Working with documents
Converting PDFs to editable Word
Drop a PDF, pick DOCX as the target. The output opens in Word, Google Docs, or LibreOffice as an editable document. The result is good for text-first PDFs and gets messier as the source design gets more elaborate — multi-column magazines, infographics, and annotated forms all need cleanup. For scanned PDFs (a PDF that's really just an image inside a wrapper), run OCR first to extract the text into a real text layer.
Converting Word to PDF
Drop a DOCX, pick PDF. We use a headless LibreOffice pipeline that's the most faithful open-source converter we've found — fonts, headings, tables, page breaks, and bookmarks all carry over. Standard fonts (Arial, Times, Helvetica, Calibri) render exactly. For custom fonts, embed them in the source DOCX before converting (most word processors have an "embed fonts" option).
Converting between image formats
Most image conversions are sub-second. Things to know:
- Going PNG → JPG shrinks the file but adds compression artifacts and flattens transparency to white.
- Going JPG → PNG grows the file (often 5–10×) but doesn't add quality back — JPG already discarded data the PNG can't recover.
- WebP is usually the smallest at equivalent quality and supports transparency. It's the right default for the modern web.
- BMP and TIFF are big — they're for legacy tools and archival, not day-to-day use.
See PNG vs JPG vs WebP for the full trade-off.
Converting HTML to PDF
Drop an HTML file (or save a webpage as HTML and drop that), pick PDF. The renderer respects most CSS, including print stylesheets — if your HTML has a @media print block, that's what shows up in the PDF. Background colors and images render by default; web fonts load from CDNs at conversion time. More on HTML → PDF.
OCR — extracting text from images and scans
Pick OCR (Extract Text) as the target for any image or PDF. The result is a plain-text file with the recognized text. OCR uses Google Cloud Vision, which is high-accuracy on clear, high-contrast scans and noisier on faint, skewed, or handwritten pages. For best results: scan at 300 DPI or higher, use good lighting, and crop tightly. How OCR works under the hood.
File size and batch limits
20 MB per file, 5 files per upload. The 20 MB limit is per source file — converted output may be larger or smaller. If your file is bigger:
- Photos: compress in Preview (Mac) or any image editor — most photos can drop below 20 MB without visible loss.
- PDFs: split into smaller chunks, or compress with Preview's "Reduce File Size" / Adobe Acrobat's optimizer.
- Videos and other unsupported formats: Formatly only handles documents and images; we don't convert video, audio, or archives.
- Genuinely larger files you need converted: write in with the use case and we'll see what we can do.
Privacy and what we keep
Files are deleted one hour after upload — both source and output. We don't keep backups, audit trails, or analytics copies of file contents. We log the conversion type and file size for capacity planning ("docx → pdf, 3.2 MB"), but not filenames, contents, or IP addresses tied to specific files. The full security writeup is on /security; the legal version is on /privacy.
Common errors and fixes
"File too large"
The source is over 20 MB. Compress or split the file and re-upload. For photos, dropping JPG quality from 100 → 85 typically halves the file with no visible change.
"Unsupported format"
The file extension or MIME type isn't in our supported list. Check /formats for what we handle, or rename the file with the correct extension if you know it's actually a supported type misnamed.
"Conversion failed"
Usually one of: corrupted file, password-protected PDF, or a sub-format we don't handle (PDFs with embedded forms, DOCX with macros). Try re-saving the source from the original tool, or strip the password / forms first.
"Network error" during upload
Slow or interrupted connection. Refresh and try again. On flaky mobile data, switch to Wi-Fi for files over a few MB.
Stuck "Processing…"
Most jobs finish in seconds; OCR on a long PDF can take a minute. If the status hasn't moved after two minutes, the worker has likely failed silently — refresh the page and re-upload. If it keeps happening with the same file, email bugs@formatly.app.
Download link 404s
Files are deleted one hour after upload. An old link will 404 — re-upload to get a fresh one.
Browser and platform notes
- Chrome, Firefox, Safari, Edge: all current versions work.
- Internet Explorer: not supported.
- iOS Safari: works fully — uploads from photo library, files app, and any cloud drive that's installed.
- Android Chrome: same — uploads from gallery, files, Google Drive, etc.
- Privacy extensions: some ad blockers and privacy extensions occasionally intercept upload or download requests. If something's misbehaving, try a private/incognito window.
Keyboard shortcuts
- ⌘↵ (Mac) / Ctrl+Enter (Windows/Linux) — run the conversion without clicking Convert.
- Esc — clear the current selection and reset the form.