time. The handling details are in the IsonAI privacy policy (deleting your account removes the associated data). For the most sensitive data, doing the work entirely offline remains the most conservative choice.
Common mistakes and fixes
| Symptom | Cause | Fix |
|---|---|---|
| Text can’t be selected | Scanned PDF (an image) | OCR first, then edit/search |
| Layout breaks after editing | Original font not embedded | Embed/substitute a similar font; check alignment |
| ”Redaction” still readable | Black box, not redaction | Use true redaction; test with copy-paste |
| Can’t edit at all | Locked/password-protected PDF | Open with the legitimate owner password; don’t try to crack it |
| Signature becomes “invalid” | File edited after digital signing | Edit first, sign last; or request re-signing |
| File too large | High-resolution images | Compress the images, don’t degrade the whole document until text blurs |
Frequently asked questions
Why can’t I select text in a PDF?
Most likely it’s a scanned PDF—its content is an image, not text. Run OCR so the text can be selected, searched, and edited.
Why did the formatting change after I edited the text?
Because the original font isn’t embedded in the PDF, so the editor substitutes another font and the layout shifts. Embed the same font or pick the closest substitute, then fix the alignment.
Is a black box enough to censor data?
No. A black box only covers the display; the text beneath remains and can be copied or searched. Use redaction that truly removes the text from the file’s contents.
How do I edit a locked/password-protected PDF?
If you hold the owner password, open with it and edit. If not, respect the protection—don’t try to crack a document that belongs to someone else.
Is an image signature legally valid?
A signature image acts as a visual approval, but it doesn’t prove the document’s integrity. For stronger legal weight, use a certificate-based digital signature from a recognised provider.
How do I shrink the size without blurring text?
Compress the images inside the PDF (lower the resolution of photos/scans), not the whole document’s quality. Vector-based text stays sharp as long as you don’t rasterise the whole page.
Let IsonAI do the editing
If you’d rather not hop between apps: upload the PDF to IsonAI, say the pages and the change, then check the result file. Honest about limits: bad scans, custom fonts, password-protected PDFs, and digitally signed documents stay challenging in any tool—including IsonAI. For those, prepare a cleaner source or do the sensitive part offline, and always reopen the result file to confirm the text, layout, and (if any) redaction are correct before you send it.
Sources & review
This guide was reviewed in July 2026 against the following sources.
- U.S. Court of Federal Claims — PDF File Redaction Best Practices (PDF) (a black box does not remove text; true redaction is required)
- Adobe — Permissions and limitations of signed PDFs (editing after a digital signature invalidates it)
- IsonAI Privacy Policy (handling of uploaded documents and data deletion)
Last updated: