Word / PDF diagnosis
Why can a file appear without searchable text?
Appearing in the file list does not automatically mean the words inside can be searched yet. Text-based documents extract automatically; for scanned PDFs or words inside images, turn on offline Image text recognition (OCR) in Settings and wait for recognition to finish.
Start from what you can observe
| Observed state | Possible explanation | Check and next action |
|---|---|---|
| Name found, known body text absent | Only the record is ready, wrong mode, narrow filters, or pending text and OCR processing | Select Content, clear filters, expand the bottom status bar to check file processing and OCR progress, and retry the same text |
| DOC/DOCX/PDF opens with readable text | Some prerequisites are met, but not necessarily complete extraction | Search a distinctive source phrase and inspect snippets; check the file state if missing |
| PDF is mostly scanned pages, or words are inside images in Word/PPT/Excel | The document lacks extractable text layers, Image text recognition is off by default, or the account only covers standalone images | Try selecting text in a reader as an initial check; to search scanned PDF pages and images inside DOCX/PPTX/XLSX, sign in with Pro access, enable Settings → Image text recognition, and wait for OCR progress in the status bar to finish |
| Standalone image shows text without matches, or recognized file still misses words | Image text recognition is disabled, text is blurry or tiny, or the image uses an unsupported encoding such as EMF/WMF | Sign in to a registered account and enable Settings → Image text recognition for standalone images; if enabled, check file details for partial recognition failures and retry, or fall back to filename and folder clues |
| Password required or the original app reports an error | Protection, corruption, or permissions | Obtain an authorized readable copy or inspect in the original app; renaming the extension is not a repair |
| Very large, occupied, or still processing | A reading limit or temporary inability to read | Let other apps finish writing, then rescan and retry without altering the source |
| Text matches but built-in preview is unavailable | Extraction and preview support differ | DOC text reading and DOCX built-in preview have different scopes; confirm with the default app |
Check three distinct file conditions
A readable text document
Choose a DOCX or PDF with text that opens normally. Note a phrase, add the folder, wait for processing, and search it. Verify the snippet and full path.
A scanned PDF or document with images
Confirm your sign-in and tier (a registered account covers standalone images and scans; Pro adds PDF pages and images inside DOCX/PPTX/XLSX), enable Settings → Image text recognition, and wait for OCR progress in the expanded status bar. Then check the triple-backtick fenced OCR text in preview and search again.
A state you cannot identify
Record the format, whether the original app can open it, status bar processing and OCR progress, partial OCR status in file details, and search mode separately. A file record alone cannot identify the failure.
Format, OCR, and version boundaries
Image text recognition models ship with the app and run offline without uploading files or results, covering common standalone images (PNG, JPG, JPEG, BMP, GIF, WebP) with automatic Windows DirectML GPU acceleration when compatible. Accuracy still depends on clarity, text size, layout, unsupported image encodings, and text-budget limits, so complete recognition is not promised for every scan. Consult the manual for ordinary extraction and preview lists, and check Features and pricing for account tier details.