Import a PDF, DOCX, or document image

Start in the document editor and choose the import option. The product first uses selectable text when a PDF contains it. A single standard DOCX is converted locally into the editor's supported semantic HTML; it does not use OCR or AI cleanup. When a PDF has no selectable text, the workflow can render its scan pages for OCR. PNG, JPEG, and WebP image pages can also follow the image-OCR path.

A DOCX import is not a page-perfect Word conversion or native DOCX round trip. Word-specific layout, comments, tracked changes, unsupported tables, floating objects, and other unsupported features may not be retained. The editor treats the result as one semantic document rather than claiming the source file's original page count.

Before importing

• Use a legible source with upright pages, clear contrast, and all required pages present.
• For a DOCX, import one standard .docx file (up to 8 MB) and keep the original available for comparison.
• Choose the correct available OCR language for the document where the editor offers a language choice.
• Do not upload a document unless you are authorised to handle its contents.
• Keep the original file available so you can compare it with the extracted draft.

Review the extracted draft

1. 1Compare party names, addresses, identifiers, dates, amounts, headings, and clause numbering against the original.
2. 2Check table rows, page breaks, signatures, stamps, handwritten marks, and image-only sections separately; they may not become editable text exactly as shown.
3. 3Correct or remove any OCR error before exporting, sharing, or signing the document.
4. 4Use professional review where an extraction error could materially affect the transaction or legal position.