Extract form data

The CSV shape depends on which pivot you choose. "One row per field" produces a long table with the header filename,fieldName,fieldType,fieldValue and one line for every field of every PDF you dropped in — the shape that pivots cleanly in a spreadsheet. "One row per file (fields as columns)" produces a wide table whose header is filename followed by every distinct field name found across all your files, sorted alphabetically, with one line per uploaded PDF and an empty cell where a file lacks that field. Both are written RFC 4180 style: CRLF line endings, and any cell containing a comma, a double quote or a line break is wrapped in quotes with internal quotes doubled. The download is named form-data.csv.

The JSON mirrors the same two pivots. In long mode the file is a flat array of objects, each carrying filename, fieldName, fieldType and fieldValue keys — the same four values as the long CSV, in the order the fields appear in the documents. In wide mode it is an array of objects shaped as filename plus a fields object mapping each field name to its value, one entry per uploaded PDF in upload order. Wide mode is deliberately an array rather than an object keyed by filename, because two different PDFs both called application.pdf are entirely normal and an object would silently drop one of them. Output is indented with two spaces and saved as form-data.json, and every value is a string.

Each field type exports as a predictable string. A check box becomes the lowercase text true or false. A radio group exports the export value of the selected option as stored in the PDF, which is often something like Yes or Choice1 rather than the label printed beside the button, and an empty string when nothing is selected. Dropdowns and multi-select option lists export their selected values joined with the pipe character, so a list with two selections reads Blue|Green. Text fields export verbatim, line breaks included, which is exactly why the CSV quotes them. The fieldType column tells you which rule applied, using the values text, checkbox, radio, dropdown and option-list.

A file is skipped when it yields no readable values, and the file list states which case applied. "Skipped — no fields" means the PDF has no AcroForm at all, which covers plain documents and scans. "Skipped — no readable fields" means fields were found but none held an exportable value, typically a document whose only widgets are push buttons or signature fields; those are counted separately next to the filename. An XFA form from Adobe LiveCycle also reads as fieldless, so print it to a new PDF in Acrobat Reader first. An encrypted file shows Error instead — remove the password with the Remove PDF password tool and re-drop the unlocked copy. Individual fields that could not be read are counted as unreadable rather than exported as blank.

That is the field's internal name, and it is what the PDF actually stores. Government and enterprise forms built in Acrobat or LiveCycle generate hierarchical names from the form's own structure, so the label a human reads as "Full name" may be stored as Topmostsubform[0].Page1[0].FullName[0]. Many forms also carry a friendlier alternate name in the TU entry — the tooltip you see when hovering a field in a reader — but that is a separate property and not what identifies the field, so it is not used as a column heading here. If you need the readable labels, open the same PDF in the Fill PDF form tool, which shows each field's tooltip beneath its input. Renaming the columns afterwards in your spreadsheet is usually quickest.

The values never leave your browser — extraction, preview and export all happen in the tab. Each PDF is read from disk with the browser's File API, parsed with pdf-lib, and the CSV or JSON is assembled as a string in memory and handed to your downloads folder. No request carries the data anywhere, because there is no backend behind this page. That is the point for the documents this tool exists for: returned job applications, timesheets, medical intake forms, expense claims and survey responses are exactly the files that should not be uploaded to a stranger's server to be turned into a spreadsheet. The tool is free with no limit on how many PDFs you drop in at once, and adds nothing of its own to the export.