CSV Formatting Rules for Ecommerce Imports

The CSV Standard Ecommerce Platforms Follow

All major ecommerce platforms — Shopify, WooCommerce, Etsy, eBay, Amazon — expect files that follow RFC 4180, the common CSV standard. That standard defines exactly how rows, columns, quoting, and line endings must be structured.

This guide covers every rule that affects import success. Violating any of these causes parsing errors, wrong column mapping, or silent data corruption.

Rule 1: Commas Are the Delimiter

Use commas to separate columns. Do not use semicolons (common in European Excel locales), tabs (TSV format), or pipes (|).

Correct:

Handle,Title,Vendor,Variant Price
blue-shirt,Blue T-Shirt,Acme,19.99

Wrong (semicolons):

Handle;Title;Vendor;Variant Price
blue-shirt;Blue T-Shirt;Acme;19.99

If your Excel uses semicolons by default (because your system locale uses a comma as the decimal separator), change the system list separator to a comma or export via Google Sheets, which always produces comma-delimited output.

Rule 2: The First Row Is Headers Only

Row 1 must contain column names and nothing else. No blank rows above it, no title rows, no merged header cells.

Correct:

Handle,Title,Variant Price

Wrong:

Product Export -- June 2026
Handle,Title,Variant Price

The platform reads row 1 as headers. Any non-header content in row 1 is treated as column names, causing every header to be mismatched.

Rule 3: Headers Are Exact

Column names must match the platform's expected names exactly:

• Case-sensitive: Handle is not the same as handle or HANDLE on Shopify
• No leading/trailing spaces: Title (space before) will not match Title
• No line breaks inside headers: a header cell must be a single line
• No duplicate headers: if two columns have the same name, most parsers use only the last

Rule 4: Quoting Required for Commas, Quotes, and Newlines

Whenever a cell value contains a comma, double-quote character, or newline, wrap the entire value in double-quote characters.

Correct:

"Leather Wallet, Brown",29.99,"Handmade in the USA, ships free"

Escaping quotes inside a quoted value — double the inner quote character:

"She said ""great product"""

This represents: She said "great product"

| Situation | Correct format | |---|---| | Value with comma | "value, with comma" | | Value with double quote | "value ""with"" quotes" | | Value with HTML | "<p>Description here</p>" | | Value with newline | "Line one\nLine two" (inside quotes) | | Plain number | 19.99 (no quotes needed) |

Rule 5: UTF-8 Without BOM

Save your file as UTF-8 encoding, without a BOM (Byte Order Mark).

• UTF-8 is the only encoding all platforms accept
• The BOM (added by Excel on Windows) corrupts the first column header name
• ANSI, Latin-1, and Windows-1252 files produce garbled characters for accented letters, curly quotes, and currency symbols

Check: Open the file in VS Code. The status bar shows the encoding. It should say UTF-8 — not UTF-8 with BOM, Windows 1252, or ISO 8859-1.

Rule 6: Consistent Line Endings

Use either CRLF (\r\n, Windows) or LF (\n, Unix/macOS) throughout the file. Never mix them.

Mixed line endings cause some parsers to see blank rows between data rows, or to miscount columns in multi-line values.

Check in VS Code: The status bar shows CRLF or LF. Click it to change the line ending, then save.

Rule 7: One Row Per Record (Except Platforms With Multi-Row Products)

Each row is one independent record. The exception: Shopify uses multiple rows sharing the same Handle to represent a product with variants.

For all other platforms, a row = one product. Multi-row product structures only apply where the platform's official documentation explicitly describes them.

Rule 8: No Blank Rows

Blank rows between or after data rows cause parsers to:

• Insert empty records into the import
• Miscount the number of products
• Stop processing at the first blank row (some platforms)

Delete all blank rows before saving. In Excel, press Ctrl+End to jump to the last used cell — if it's beyond your data, blank rows exist and must be deleted.

Rule 9: No Summary or Formula Rows

Remove any rows that are not product data:

• Total/subtotal rows at the bottom
• SUM() formula rows
• Filter header rows added by Excel's auto-filter
• Section dividers

These rows appear as product records in the import and cause type errors (text where numbers are expected, missing required fields, etc.).

Rule 10: No Merged Cells

Merged cells in Excel export as a value in the top-left cell and empty strings in all others. The importer sees missing required fields on every row after the first.

Fix: Select all → Unmerge cells → Fill blank cells with appropriate values before saving as CSV.

Quick Validation Checklist

• [ ] Comma-delimited (not semicolons, tabs, or pipes)
• [ ] Row 1 is column headers only — no blank rows above it
• [ ] Headers match platform's expected column names exactly (case and spaces)
• [ ] No duplicate column names
• [ ] Values with commas or quotes are wrapped in double quotes
• [ ] File encoding is UTF-8 without BOM
• [ ] Line endings are consistent (all CRLF or all LF)
• [ ] No blank rows anywhere in the file
• [ ] No summary rows, formula rows, or merged cells
• [ ] File has a .csv extension and is plain text (not .xlsx renamed to .csv)

Fix This Automatically with StriveFormats

Upload your CSV and StriveFormats checks every rule above automatically. It flags violations with exact row and column locations, auto-fixes safe issues, and exports a clean file.

Open CSV Fixer | View CSV Templates