Every week I saw the same thing at work: someone opening a PDF, manually copying a table into Excel, cell by cell. 30 minutes of work that should take 2 seconds.
So I built a tool to fix it.
What it does
python extract_tables.py financial_report.pdf
That's it. You get a formatted .xlsx file with every table from the PDF, organized into separate sheets with styled headers, alternating rows, and auto-fitted columns.
The tech stack
Three libraries. That's all:
- pdfplumber — opens the PDF and extracts table data page by page
- pandas — structures the raw data into DataFrames
- openpyxl — writes to Excel with formatting
No OCR. No Java. No external APIs. Pure Python.
The core extraction — 15 lines
import pdfplumber
import pandas as pd
def extract_tables_from_pdf(pdf_path):
results = []
with pdfplumber.open(pdf_path) as pdf:
for page_num, page in enumerate(pdf.pages, start=1):
tables = page.extract_tables()
for t_idx, raw_table in enumerate(tables, start=1):
header = raw_table[0]
rows = raw_table[1:]
df = pd.DataFrame(rows, columns=header)
results.append({"page": page_num, "table_index": t_idx, "dataframe": df})
return results
pdfplumber does the heavy lifting. It detects table boundaries from the PDF's internal structure — no guessing, no heuristics for basic cases.
Why it's useful for real work
PDF reports are everywhere: government data, supplier invoices, financial statements, logistics documents. Most of them have tables. None of them are easy to work with.
This tool is for the analyst who gets a 20-page PDF every Monday morning and needs the data in Excel by 9am.
Get the full code
GitHub: github.com/YOUR_USERNAME/pdf-to-excel
Includes the full CLI script, a sample PDF generator for testing, and instructions to run it in under 2 minutes.
What's coming next
- Batch mode for processing multiple PDFs at once
- Streamlit UI for non-technical users
- OCR support for scanned PDFs
This is part of a Python automation toolkit I'm building in public. If you work with PDFs regularly, try it and let me know what breaks — real-world PDFs are always messier than test cases.
SOCIAL SHARE CARD GENERATOR