Clean and Rename Excel Column Headers in Bulk

Trailing-space and punctuation headers

The classic schema-prep set: a trailing space hidden after First Name, a # in Order #, and a parenthetical unit in Date (UTC). You type clean identifiers; data rows are untouched.

Detected headers (shown trimmed):
  First Name   →  [ first_name        ]
  Order #      →  [ order_id          ]
  Date (UTC)   →  [ date_utc          ]
  Amount       →  [                   ]  (blank = keep 'Amount')

Output header row:
first_name,order_id,date_utc,Amount

Data rows below are copied through byte-for-byte.

Keeping most headers, renaming a few

You rarely need to rename everything. Leave a field blank and the original header is kept exactly. The run button counts only the fields you actually changed.

Headers:  id | Customer Name | e-mail | created at
You type: (blank) | customer_name | email | (blank)

Button reads: 'Rename 2 headers'

Output header row:
id,customer_name,email,created at

Watch out: accidental collision

There is no collision check. If you rename two columns to the same identifier, you get two identical headers — which most importers reject or silently mangle. The preview is your safety net.

Headers:  Email | Email Address
You type: email | email          ← both renamed to 'email'

Output header row (no warning):
email,email

Fix: give the second a distinct name (email_secondary).

XLSX in, XLSX out (formatting flattened)

Upload an .xlsx, get an .xlsx back. Because the tool round-trips through CSV, the output is a clean single-sheet workbook of plain values — any formulas, number formats, or cell styling in the original are not carried over.

Input:  quarterly_report.xlsx (3 sheets, formatted, formulas)
Rename: Q1 Sales -> q1_sales,  Q2 Sales -> q2_sales

Download: quarterly_report.renamed.xlsx
  - Only Sheet1 (the first sheet) is present
  - Cells hold computed values, not formulas
  - Number formats / fills are not preserved

Empty header cells

If a column has no header text in row 1, it is listed as Col N so you can still name it. This is common in exports where a key column was left unlabelled.

Headers:  id | (empty) | total
Displayed: id | Col 2 | total
You type:  (blank) | line_item | (blank)

Output header row:
id,line_item,total

The tool maps each column index to the name you type with no uniqueness check. Renaming both Email and Email Address to email produces two identical headers in the output with no warning. Most importers (pandas, PostgreSQL COPY, BigQuery) then either reject the file or silently keep only one column. Always scan the preview header row for duplicates before downloading.

Only the first sheet of an .xlsx/.xls/.ods is read (SheetNames[0]) and the download contains a single Sheet1. Other tabs are dropped. If you need every sheet renamed, process each sheet separately, or split the workbook first. The tool does not merge or iterate sheets.

XLSX input is converted to CSV in-browser, then the renamed CSV is written back to a brand-new XLSX. Formulas become their last computed values, and number formats, fills, fonts, and merged cells are not preserved. If you need the original formatting intact, rename headers in Excel directly — this tool is for producing a clean, import-ready values-only file.

Some exports put a title in row 1 and the real headers in row 2. The tool always treats row 1 as headers, so a two-row or banner header is misread. Delete the banner row in your source before uploading, or the rename targets will be the wrong strings.

Headers are trimmed when listed, so First Name shows as First Name and the replacement you type is a clean string. The original whitespace does not survive into the output header because you are supplying the new value. This is helpful for schema prep where a trailing space would break an exact-match column lookup downstream.

The run button stays disabled until at least one header is changed — it reads Edit any header to rename when no field has a value. There is no no-op pass that re-emits the file unchanged; rename at least one header (or use a sibling tool if you only need a format conversion).

Header Rename is Pro-gated (minTier: pro). Free accounts can open the page but hit the upgrade wall before running. The Excel hub card advertises the Excel free limit (5 MB / 10,000 rows), but the rename action itself requires Pro and enforces CSV-family limits (2 MB free / 100 MB Pro) when run.

The tool stores whatever you type as the new header — including 2024_revenue. It does not validate identifiers, so a name starting with a digit is written as-is. That is fine for a CSV column label but illegal as a Python attribute or unquoted SQL identifier; type a safe form like revenue_2024 if the downstream consumer needs it.

There is no reserved-word or character filter. If you type select or order by as a header, it is written verbatim. For database targets, prefer non-reserved snake_case; the tool will not stop you or quote the name for you. See the PostgreSQL import guide for safe-identifier rules.