Turn a messy email thread into a table you can paste into a spreadsheet

Plenty of spreadsheets start life as somebody retyping an email thread. Don't. Paste the mess, name the columns you want, and let the AI do the typing.

Extract the items from the text below into a table with these columns:
[COLUMN 1], [COLUMN 2], [COLUMN 3], [COLUMN 4].
Rules:
- One row per [ITEM, e.g. action, supplier, incident].
- If a value isn't in the text, write "NOT STATED". Don't guess.
- Dates as DD/MM/YYYY. Amounts as plain numbers with no currency symbol.
- Output as CSV in a code block, with a header row.
- After the table, list anything in the text that didn't fit a row.

Text: [PASTE]

Three lines in there do most of the work. "NOT STATED, don't guess" stops blank cells filling up with sensible-sounding nonsense, which in a spreadsheet looks exactly like data. The fixed formats stop "next Tuesday", "1st Oct" and "£1,200.00" turning up in the same column. And the leftovers list tells you whether anything got dropped.

CSV in a code block is the format most likely to survive the trip. Save it as a .csv, or paste it into a spreadsheet and use Data > Text to Columns. Try one paste before you rely on it, because chat windows have a habit of turning tabs into spaces.

Then check it: count the rows against the source and spot-check three. If there's a lot of data, leave the totals to the spreadsheet.

Anything personal or confidential stays out unless the tool's approved for it. Redact first if you need to.