Train Your Team to Record Expenses with AI
Build a shared expense recording process that works the same way for everyone
Each person sends expense data in a different format. You spend hours reformatting entries and chasing missing details before you can reconcile accounts.
That is the gap this drop closes. It is a finance & admin skill built for Gemini, and it takes about ten minutes to set up the first time. After that it runs in under a minute.
Who should use it
Finance managers and team leads rolling out AI expense tools to staff
How it works
The skill file does five things, in this order.
1. Define your required fields once. List exactly what data every expense entry must contain: date, vendor, category, amount, VAT, payment method, project code.
2. Create a standard team prompt. Write one master prompt that extracts those fields consistently, then share it as a template everyone copies.
3. Set validation rules in the output. Ask Gemini to flag missing VAT numbers, dates outside the current month, or amounts without currency codes so errors surface immediately.
4. Run a test round with sample receipts. Give the team five practice receipts, have them submit results, then compare outputs to catch misunderstandings before real expenses flow in.
5. Document edge cases as they appear. When someone encounters a receipt type the prompt handles poorly, update the shared template and notify the team so the fix applies everywhere.
What comes back
A documented expense recording process that produces uniform data quality regardless of who submits
The mistake to avoid
Create a shared spreadsheet where the team logs which receipts gave Gemini trouble and how they fixed it. This becomes your living troubleshooting guide and reduces repeat questions.
Running it
Save the prompt as a Gem so it is reusable, or paste it into Gemini with your files attached. If the task touches Sheets, Docs or Gmail, run it from the Workspace side panel so it can read the file you already have open.
Where this fits
On its own, one skill saves an hour a week. The compounding happens when three or four of them run in sequence on the same input, the same transcript that produces a scope of work also produces the follow-up email and the project brief. That is the point at which it stops being a prompt and starts being an internal tool. If you want that wired into the systems your team already uses, that is the work 67 Digital does.
In the file
You are an expense data extraction assistant for our finance team.
Extract the following fields from the receipt image provided:
- Transaction date (format: DD/MM/YYYY)
- Vendor name
- Expense category (meals, transport, supplies, equipment, other)
- Total amount with currency code
- VAT amount if shown
- VAT registration number if present
- Payment method (cash, card, bank transfer)
- Project code if mentioned on receipt
Validation requirements:
- Flag if VAT registration number is missing for amounts over {vat_threshold}
- Flag if transaction date is outside {current_month}
- Flag if currency code is not AED, USD, or EUR
- Flag if any required field cannot be read from the image
Return the data in this exact structure:
{format_example}
If multiple items appear on one receipt, extract each as a separate line.
For any flagged issues, add a note in the "Validation alerts" field explaining what needs manual review.