Compare two columns

How to compare two columns in Excel

1. Open the file. Drop the .xlsx or .csv onto the box. It is read by JavaScript in this tab; no copy is sent anywhere.
2. Pick the two columns. Choose them from the dropdowns, which list the header names from the first row so you do not have to translate "column F" in your head.
3. Read all four answers. In both columns, only in the first, only in the second, and the repeats inside each column — computed together, not one at a time.
4. Take the result with you. Copy any of the three lists to the clipboard, or download it as CSV to paste back into your sheet.

"Compare two columns" is four questions, and the formula answers one

Almost every guide to comparing two columns in Excel gives you the same helper column: =COUNTIF(B:B, A2)>0, or a VLOOKUP wrapped in IFERROR, or conditional formatting with a duplicate-values rule. They work. What they have in common is that each one answers a single question per pass:

• Which values are in both columns — the overlap. Customers who are on the renewal list and also on the paid list.
• Which are only in the first — invoices you issued that the bank statement does not show.
• Which are only in the second — charges on the statement with no invoice behind them.
• Which repeat inside one column — usually the reason the two totals disagreed in the first place.

Running the COUNTIF in the other direction to get the second answer means writing it again against the other column, and the fourth question needs a different formula entirely. This page computes all four in one pass and shows them side by side, each with its own count, copy button and CSV download.

Columns are chosen by their header, not by a letter

The dropdowns list the header names from the first row, so you pick Email and Billing email rather than counting across to column F and column N. Turn the header checkbox off and the columns are labeled A, B, C instead, and the first row is treated as data.

The two columns do not have to be the same length, or in the same order, or next to each other — matching is by value, not by row position. They do have to be in the same sheet; for two separate files use the file comparison tool, and for values you have already copied to the clipboard, compare two lists takes pasted text.

The invisible differences that make a match fail

A comparison that reports zero matches on data you can see is identical is nearly always one of four things, and this page handles three of them without being asked:

• Trailing spaces. Always trimmed. A value exported from another system usually carries one, and it is invisible in the cell.
• Full-width characters. ＡＣＭＥ and ACME are different codepoints and compare as different text. They are folded together here (Unicode NFKC), which also fixes half-width katakana and accents typed as a letter plus a combining mark.
• Upper and lower case. Treated as different unless you turn that off — which matters for email columns, where Bob@example.com and bob@example.com are one mailbox.
• Numbers stored as text. This one you have to fix in the sheet: 1001 and "1001" look identical and are not the same cell value in Excel either.

Close but not identical: use fuzzy matching instead

This page matches values exactly, after the folding above. If your two columns hold the same customers spelled differently — Acme Corp. against ACME Corporation — exact matching reports two unmatched values and it is right to. That is what fuzzy matching is for; it scores every pair and lets you set how close is close enough.

Nothing leaves this tab

The columns people compare are email lists, customer IDs, invoice numbers and account numbers. The file is read in your browser and the comparison runs there: there is no upload to intercept, log or retain. Open the developer tools on the Network tab and drop a file — no new request appears.