23 questions foundML improves matching by learning from historical application patterns, parsing unstructured remittance (emails, PDFs), predicting likely invoices for a payment, and auto-resolving common deductions. It raises straight-through rates beyond rule-based systems and adapts to customer behaviors, leaving fewer, harder exceptions for staff.
Real-world example An ML model reads emailed remittance and lifts the auto-match rate from 80% to 92% over a few months.
Common follow-ups: How does ML handle unstructured remittance? | Why does it outperform static rules?
Reconciliations Aging Analysis Cash Application
Applying a receipt reduces the specific open invoice balances it's matched to and lowers the customer's total AR. Fully paid invoices are marked closed; partially paid ones show a residual. The GL AR control account and the customer's sub-ledger both decrease by the applied amount.
Real-world example Applying $5,000 across two invoices closes them and drops the customer's AR balance by $5,000.
Common follow-ups: What happens to a partially paid invoice? | How do the GL and sub-ledger move?
Aging Analysis Reconciliations Cash Application
When a customer pays early and deducts an agreed settlement discount, apply the cash received and record the discount as a reduction of revenue (sales discount) or expense, clearing the invoice in full. The discount is the difference between the invoice and the cash received under the agreed terms.
Invoice 1,000, 2% settlement discount taken, cash 980:
Dr Bank 980 Dr Sales Discounts 20 Cr AR 1,000
Real-world example A customer pays a $1,000 invoice at $980 under 2/10 terms; AR clears fully and the $20 discount is booked.
Common follow-ups: Where is a sales discount recorded? | How does it clear the invoice fully?
Credit Control Reconciliations Cash Application