OCR PDF to Excel: Turning Scans and Photos into Sheets

Quick Answer: A scanned PDF is a picture, so Excel cannot read it until OCR (optical character recognition) turns the picture into characters. For a page or two, Excel's free Data from Picture feature reads a table straight into cells. For whole documents, Acrobat Pro's export runs OCR automatically, and some online converters do too. Whatever you use, proof every number afterwards.

OCR to Excel is a two-stage job: recognition turns the scanned image into text, and table reconstruction arranges that text into rows and columns. Tools differ mostly in the second stage. A clear, straight, high-contrast scan converts well in nearly any of them; a skewed phone photo of a crumpled page converts badly everywhere, so the quality of the source matters more than the software.

How can you tell if a PDF is scanned?

Three quick tests, any of which settles it:

1. Try to select a word. Drag across a line of text. If individual words highlight, the PDF has a text layer. If a box appears around the whole page, or nothing highlights, it is an image.
2. Search for a word you can see. Press Ctrl+F (Cmd+F on a Mac) and type a word from the page. No match means no text layer.
3. Zoom to 400%. Real text stays sharp at the edges. Scanned text turns fuzzy or pixelated.

Some PDFs are hybrids: a scan with an invisible OCR layer added by the scanner. These pass the selection test, but the hidden text may be poor. If a converter gives you garbled output from a file that looked like text, that hidden layer is the likely cause, and re-running OCR with a better tool can help.

How do you prepare a scan so OCR reads it well?

Fixing the image before recognition saves more time than correcting the output after.

• Resolution. 300 dpi is the usual sweet spot for printed documents. Lower and small digits blur; much higher makes files large without improving accuracy.
• Straighten. Rotate pages upright and correct skew. Google's own OCR help says documents must be right-side up.
• Contrast. Black text on a white background. Remove coloured shading if your scanner settings allow it.
• Crop. Trim away hole punches, staples, handwritten notes and anything outside the table.
• Phone photos. Shoot head-on, in even light, with the page filling the frame. Use your phone's document-scan mode, which flattens perspective.

Which tools convert a scanned PDF to Excel?

Excel's Data from Picture

Microsoft's Data from Picture instructions cover each platform. On Windows choose Data > From Picture, then Picture From File or Picture From Clipboard for a screenshot. On a Mac the same choices sit under Data > Data From Picture, and there is also a Scan Documents option on the right-click menu that uses an iPhone camera through Continuity Camera. In the phone apps it is Insert > Data from Picture. Excel analyses the image, highlights cells it is unsure of, and lets you fix them before you click Insert Data.

To use it on a scanned PDF, export or screenshot one page at a time. It is the most accurate free table OCR for small jobs, and the review step catches many errors before they reach the sheet. Windows requires Windows 10 version 1903 or later with Microsoft Edge WebView2.

Acrobat Pro

According to Adobe's PDF to Excel how-to, you open the PDF, choose Export PDF, pick Spreadsheet and Microsoft Excel Workbook, and export; when the file is a scan, "Acrobat will run text recognition automatically." This is the practical choice for long scanned documents. The free online Acrobat converter does not include OCR; Adobe reserves it for Pro and the trial.

Google Drive

Right-click the file in Drive and choose Open with > Google Docs. Google's conversion help says the file should be 2 megabytes or smaller and text at least 10 pixels high, and that tables and columns are not likely to be detected. Expect a page of text that you paste into Sheets and split into columns. It handles over 100 languages, which helps with non-English documents that Excel's picture feature does not support.

Online OCR converters

Investintech's free converter applies OCR to scanned PDFs and needs no account. Smallpdf offers OCR conversion on its Pro plan. Both upload your file to their servers, so weigh that against what the document contains.

How do you proof OCR numbers in Excel?

OCR errors are not random. They cluster in characters that look alike, and those are exactly the characters in amounts and dates.

A proofing routine that catches most of them:

1. Find text masquerading as numbers. Select the amount column; if the status bar shows Count but no Sum, some cells are text. =ISNUMBER(C2) in a helper column shows which.
2. Search for letters in numeric columns. Use Find for O, l, S and B within the amount range.
3. Check magnitudes. Sort the amounts. A misplaced decimal point shows up as one value a hundred times bigger than its neighbours.
4. Reconcile to printed totals. Sum the column and compare with the total, subtotal or closing figure on the page. If they differ, the difference often points at the faulty row.
5. Check dates. Sort by date; out-of-sequence or impossible dates (13th month, 31 April) reveal misreads.
6. Spot-check against the image. Compare ten random rows with the original side by side.

What about scanned bank statements?

Statements are among the hardest scanned documents: dense small print, signs that depend on which column an amount sits in, and a running balance that must chain from row to row. The running balance is also your best proofing tool, because each row's balance should equal the previous balance plus the transaction. Any break in that chain flags an OCR error.

Our dedicated guide to converting a scanned bank statement to Excel walks through that check, and the broader OCR for bank statements guide covers tools. QuickBankConvert reads scans and phone photos, though less accurately than PDFs downloaded from online banking, so if you can download the original statement from your bank instead of scanning a paper copy, do that first and convert the clean file with the statement to Excel converter.

Related guides

• Scanned statements into a spreadsheet
• Choosing OCR software for statements
• Turning a digital PDF into Excel