PDF to Excel
Drop a PDF here or click to choose
Choose a PDF with real text. Scans will not become rows.
Guess spreadsheet columns from gaps in a PDF's text layer and download XLSX or CSV. This is a spacing guess, not a reconstruction of the original table.
Read the guide: PDF to Excel is a spacing guess
How to use
- Choose a PDF where you can already select the text. Scans do not become rows.
- Click Build sheet. The box shows the first rows so you can see whether the columns split in a useful place.
- Download XLSX or CSV. CSV is UTF-8 with a byte-order mark so Excel is more likely to open accents correctly.
How columns are chosen
Words that sit close together stay in one cell. A wide gap, including the large space PDF readers insert between columns, starts a new cell. The same horizontal positions are lined up down the page. A title that spans the page sits in the left cell and leaves the other cells blank.
If every line becomes one cell, the file did not have a gap wide enough to call a column. That is reported in the status line. It is still a usable one-column sheet. Pages are stacked in one worksheet. When the PDF has more than one page, a row reading Page N is inserted so you can see the break. There is no second sheet per page.
Merged cells, ruled lines, and multi-line cells are not reconstructed. A scanned page has no cells unless the PDF already contains real text. There is no OCR on this page. Password-protected files are refused. The limit is 80 pages.
A worked example
A bank statement exported as PDF, with a date, a description, and an amount separated by wide spaces, often splits into three columns. You still need to check the description column: a long payee name can be cut, and a wrapped description can become two rows.
A price list set in a tight font, where the gap between name and price is about the same as a normal word space, often lands in one cell. That is the tool refusing to split a sentence into words. Widen the source, or split the cells in Excel. A photo of a spreadsheet has nothing to measure. Use PDF to Text with English OCR if you only need the words, or type from PDF to JPG.
When not to use this
- The PDF is a scan or a picture of a grid. There is no text layer to measure.
- You need formulas. PDF tables are not spreadsheets. Numbers come back as text.
- The layout is a magazine page with sidebars. You will get lines, not the designer's grid.
Questions
Why did two columns land in one cell?
The gap between them was small, similar to a normal word space. The tool would rather keep a sentence together than split every word. Split the cells in Excel, or use a source with wider gaps.
Can it read a photo of a spreadsheet?
No. OCR is not part of this page. A picture has no columns to measure.
Will formulas survive?
No. You get text, including numbers stored as text. There are no formulas to keep.