Bring a PDF table into a spreadsheet

Check selectable text first

1. Open your PDF or try the two-page table.
2. Draw the table region, adjust columns and check the first rows shown.
3. Choose pages to combine, recover the table, edit cells and download your CSV.

Open the PDF in your reader and try selecting a few words in the table. A document showing numbers may still be just a photograph. This first version does not perform OCR: it uses the positions of text already in the PDF. A scan with an existing OCR layer may work, but recognition mistakes will appear in the cells. Always compare values with the visible page.

Start with one representative page. A price list with a reference, a description and an amount is easier to recover than a layout full of merged cells. The result does not reproduce colours, borders or spreadsheet formulas. Your PDF stays unchanged. The built-in example provides two fictional pages to try without using your own document.

Select the region and place dividers

Draw a rectangle around the useful rows, keeping the header if you want to export it. Leave out the page title, notes and footer. Four position fields offer the same operation using a keyboard. Percentages refer to the visible page: left and top locate the corner, while width and height set its size.

Choose the number of columns, then move their dividers using the sliders. Put them in gaps between text. A warning identifies text blocks crossing a divider. A PDF does not always expose individual letter positions, so a block crossing a boundary is not artificially cut into letters. Adjust the region or correct the result. The row-grouping setting joins nearby text fragments, but a high value can merge neighbouring rows.

Opening a PDF does not validate a region. Draw your own or change the keyboard position fields. The preview immediately shows up to five rows from the displayed page: if titles or notes appear, tighten the region. Recovery stays disabled until you choose a frame.

Combine pages without silently losing rows

Enter pages such as 1-3, 5. They are processed in your specified order without duplicates. The same region and columns are reused proportionally to each page’s visible dimensions. Browse every preview first: a later page with a larger title or a different layout may need its own extraction.

The header option does not blindly remove one row per page. It removes the first row on a later page only when every cell exactly matches the initial header. The removed count is shown. A selected page without text in the region raises a message and preserves your previous result. Empty pages are never silently skipped.

All pages are preselected for documents of up to 50 pages. The selected count and page numbers stay visible. You can choose the displayed page or a custom list instead. For a longer PDF, the displayed page is initially selected and All pages is disabled.

Edit and import the CSV

Each text line becomes a result row. Edit cells directly, manually combine multiline descriptions and remove rows you no longer need. The source column shows the PDF page for checking; it is not exported. Editor pagination does not limit the download: all remaining rows are included.

Choose semicolon, comma or tab. The file uses UTF-8 with an Excel marker and escapes quotes and line breaks. A protective apostrophe is added to text that might become a formula, including negative numbers. No amounts are calculated. To keep codes such as 0012 and preserve dates, use your spreadsheet’s import command and choose text columns. Compare some amounts and the first and last row of each page afterwards. Files stay in this tab without uploading; download before closing.