Back to library
Copilot#data analysisStandard promptSaves ~15min

Find anomalies in an Excel expense column

Copilot in Excel scans an expense column for statistical anomalies, duplicates, suspicious round numbers, and new vendors, then ranks the top 10 to review.

rach_maeve4 June 2026
0.0(0)

Fill in the Template

0 / 8 filled

Your Built Prompt

Updates live as you type
I have an expense column in Excel and I want Copilot to flag the rows that look suspicious so I can review them before BAS.

In Excel, open the workbook and click Copilot in the ribbon. Make sure the data is in a Table (Insert > Table) so Copilot can reason about it.

Sheet and table name: {{sheet_table}}
The expense column (header + letter): {{expense_col}}
Date column: {{date_col}}
Vendor column: {{vendor_col}}
Category column: {{category_col}}
Time window to scan: {{window}}
What "normal" looks like for me (typical range, frequency): {{normal}}
Specific red flags I want surfaced: {{red_flags}}

Ask Copilot:

Analyse the {{expense_col}} column in {{sheet_table}} for {{window}}. Flag any rows where:
1. The amount is more than 2 standard deviations from the median for that vendor.
2. The same vendor has duplicate transactions within 3 days for similar amounts.
3. The amount is a round number over $500 (possible estimate, not real receipt).
4. The vendor is new (first appearance) and the amount is over $200.
5. Any item in {{red_flags}}.

Output:
- A summary count of flagged rows by category of anomaly.
- A new column in the table called "Review flag" with the reason for the flag.
- The top 10 flagged rows in a short list, ranked by dollar impact.
- The one row I should look at first, with a one-line reason.

AU English. AUD amounts.
Run in

Fill in the required fields above to copy or run this prompt.

This prompt produces general information, not financial or tax advice. Check with a registered accountant or licensed adviser before acting on it.

Tags

Community Feedback

0 comments
Be the first to leave feedback on this prompt.

Related Prompts

More in Data Analysis