How to convert a PDF table or invoice to Excel

Why does a table fall apart when you copy it from a PDF into Excel?

Because a PDF does not store cells. It stores pieces of text, each with a position on the page: this word goes here, that number goes there. The rows and columns you see are a visual effect; inside the file there is usually nothing that says where one cell ends and the next begins.

When you copy and paste, Excel receives a stream of text and has to guess the structure. The usual results are familiar: a whole row crammed into one cell, columns shifted wherever a cell was empty, or a long description split across two rows.

How can I tell if my PDF has real text or is a scanned image?

Try to select a single word with the cursor. If that word highlights, the PDF contains real text. If nothing highlights, or the whole page gets selected as one block, you have an image wrapped in a PDF. Another quick test is to search with Ctrl+F for a word you can see on the page: if it is not found, there is no text.

What are the manual ways to convert a PDF to Excel?

Excel's PDF importer

Recent versions of Excel for Windows include an importer under Data > Get Data > From File > From PDF. The exact path may vary by version, and not every edition has it. Excel lists the tables it detects in the file and lets you choose which one to load.

It works well on PDFs with real text and clearly ruled tables. It does not read scans, and with borderless tables or merged cells it may split the columns in the wrong places.

Paste the text and split it into columns

Copy the table and paste it into a plain text editor first to see what separates the values: spaces, tabs, or line breaks. Then paste it into Excel and use Text to Columns, usually found on the Data tab. The limitation is obvious: if descriptions contain spaces, splitting on spaces breaks each description into several cells.

Retype the table

For a table of five or ten rows, typing it in is often the fastest and most reliable option. That stops being true as volume grows, because fatigue produces errors too, and those leave no trace.

How do I get a photo or a scanned document into Excel?

With OCR, optical character recognition: software looks at the image and decides which letter or digit sits in each spot. It is an interpretation rather than an exact reading, and its quality depends heavily on the image it gets. These are the things that hurt it most:

• Skew and perspective: a sheet photographed at an angle bends the rows and blends columns together.
• Shadows and glare: they darken or wipe out part of the text, especially on glossy paper.
• Low resolution: small digits get confused with one another, such as 3 with 8 or 1 with 7.
• Handwriting: it is recognized much less reliably than print.
• Stamps, signatures, and creases that run across the figures.

Before you take the picture, lay the sheet on a flat surface, use even light, hold the camera parallel to the paper, and keep the whole page in frame. A good photo saves more time than any amount of fixing later.

If you would rather not chain several programs together, the tool to extract data to Excel accepts both PDFs and photos of invoices, tables, and forms.

How can I extract the table automatically?

You upload the PDF or photo, the tool shows the tables it found on screen, and you download an .xlsx file. The workbook comes organized in two parts: the line items (each product or service with its quantity and amount) on one sheet, and the document's standalone fields (issuer, number, date, totals) on another. Header data stays out of the rows.

Extraction is automatic and can be wrong. Check amounts, dates, and totals against the original document before the spreadsheet goes into a tax filing, a payment, or a report.

If the invoice may be needed as a supporting record later, keep the original file unmodified. You can certify the file to have a record that it existed, unchanged, on a given date.

What errors should I check for after converting?

The same ones show up again and again, whatever the method. Nearly all of them come from how Excel interprets what it receives:

How do I verify that the spreadsheet matches the original?

With a short routine that doesn't require rereading everything. It takes a few minutes and finds most problems:

1. Count the rows. The number of line items in Excel should equal the number in the document. One too many or too few means a row was split or merged.
2. Add it up again. Sum the amount column and compare it with the printed subtotal. Do the same for tax and the grand total.
3. Spot-check rows. Take the first, the last, and two or three from the middle, and compare them field by field.
4. Look at the extremes. Sort the amount column and look at the highest and lowest values: a misread separator tends to surface there.
5. Confirm the header data: issuer, document number, and date.

If the totals do not match, don't edit the figure by hand to force agreement. Find the row causing the gap: it is almost always a misread digit or a shifted decimal.

What mistakes should I avoid?

• Trusting the sheet because it looks tidy. A neat table can still hold a wrong amount.
• Fixing the only copy. Save the extraction as it came out and work on a duplicate.
• Summing without checking formats. Numbers stored as text are left out, so the total looks plausible but comes up short.
• Throwing away the original. The spreadsheet is a working copy; the document is the record.

A checklist before you use the data:

1. I checked whether the PDF has text or is an image.
2. The row count matches the original.
3. Subtotal, tax, and total add up again.
4. Dates have the right day and month.
5. Codes kept their leading zeros.
6. I kept the original document unmodified.