Skip to content

CPD lab · AR

Remittances and split payment matching

A week of customer receipts for a wholesaler: match every payment to the invoices it pays, with your AI tool, and catch the payments it gets wrong.

All the ways to get this lab

Comes with AR, credit and cash, and with the year. 29 labs open now.

The brief

You work in credit control at Harlow & Pike Wholesale Ltd, a catering and hospitality wholesaler. It is Friday 30 October 2026 and this week's receipts need applying before the ledger closes.

The pack has the bank extract (15 receipts), the open-items ledger (40 invoices and 3 credit notes) and 12 remittance advices as they arrived: emails, letters, system printouts and one scanned by post. Every company and figure is invented, so you can paste it into any AI tool.

Use your AI tool to match every receipt to the invoices and credit notes it pays. Post nothing: your job is the cash application table, for a person to review and post.

Give one line for each invoice or credit note a receipt pays. A credit note taken off a payment is its own line with a negative amount. The amounts on a receipt's lines must add up to the receipt.

Action, one per line. MATCH: the evidence supports applying this amount to this item. QUERY: applied to the invoice, but the customer has paid short with no agreed reason, so the difference goes back to the customer. UNAPPLIED: the cash can't be matched on the evidence; leave invoice_ref blank and hold it unallocated. ESCALATE: applied to the invoice, but the difference needs someone with authority (a write-off, say).

A remittance is evidence, not proof: check every one against the ledger and the bank. A bank reference that quotes an invoice number and agrees to the amount is enough evidence on its own.

Then write a short review note: what the AI got wrong, what you checked, and what goes to a person.

Objective: Apply cash allocation controls when using an AI assistant to match customer receipts to open invoices, identifying combined, part and short payments and escalating unsupported items without posting entries.

What's in the pack

Every name, figure and email in the pack is invented. Use your own AI tool: Claude, ChatGPT, Copilot or Gemini.

  • bank-extract.csv

    Bank receipts, 26-30 October 2026 (15 lines)

  • open-items-ledger.csv

    Sales ledger open items at 26 October 2026 (40 invoices, 3 credit notes)

  • remittance-01-thornbury.txt

    Thornbury Garden Centres: system payment advice

  • remittance-02-pell.txt

    Pell & Daughters Catering: letter

  • remittance-03-ashdown.txt

    Ashdown Hotels: ERP remittance printout

  • remittance-04-meresby.txt

    Meresby Farm Shops: email

  • remittance-05-corbel.txt

    Corbel Interiors: payment advice

  • remittance-06-kilnbrook.txt

    Kilnbrook Cafe (Ireland): bank payment confirmation forwarded by the customer

  • remittance-07-dunmore-email.txt

    Dunmore Leisure: emailed remittance

  • remittance-08-dunmore-post.txt

    Dunmore Leisure: remittance received by post

  • remittance-09-wyke-scan.txt

    Wyke Valley Hotels: posted remittance, scanned and run through OCR

  • remittance-10-saltmarsh.txt

    Saltmarsh Deli: short email

  • remittance-11-osprey.txt

    Osprey Events: remittance exported from their accounts package

  • remittance-12-glenholt.txt

    Glenholt Care Homes: supplier portal notification

The traps

The pack has 8 deliberate traps: cases where AI tools often get it wrong and a person must check. We don't list them here. Finding them is the job, and your feedback names each one you missed.

What you submit

  1. Your result, as a table with the columns receipt_ref, invoice_ref, amount_applied, action, note.
  2. The prompt you used.
  3. A short review note, 150 to 350 words: what the AI got wrong, what you checked, and what goes to a person.

How it's marked

  • Your result, checked line by line against the answer key60%
  • Your review note, marked on traits40%

The review note's traits:

  • Split and combined payments found
  • Deductions queried, not written off
  • Unmatched items escalated
  • Nothing posted without evidence
  • Customer data kept out of personal accounts

Pass at 80. Under 80, read the feedback and try again. Then write your reflection and copy the entry for your CPD record.

Price

  • AR, credit and cash, £79. The role track and its labs, this one included. No end date.
  • Everything for a year, £199. Every course, role track and lab for 365 days.
  • Monthly, £19.99 a month, cancel any time. Every lesson, plus 8 marked tasks a month, across every course, role track and lab.

Buying for a team, or want a lab built on your own process? CPD for employers.