PROMPT BRO
Accounts desk

Find likely matches for unknown receipts

Bank narration: 'TRANSFER'. Thank you for the detailed explanation.

Tested Claude Opus 5.5, medium effort · 15 Nov 2026 · on the practice files

By hand1 hr
With AI10 min
Per week · estimate
Easier on your office computer.Send to my computer
I usePick yours. We'll show the right steps.

Get receipt and invoice lists

Use receipt ref, amount, narration hint and open invoice ref, customer code, due amount. Replace bank and UPI identifiers first.

First time? Try it on fake data first:

Check the amount basis

Use actual credited and outstanding amounts; no assumed TDS or fees.

Privacy Keep receipt reference and amount; remove account numbers, customer UPI IDs and phone numbers. Delete bank account numbers; replace customer UPI IDs and phone numbers.

Fill the orange boxes
You send (the "prompt": the message you paste into your AI tool)
receipts.xlsxopen-invoices.xlsx

Whenever I send masked receipts and open invoices for [Period], find possible matches. Rule: [Match rule]. A. For each receipt list exact amount candidates; if hint names a matching invoice, show it. If multiple candidates without hint, mark ambiguous; if none, no match. Treat likely matches as proposals only. If two receipts have the same invoice as their explicit matching hint, hold both as a conflict; do not let row order decide. If an unhinted receipt has several exact candidates in the original invoice list, keep it ambiguous even if another receipt has a likely hint. Never propose one invoice for two receipts. Hold invalid, missing or non-positive amounts, duplicate receipt or invoice references and hints that conflict with the amount; ask for verified allocation details. B. Table of likely, ambiguous and no match with receipt and invoice refs, amount and reason. C. Total received and amounts by status; draft questions for accounts/customer to verify. No automatic allocation or TDS guess. Use only my data. GST rates, HSN and TDS rates only from my supplied figures; never guess tax treatment. This is a first pass; a person checks and posts any final entries in Tally/Busy.

Set it up once. Save the matching rules; keep actual data out until needed. Pick your tool at the top to see how.

After that, each time: send new masked receipts and current open invoices.

First time? Try it in a normal chat: paste and send.

Which model? Use your tool's strongest model with thinking turned on.

In your AI tool: attach the files with the + button, paste, send.

What you'll get back (example, from the practice files)
3likely
1ambiguous
1no match
  • R2₹5,000 ambiguous. I2 and I3 have the same amount; no hint.
  • R4₹3,000 no match. No invoice has this outstanding amount.
  • R5I3 hinted. ₹5,000 matches I3; keep R2 unresolved until verified.
Before you trust it

Then send this: Recheck every amount and hint, count original candidates, identify competing proposals and duplicate refs, and reconcile status amounts to total known receipts. List invalid/missing amounts separately.

If it goes wrong
It picks one equal amount

Reply: "List all candidates; don't choose without evidence."

It guesses TDS

Reply: "Exact amounts only; no deductions assumed."

Narration has personal data

Replace it with the useful invoice hint only.

Sir ne bola urgent. Ab bolo: ho gaya.

Did it work?