How to Get Data from PDF to Excel: 4 Ways That Actually Work

Getting data from a PDF into Excel is one of the most common office tasks there is, and one of the most quietly frustrating. The data is right there on the page; it just refuses to arrive in your spreadsheet as usable rows and columns. There are four practical ways to do it, and each fits a different kind of document. This guide covers all four honestly, including when the free built-in options are all you need.

If your document is a scan or the table is complex, you can skip straight to the AI PDF to Excel converter: the first page converts free with no signup.

Method 1: Copy-Paste (Fast, Fragile)

Select the table in your PDF viewer, copy, and paste into Excel. For a small, simple, digital table this sometimes just works, and it's always worth ten seconds to try.

Why it usually fails: PDF stores characters and positions, not tables. On paste, multi-line descriptions split into extra rows, columns merge, and numbers arrive as text that breaks SUM. On a scanned PDF there is nothing to select at all.

Use it when: the table is small, digital, and single-page, and you don't mind fixing a few cells.

Method 2: Excel's Built-In "Get Data from PDF" (Free, Digital PDFs Only)

Excel can import PDF data directly, and surprisingly few people know it:

1. Open Excel (Microsoft 365 on Windows).
2. Go to Data → Get Data → From File → From PDF.
3. Pick your PDF. The Navigator shows each table and page Excel detected.
4. Select a table, click Load (or Transform Data to clean it in Power Query first).

This is genuinely good for clean, digital PDFs with simple tables, and it's already on your machine. Its limits are structural: it cannot read scanned PDFs (no text layer, nothing detected), it frequently mis-splits tables with merged cells or wrapped text, and each page's table arrives separately, so multi-page statements need manual stitching.

Use it when: the PDF is digital, the table is well-behaved, and you're comfortable with Power Query for cleanup.

Method 3: Adobe Acrobat's Export (Decent, Subscription)

Acrobat Pro's "Export a PDF → Spreadsheet" produces an .xlsx and handles simple documents reasonably. It has limited OCR for scans, struggles with complex and multi-page tables, and requires the subscription. If you already pay for Acrobat, try it; if you don't, it's not worth subscribing for this task alone.

Use it when: you already have Acrobat Pro and the document is straightforward.

Method 4: AI Extraction (Scans, Complex Tables, Handwriting)

AI-based extraction reads the rendered page the way a person does: it finds the table, understands which values belong together, and rebuilds the structure as real rows and columns. That difference is what handles the documents the first three methods can't:

• Scanned PDFs and photos, where there are no characters to extract: OCR and table detection run together
• Complex tables: merged cells, multi-line descriptions, empty cells, unusual layouts
• Multi-page tables, stitched into one continuous list with repeated headers removed
• Handwriting, including filled-in forms and handwritten registers
• Numbers as numbers, with currency symbols stripped and formats like 1,00,000 parsed correctly

With ScanPilot's PDF to Excel converter, you upload the PDF, preview the first page as a table in seconds (free, no signup), and download the full document as XLSX, CSV, or JSON. For specific document types, there are tuned versions: bank statements, scanned PDFs, and handwritten documents.

Use it when: the document is scanned, the table is complex, the file is long, or you've already watched methods 1-3 scramble it.

Which Method Should You Use?

Key Takeaways

• Try the free built-ins first on digital PDFs: copy-paste for tiny tables, Excel's Get Data for clean ones. When they work, they're instant.
• Scans change everything. No text layer means copy-paste and Excel's importer find nothing; you need OCR plus structure detection.
• Misaligned columns are a method problem, not a you problem. Position-based reconstruction breaks on real-world tables; AI reads the layout instead.
• Judge on your own document. The first page converts free, which tells you more than any comparison article, including this one.