How to convert a PDF to Excel and get a table you can use

Which PDF tables convert cleanly, and which do not?

A PDF does not contain a table the way Excel does. It contains text placed at coordinates and some lines drawn around it. A converter looks at those positions and guesses where the rows and columns are, and the guess is reliable in some cases and hopeless in others.

Tables that convert well: exported from accounting software, a bank, a payroll system or a database; simple grids with one value per cell; a header row and consistent columns on every page. Bank statements, invoices, price lists and financial statements generated by software usually come out ready to sum.

Tables that convert badly: scanned or photographed pages, where there is no text at all, only pixels; cells with several lines of text, which land in separate rows; merged header cells spanning columns; tables with no ruling lines and uneven spacing; and layouts where two tables sit side by side. These can still be converted, but expect to fix the result by hand or to use OCR first.

Is your PDF a scan? Check before you convert

Open the PDF and try to select a few words with the mouse. If the text highlights and you can copy it, the file has a text layer and any converter can read it. If dragging draws a box over the whole area, or nothing happens, the page is a picture: a scan, a fax or a phone photo. No PDF to Excel tool can read a picture directly.

In that case run OCR first. It reads the text in the scanned pages and adds a real text layer, so you can search, copy and convert the file. Convert the OCR output to Excel as a second step. If the table is a photo or a screenshot rather than a PDF, Image to Excel does both jobs in one pass and returns an XLSX.

OCR confuses 1 and l, 0 and O, 5 and S, and can drop a minus sign or a decimal point, so a scanned table needs the checks at the end of this guide more than any other.

Convert PDF to Excel online for free

PDF to Excel pulls the tables out of a PDF into an XLSX spreadsheet. It is free, needs no account and leaves no watermark. Files up to 300 MB and up to 50 files per batch are accepted, so a year of monthly statements converts in one run. It works in the browser on a phone or computer, runs on its own servers, and files are processed in memory and deleted after 30 minutes.

Expect one sheet per table or per page, depending on how the PDF is laid out. Open the XLSX: if the numbers are right-aligned, Excel read them as numbers. If they are left-aligned, they came in as text and need the cleanup step below.

1. Open PDF to Excel and add the file. For a scan, use the OCR output rather than the original.
2. Run the conversion and download the XLSX.
3. Open it and check the header row and one column of numbers.

Use Excel's own Get Data from PDF on Windows

Excel for Microsoft 365 on Windows can import a PDF directly through Power Query. It shows you every table it found, lets you pick one or several, and loads them as proper Excel tables. You can also refresh the import later when a new version of the same report arrives.

The feature is not in Excel 2016 or 2019, not in the free web version, and Excel for Mac has only part of Power Query, so check your Data tab. If From PDF is missing, use the online converter above. Power Query also cannot read scanned PDFs; a scan shows up as zero tables found.

1. In Excel choose Data, then Get Data, then From File, then From PDF, and pick the file.
2. In the Navigator window, click each Table or Page entry on the left to preview it on the right. Tick Select multiple items to load several at once.
3. Click Load to put the table in a new sheet, or Transform Data to fix column types and remove header repeats in the Power Query editor first.
4. Right-click the loaded table and choose Refresh whenever you replace the PDF with a newer one.

The Google Sheets paste trick

For a small table on one page you can skip converters entirely. Open the PDF in Chrome, select the table by dragging from the first cell to the last, copy it, and paste into a blank Google Sheet. Sheets often splits the text into columns at the spaces or tabs it finds,.

When everything lands in column A, select that column and choose Data, then Split text to columns, and pick a separator (space, comma or custom). This goes wrong when values contain spaces, such as company names, because the split cuts them in half. Pasting into Google Docs first, then copying from Docs into Sheets, sometimes keeps the columns better because Docs recognizes the table structure.

This method is good for a quick look and poor for anything longer than a page, where the copy picks up headers, footers and page numbers you then have to delete.

Clean up the converted spreadsheet

Converted tables arrive with predictable dirt. Numbers come in as text because they carry a dollar sign, a comma, a trailing space or parentheses for negatives, so a SUM over the column returns 0. Empty rows appear where the PDF had a page break. Header rows repeat every 40 lines. Dates arrive as text.

Clean up spreadsheet handles the mechanical part: it removes duplicate and empty rows, trims stray spaces and fixes numbers stored as text in one pass. Run it on the converted XLSX, then do the remaining fixes in Excel.

The remaining fixes are usually: delete repeated header rows (filter on the header text to find them), turn parentheses negatives into minus signs, split a column that carries two fields into two, and set the date column to Short Date once Excel recognizes the values. Save the cleaned file under a new name and keep the raw conversion for reference.

Check the numbers before you trust them

A converted table that looks right can still be wrong in one cell, and that is the cell that ends up in a tax return or a board slide. Spend five minutes verifying against the PDF.

Start with the totals. If the PDF shows a total row, sum the converted column and compare. A mismatch tells you a row was dropped, a value was read as text, or a digit changed. Then count the rows: 212 transactions on the statement should be 212 rows in the sheet, no more and no fewer. Spot-check five rows digit by digit, including the sign.

For OCR output, add a pass for the classic confusions: an l in a number column, an O where a 0 should be, a decimal point dropped so 1,250.00 became 125,000. When the stakes are high, convert the same PDF with a second method and compare the two results with a simple = check between the sheets; the cells that differ are where to look.

• Sum each column and compare to the total printed in the PDF
• Count rows against the statement or report
• Spot-check five rows digit by digit, signs included
• Filter for numbers stored as text and fix them
• For scans, compare two conversions and inspect the cells that differ

XLSX or CSV: which should you save?

Keep the XLSX as your working file; it holds formatting, several sheets and formulas. If the data is going into accounting software, a database or another system, those tools usually want CSV. Save a copy with File, Save As, CSV (Comma delimited), one per sheet, or use Excel to CSV to export every sheet at once with ISO dates and plain numbers, which import systems prefer.

A CSV has no types: dates and leading zeros (ZIP codes starting with 0, account numbers) are at risk when the file is reopened in Excel. Keep the XLSX as the master and treat the CSV as a one-way export.