Detect Anomalies in Excel

Find the errors in a spreadsheet before they cost you

Most bad data is not really bad. It is a trailing space that breaks a lookup, a date written the American way in a British column, one duplicated ID, a price with an extra zero. This tool reads a sheet and reports what looks wrong, row by row, so you can fix it before it reaches a mail merge, an import, or a board pack.

What it checks

• Duplicate rows — both exact repeats and near-duplicates that differ only in spacing or capitalisation
• Formatting slips — leading and trailing whitespace, double spaces, unusual characters, inconsistent capitalisation, and dates written in a different format from the rest of their column
• Missing values — empty cells in a column that is otherwise almost always filled in, which usually means something was skipped rather than genuinely blank
• Unexpected types — text sitting in a column that is otherwise numeric, or a number in a column that is otherwise text
• Outliers — numeric values far outside the normal range for their own column, found using the interquartile range rather than a fixed threshold
• Repeated identifiers — the same value twice in a column that looks like it should be unique, such as an ID, email, SKU or username

Every column is judged against itself

The thresholds are deliberately conservative. Each check compares values against the rest of their own column rather than against a universal rule, so a column that is legitimately full of symbols, or entirely lowercase, is not flagged simply for being that way. Only the minority that breaks its column's own pattern gets reported. The result is a short list worth reading rather than hundreds of findings you learn to ignore.

How it works

1. Upload a .xlsx or .xlsm file and choose a sheet.
2. Run the check. Results appear on screen as a table: the row number as you would see it in Excel, the column, what kind of issue it is, and a plain-English explanation of why it was flagged.

Your file is never modified. This tool only reads and reports, so nothing is changed or corrected behind your back — you decide what is a real problem and fix it in Excel.

What people use it for

• Checking a mailing list before a merge, so nobody gets two letters
• Cleaning a survey or form export before analysis
• Sanity-checking a finance or stock sheet before it is circulated
• Reviewing data inherited from someone else before trusting it
• Catching import errors before they reach a CRM or database

Common questions

Does it change or fix my file?

No. It is read-only. You get a list of what looks wrong; the corrections are yours to make.

Why was something flagged that is actually fine?

Because it differs from the rest of its column. A genuinely one-off value, an unusual but correct name, a deliberate outlier — all can be flagged. The tool reports what stands out; judging what matters is your call.

Why was something not flagged?

Checks need enough data to establish a pattern. A very short sheet, or a column where the unusual values are more common than the usual ones, will produce fewer findings by design.

Can I download the results?

Yes. Export gives you the whole list as an .xlsx: a summary sheet with the counts by issue type, and one row per issue with its row number, column, type and detail. It draws on the same free allowance as the check itself, so exporting a sheet you have just checked doesn't cost you a second file.

What are the limits?

.xlsx and .xlsm, up to 10 MB, one sheet at a time. Three checks a month without an account. Your file is deleted within the hour.