PDF to Excel
Lift the rows of a text-based PDF into an editable workbook, one worksheet per page, with the columns inferred from the spacing. A scanned PDF has no text layer and is refused by name. The file is sent to the tool server for that run and not stored.
What PDF to Excel does
Lifts the rows of a text-based PDF into a spreadsheet you can edit. The columns are inferred from the spacing: the PDF text engine keeps the position of every word, and a gap of two or more spaces between words is treated as a cell boundary. That is the whole heuristic, and saying so is more useful than claiming table detection. A price list, an invoice, a bank statement with aligned columns or a report table written with a tab stop all arrive as rows.
Two ways to read the page
Choose Columns when the page has more than one column: each line becomes a row and each run of two or more spaces starts a new cell. The first row is bold and stays in view while you scroll, because it is almost always a header. Choose Lines when the page is prose or a single-column list: every line becomes one cell in column A, which keeps an address or a paragraph intact instead of splitting it at a stray gap.
How to use it
Choose a .pdf file, pick a mode, optionally name the pages you want with the 1-3,5 grammar, and press Run. Blank pages means every page. The selection is applied before extraction, so only those pages are read, and the PDF may have at most 200 pages. Each selected page becomes one worksheet named Page 1, Page 2 and so on, and the result is a .xlsx you download with its page count and size shown beside it. A column is as wide as its longest cell, up to 60 characters.
What it will not do
This is spacing-based extraction, not a layout model. A merged cell, a table split over several lines, a nested table, a column that spans others and a proportional font whose gaps fall to a single space are all approximated or arrive as one column. Numbers stay text, so a currency symbol or a thousands separator is not reformatted and a formula is not rebuilt. If every selected page has no text layer — a scan, a photo or an exported image — the tool refuses with a sentence rather than returning a workbook full of apology notes, because OCR is a different job and is not part of this tool.
Privacy. The file is sent to the tool server for this run and is not stored, and the temporary directory holding it is deleted when the run ends.
Upload a text-based PDF and this page lifts its rows into an Excel workbook, one worksheet per page, with columns inferred from the spacing the typesetter left between cells.
Two modes decide how a line becomes cells. Columns splits each line on two or more spaces, which is what a table row looks like in a layout-preserving text extraction. Lines puts the whole line into column A, which is the escape hatch for prose or a single-column list.
It is spacing-based extraction, not table detection. There is no OCR, no learned layout model, no merged cells, no multi-line cells and no column spans, and a scanned PDF with no text layer is refused by name rather than turned into a workbook of apologies.
What the workbook contains
Each selected page becomes a worksheet. The first row is treated as the header, formatted bold and frozen so it stays visible while scrolling, and each column's width is taken from its longest cell up to a ceiling.
Cells that look like numbers are written as numbers so a column can be summed. A page whose selected rows are all empty is not written as a blank sheet.
- One worksheet per selected page
- Bold, frozen header row and widths from the longest cell
- Numbers stored as numbers, not text
When the PDF has no text
A scan is a picture of text and contains no characters for the extractor to read. When every selected page comes back empty, the run is refused with the sentence that names OCR as the missing step.
Page selection uses the 1-3,5 grammar and is applied before extraction, so only the pages you asked for are read. The upload is capped at twelve megabytes and the document at two hundred pages.
How to use the mode choice
Upload the text-based PDF in the Upload your file step, pick the mode in Options, Columns splits each line on two or more spaces for table rows, Lines puts the whole line into column A for prose, and optionally narrow the 1-3,5 page selection, then press Run. Each selected page becomes a worksheet with the first row bold and frozen and number-like cells written as numbers.
A scanned PDF with no text layer is refused by name rather than turned into apologies: there is no OCR here. The header row stays visible while scrolling, a fully empty page is not written as a blank sheet, and the upload is capped at twelve megabytes and two hundred pages.
What to Expect
Best for
- Getting a simple table out of a digital PDF
- Splitting a long report into one sheet per page
- Extracting a single-column list into column A
Not ideal for
- Scanned documents, which need OCR first
- Tables with merged cells or cells that wrap across lines
- PDFs where columns are separated by a single space
What this tool keeps
- The page order you selected
- A header row kept visible while scrolling
- Numbers as numbers, so totals work
What may need cleanup
- A cell split in the wrong place because the gap was one space
- A two-line cell broken into two rows
- A header row that was not actually a header
Common errors
- Uploading a scan and expecting OCR
- Using columns mode on prose, which splits sentences at wide gaps
- Expecting merged cells, which this extraction cannot represent
Example use cases
- Moving a printed price list into a spreadsheet
- Extracting a statement's rows for reconciliation
- Turning a one-column address list into cells
Sample input
Upload a two-page digital invoice PDF, keep columns mode and leave the page field blank.
Sample output
A workbook with two worksheets, each with a bold frozen header row and numeric amount cells, named pdf-to-excel.xlsx.
Who this is for
- Bookkeepers extracting statement rows
- Analysts moving a PDF table into Excel
- Anyone with a digital PDF and a spreadsheet to fill
Frequently asked questions
How does PDF to Excel infer columns?
It lifts the PDF's rows into a worksheet per page and infers columns from the spacing the typesetter left between cells.
Can it read a scanned PDF?
No. A scanned document has no text layer and needs OCR first.
Why is a cell split in the wrong place?
A gap of only one space between columns can be read as a cell boundary, so a table with tight spacing may not line up.
Build custom tools for your store
Join the SDK program to create useful ecommerce utilities for your store.