Can Excel Extract Data From a PDF? Yes, With Two Limits

Last updated July 2026.

Yes, Excel can extract data from a PDF, with two hard limits. The Get Data From PDF connector reads tables out of digital PDFs that contain a text layer. It cannot read a scanned PDF, because Power Query has no OCR step. And it returns cells, never labeled fields, so a human still decides which number was the total.

Those two limits explain almost every frustrated hour spent on this problem. People try the connector, it works beautifully on the vendor who emails a real PDF, then returns nothing at all on the vendor who scans a paper bill. Same button, same file extension, completely different outcome.

How do I import a PDF into Excel?

Open a blank workbook. Choose Data, then Get Data, then From File, then From PDF, and pick your file. Excel opens a Navigator window listing every table it detected, plus whole-page options. Preview one, click Load, and it arrives on a worksheet as a proper table with columns intact.

Tick Select multiple items in that Navigator if you need several tables at once. If a single table runs across pages, Power Query sees each page as a separate table, so use Combine and then Append Queries to stack them. Watch the repeated header rows when you do: they will happily paste themselves into the middle of your data and flip the column from numeric to text, at which point every formula below returns zero.

Why does Excel say no tables were found in my PDF?

Almost always because the PDF is a scan. Test it in five seconds: open the file and try to select a word with your cursor. If the text highlights, there is a text layer and the connector should work. If your cursor draws a box across the page like you are cropping a photograph, the file contains an image of a document, not a document. Power Query reads text. There is no text to read.

The second cause is a table drawn without ruling lines or consistent spacing, which the detector simply does not recognize as a table. In that case, load the page as text rather than as a table and split it with Text to Columns, or move to a tool that reconstructs the grid from character coordinates.

Can Excel read a scanned PDF?

Not through the PDF connector. Something has to run optical character recognition first. Your options are to OCR the file elsewhere and then import it, to let Adobe Acrobat handle both steps at once (it runs OCR automatically when exporting a scan to Excel, and produces a workbook that mirrors the page layout), or to use a document extraction tool that recognizes the characters and identifies the fields in a single pass.

Scan quality decides how well any of those go. Adobe documents 300 dpi as the resolution that produces the best text for conversion, notes that accuracy drops slightly at 150 dpi, and will not apply OCR at all to source images below 72 dpi. A page emailed through a phone and compressed twice can fall under that floor without looking any different on screen. The full behavior is in does Adobe Acrobat do OCR.

Does Excel have OCR?

Excel has one OCR-adjacent feature, and it is not part of the PDF connector. Insert Data From Picture takes a photograph, a screenshot, or an image file of a table and converts it into editable cells. Microsoft documents it on Windows, on Mac (including capturing straight from an iPhone with Continuity Camera), on Excel for the web, and on iOS and Android. You review and correct the values before inserting them.

Its documented limits are worth knowing before you rely on it. Microsoft lists a fixed set of supported character sets, all Latin-script European languages plus Turkish, so it is not a general OCR engine. Images should be shot head on and in focus, without perspective or angle, and should contain only the data you want. And like every method on this page, it returns a grid of cells. It does not know which one held the invoice total.

What are the limits of Excel's PDF import?

Can Excel extract data from a receipt or an invoice?

It can pull the printed grid off one, if the file is digital and the layout is tidy. It cannot tell you that the 8.25 is sales tax and the 108.25 is the total, and that is the whole job. The proof is scale: import twenty invoices from twenty vendors with Power Query and you get twenty differently shaped worksheets, because twenty designers put the total in twenty places. Someone now reads each sheet and copies four values into the ledger.

Field extraction inverts that. It reads the document the way a person does, returns vendor, date, subtotal, sales tax, and total as named values from any layout, and stacks twenty documents into twenty rows with the same columns. That is what receipt to Excel converter does for expense receipts and what invoice PDF to Excel converter does for vendor bills, including the line items. The general case, mixed documents of every kind, is PDF data extraction.

When to stop fighting Excel

Excel's connector is genuinely good and it is already paid for. Use it when the PDF is digital, the table is clean, and you need the data once. It will save you an hour and cost you nothing.

Stop when you notice yourself doing any of these: opening the same import three weeks running, fixing the same merged-cell shift by hand, re-typing values because they imported as text, or emailing a vendor to ask for a real PDF instead of a scan. Those are all symptoms of using a layout tool for a data problem. A hundred receipts a month go through bulk receipt scanner in one upload, and receipt OCR software returns the same columns whether the paper came from a gas station or a hotel.

The output only has to be shaped for whatever comes next. If the destination is a spreadsheet, take the Excel file. If it is your books, export a CSV, and if your accounting software wants a bank-format file rather than a sheet, you can convert the CSV into a QBO file before importing. If the destination is your own application, call the extraction over an API and skip the workbook entirely.

Curious what field extraction returns on your worst PDF? Upload it and look at the fields.