How to Convert Excel to CSV (Preserve Data, Handle Encoding, Batch Export)

Your data lives in Excel. The database you are loading it into expects CSV. The AI tool you are using to analyze it prefers plain text. The API you are calling needs comma-separated values. This happens dozens of times a day in data work, and the conversion looks trivial — until a column of phone numbers loses its leading zeros, a date column becomes ambiguous between US and European formats, or an entire CSV file displays as question marks because of an encoding mismatch. This guide covers not just how to convert, but how to do it without the common pitfalls.

1. TL;DR — Three Ways to Export CSV

2. What Gets Lost in the Conversion

CSV is a plain-text format — it stores rows and columns of values, nothing else. The gap between what XLSX supports and what CSV can represent is significant. Understanding exactly what is lost helps you decide whether CSV is appropriate and what to keep in the XLSX as the source of truth.

Formulas becoming values is usually exactly what you want when exporting for a database or API — the downstream system needs the number, not the formula. But it means the CSV is a snapshot at export time. If the underlying data changes, you need to re-export. Keep the XLSX as the live working file; treat the CSV as a read-only export for a specific point in time.

3. Method 1: Excel Save As

The native Excel export is the fastest method for single sheets and requires no additional software.

4. Method 2: Google Sheets Export

Google Sheets is useful when you need to convert an XLSX file but do not have Excel installed, or when you want to quickly inspect the file before exporting.

When to use Google Sheets over Excel:

• You do not have Excel installed and need a free desktop option
• You want to share the conversion process with someone who does not have Excel
• You need to do light data cleanup before exporting (search and replace, column reordering)
• You want to inspect the file in a different application to catch any import issues before the final export

5. Method 3: Browser-Based Converter

Convertlo's XLSX to CSV converter runs entirely in your browser using JavaScript. Nothing is uploaded to a server — your file stays on your device. This is the right choice when you need to convert quickly without opening Excel or Google Sheets, or when working with files that should not leave your computer.

6. Method 4: Python pandas

Python with the pandas library is the right tool when you need to convert multiple files, automate the conversion as part of a larger pipeline, or need to transform the data during export (rename columns, filter rows, change date formats).

Basic Single-File Conversion

import pandas as pd

# Read the first sheet (default)
df = pd.read_excel('data.xlsx')

# Export to CSV with UTF-8 encoding, no index column
df.to_csv('data.csv', index=False, encoding='utf-8')

Exporting All Sheets to Separate CSV Files

import pandas as pd

# Open the workbook
xl = pd.ExcelFile('workbook.xlsx')

# Export each sheet to a separate CSV file
for sheet_name in xl.sheet_names:
    df = xl.parse(sheet_name)
    # Clean the sheet name for use as a filename
    filename = sheet_name.replace(' ', '_').replace('/', '-') + '.csv'
    df.to_csv(filename, index=False, encoding='utf-8')
    print(f"Exported: {filename} ({len(df)} rows)")

Batch Converting Multiple XLSX Files

import pandas as pd
import glob
import os

# Find all XLSX files in the current directory
xlsx_files = glob.glob('*.xlsx')

for xlsx_path in xlsx_files:
    # Create output filename
    csv_path = os.path.splitext(xlsx_path)[0] + '.csv'

    # Read and export first sheet
    df = pd.read_excel(xlsx_path)
    df.to_csv(csv_path, index=False, encoding='utf-8')

    print(f"{xlsx_path} ({len(df)} rows) -> {csv_path}")

Fixing Date Format During Export

import pandas as pd

df = pd.read_excel('sales.xlsx')

# Convert date columns to ISO 8601 format (YYYY-MM-DD)
# This prevents ambiguity between US and European date formats
date_columns = ['order_date', 'ship_date', 'invoice_date']
for col in date_columns:
    if col in df.columns:
        df[col] = pd.to_datetime(df[col]).dt.strftime('%Y-%m-%d')

df.to_csv('sales.csv', index=False, encoding='utf-8')

Install pandas if you don't have it: pip install pandas openpyxl. The openpyxl package is required for pandas to read XLSX files.

7. Encoding Issues: UTF-8 vs UTF-8 BOM vs CP1252

Encoding problems are the most common source of "corrupted" CSV files. The symptom: you open a CSV and see question marks, boxes, or scrambled characters where accented letters (é, ü, ö), French characters (ç, à), or non-Latin scripts should be. Understanding why this happens makes it easy to fix.

Why Encoding Corruption Happens

When a CSV file is saved, each character is stored as a number. Different encoding systems map different numbers to different characters. The most common encodings:

• UTF-8 — the universal standard. Handles all languages. Every character above basic ASCII uses 2-4 bytes. The correct choice for any new CSV file.
• UTF-8 BOM (UTF-8-SIG) — UTF-8 with a Byte Order Mark at the start of the file. Some older Windows tools (including older versions of Excel) add this automatically. Most modern tools handle it correctly, but some tools or parsers choke on the BOM. Use regular UTF-8 unless a specific tool requires BOM.
• CP1252 (Windows-1252) — the legacy Windows encoding for Western European languages. Handles accented Latin characters (é, ü, ö, etc.) but corrupts anything outside Western European scripts. This is what Windows Excel uses when you choose plain "CSV" rather than "CSV UTF-8".
• Latin-1 (ISO-8859-1) — similar to CP1252, slightly different mappings for a small range of characters. Legacy Unix encoding.

Fixing an Encoding Problem in Python

import pandas as pd

# If your CSV was saved in CP1252 (Windows legacy) and shows garbled characters
df = pd.read_csv('corrupted.csv', encoding='cp1252')

# Re-save as UTF-8
df.to_csv('fixed.csv', index=False, encoding='utf-8')

Checking Encoding of an Unknown CSV File

import chardet

with open('mystery.csv', 'rb') as f:
    result = chardet.detect(f.read())
    print(result)  # {'encoding': 'CP1252', 'confidence': 0.73}

Install chardet with pip install chardet. The confidence score indicates how certain the detection is — above 0.9 is reliable, below 0.7 is a guess.

8. Handling Multi-Sheet Workbooks

Excel workbooks routinely contain multiple sheets — monthly data across 12 tabs, different data categories, summary and detail sheets. CSV is a single-table format, so exporting a multi-sheet workbook to CSV always requires multiple CSV files.

Strategy for Naming Multiple CSV Files

The naming convention matters because the downstream system will need to identify which data is which. Three common approaches:

• Sheet name as filename: january.csv, february.csv, march.csv — clear and self-documenting
• Workbook prefix + sheet name: sales-2026-january.csv — useful when multiple workbooks produce CSV files that go to the same folder
• Positional naming: data-01.csv, data-02.csv — avoid this; it loses the meaning of the sheet names

Which Sheets to Export

Not every sheet in a workbook contains data that needs to be exported. Common sheets to skip:

• Summary/Dashboard sheets — these are usually formula-driven views of the underlying data; export the source sheets instead
• Lookup tables — if the downstream system has its own lookup data, these may be redundant
• Documentation sheets — README-style sheets explaining the workbook structure are useful for humans, not for data processing
• Chart-only sheets — charts are entirely discarded in CSV; these sheets produce empty CSV files

When in doubt, export all data sheets and let the downstream system use what it needs. An extra CSV file that is not consumed causes no harm. A missing CSV file causes import failures.

9. CSV vs XLSX: What to Use When

10. Common Problems and Fixes

Problem: Phone numbers and zip codes lose leading zeros

A phone number stored as 07911123456 in Excel exports as 7911123456 in CSV, dropping the leading zero. This happens because Excel stores it as a number, and numbers do not have leading zeros. The fix: before exporting, format the column as Text in Excel (right-click the column → Format Cells → Text). Or, after exporting, import to your database with that column typed as VARCHAR rather than INTEGER.

Problem: Dates are ambiguous (01/05/2026 — May 1 or January 5?)

Date format ambiguity is a persistent problem when CSV files cross regional boundaries. Fix before exporting: in Excel, select the date column, press Ctrl+1, choose Custom format, and type yyyy-mm-dd. ISO 8601 format is unambiguous in every locale and reads correctly in all programming languages and databases.

Problem: Commas in data values break CSV parsing

A cell containing the value London, UK will break naive CSV parsers that split on every comma. Excel correctly wraps such cells in double quotes when exporting: "London, UK". Most parsers handle RFC 4180 quoting correctly, but some legacy tools do not. If you are using a strict parser, consider replacing commas within values before exporting, or switch to TSV (tab-separated values) as the delimiter when commas are common in your data.

# Python: export as TSV instead of CSV when commas appear in values
df.to_csv('data.tsv', index=False, encoding='utf-8', sep='\t')

Problem: CSV opens in Excel with all data in one column

This happens when Excel's auto-detection picks the wrong delimiter. Instead of double-clicking the file, import it deliberately: in a blank Excel sheet, go to Data → From Text/CSV → select the file → in the import wizard, set encoding to UTF-8 and delimiter to comma. In European locales where semicolons are the default delimiter, select semicolon instead.

Problem: Line breaks inside cell values create extra rows in CSV

If a cell contains a newline character (Alt+Enter in Excel), the CSV export wraps the cell content in quotes and the newline is part of the value. RFC 4180-compliant parsers handle this correctly, but simple line-by-line parsers break. Check for and remove embedded newlines before exporting:

import pandas as pd

df = pd.read_excel('data.xlsx')

# Remove line breaks from all string columns
df = df.applymap(lambda x: x.replace('\n', ' ').replace('\r', '') if isinstance(x, str) else x)

df.to_csv('clean.csv', index=False, encoding='utf-8')

Problem: Numbers formatted as text export with apostrophes or quotes

Some Excel files contain numbers that were entered with a leading apostrophe to force text storage (a common trick to preserve leading zeros). These export to CSV with the apostrophe preserved, which breaks numeric parsing downstream. Fix: in Excel, select the column, Data → Text to Columns → Finish (no changes) to reparse and remove the text flag, then re-export.

11. Frequently Asked Questions

The Excel-to-CSV conversion looks simple but has several traps that catch people regularly: encoding selection, one-sheet-at-a-time exports, leading zeros in numeric fields, and date format ambiguity. Each trap has a straightforward fix once you know where to look. For the vast majority of use cases, "File → Save As → CSV UTF-8" solves it in three clicks. For automation or bulk work, the Python pandas approach handles any volume reliably.