An accountant sends over a twenty-page bank statement as a PDF, and you need to total the turnover by counterparty. Retyping three hundred rows by hand is half a day of work and a guaranteed crop of typos in the amounts. You know the feeling: the data is right there in front of you, but it is stuck in a format you cannot just copy into a normal spreadsheet. The same thing happens with supplier price lists, reconciliation reports, timesheets, and any report that arrives as a finished PDF with no source Excel behind it.
The reason is that PDF was built for printing, not for working with data. Inside, it does not store a "table with columns." It stores a set of characters with coordinates on the page: which digit is drawn where. A person sees a grid; the program sees a scatter of numbers. So a plain mouse selection often glues everything into one row or tears the cells apart. A good converter rebuilds the table structure from where the text sits, and the cleaner the source PDF, the closer the result.
Pulling the table out, step by step
1. Open PDF to Excel and upload the PDF with the table. If the file is heavy and made of scans, run it through compression first so the upload goes faster. 2. Wait for processing and download the XLSX file. Open it right away in Excel or Google Sheets and look at the first few rows. That is the most telling area. 3. Check that each source column landed in its own column instead of collapsing into one. Pay special attention to dates and amounts, which drift apart most often. 4. Strip out the clutter: headers repeated on every page, blank rows between blocks, and the "carried forward" and "page subtotal" rows from multi-page statements. 5. Force the numbers into a number format. After conversion, amounts often come through as text, and then they will not add up. Select the column and confirm Excel treats them as numbers (values sit on the right, not the left). 6. Check a control total. Sum the final column with a formula and compare it against the "Total" from the original PDF. That is your best defense against a lost or duplicated row.
What can go wrong
- **The PDF turned out to be an image.** If the text in the file does not highlight with the mouse, it is a scan with no text layer, and the numbers will not extract as data. You need OCR first, or the output will be empty or garbled.
- **Columns were laid out with spaces.** The converter reads tables without visible lines worse, and neighboring columns can merge. If the divider is obvious (a space or a tab), split the cells in Excel through Data then "Text to Columns."
- **Amounts became text.** Thousands separators (a space or a comma) and the currency symbol break the number format. Replace them with "Find and Replace" and switch the column to a numeric type, or the formulas will not work.
- **Merged cells fell apart.** Grouped headers ("Q1" sitting over three months) often scatter. It is faster to fix those by hand than to chase perfection from the converter.
- **Wrong decimal separator.** If the source uses a dot and your system uses a comma (or the reverse), Excel reads "12.5" as text or a date. Check your regional settings and swap the separator if needed.
- **Minus signs got confused.** In accounting PDFs, negative amounts are written in parentheses: (1,200.00). Excel does not always read that as a minus, so check the sign on expense rows.
What to check
- The row count in Excel matches the number of data rows in the source PDF (headers aside).
- The column total matches the "Total" from the original down to the last cent.
- Dates are recognized as dates, not as text or random numbers.
- No column has merged with its neighbor, and there are no stray empty columns.
- Text encoding is intact and counterparty names did not turn into gibberish.
What comes next
The finished Excel is ready to use straight away. If you later need to send the summary back in a locked-down form, build it through Excel to PDF. The text part of a report is easy to edit through PDF to Word, and PDF to JPG will save individual statement pages as images. All the tools are free, no registration, and files are deleted 120 minutes after processing.