As of 13 August 2026, AI can only partly send overdue invoice reminders.

What to actually do

1. Hand it to a person

   The route this page recommends

   A person who owns the outcome does this end to end, worth it when the failure is dear.

2. Use a tool built for this

   Second choice

3. Do it yourself

   The distant third

   A chat interface, power-user skill, and roughly 30 minutes until you can act on the result.

   How to actually do it

   1. Open your accounting system and export the overdue invoice list with invoice number, customer, recipient, balance, due date, payment status and dispute status.
   2. Open your payment records and customer contact records, then remove or mark invoices that are paid, disputed, cancelled, missing a verified recipient or otherwise ambiguous.
   3. Paste the cleaned records, payment instructions, sender details and escalation policy into the prompt and ask the chatbot to produce the eligibility list, workflow and draft messages.
   4. Compare every proposed recipient, invoice number, balance, due date and payment instruction with the accounting system before approving any message.
   5. Build the approved workflow in your accounting or email automation system, with a human approval step, duplicate-send protection and exclusions for paid or disputed invoices.
   6. Run the workflow against test records or a small internal test group, check the email content and audit trail, then approve the remaining eligible reminders for sending.
   7. Record replies, bounced messages, payments and disputes in the accounting system, and stop further reminders for any invoice whose status changes.

   Prompt

   Help me design and draft an overdue invoice reminder workflow for a UK business. Use only the records and policies I provide, and do not invent invoice details, payment terms, customer names, contact details, balances, dates or legal claims.
   
   Business name: [BUSINESS NAME]
   Payment instructions: [PAYMENT INSTRUCTIONS]
   Approved sender name and email: [SENDER DETAILS]
   Tone: [PLAIN, FIRM OR FRIENDLY]
   Escalation policy: [ESCALATION RULES]
   
   Invoice records:
   [PASTE A TABLE WITH INVOICE NUMBER, CUSTOMER NAME, CUSTOMER EMAIL, INVOICE DATE, DUE DATE, BALANCE, CURRENCY, PAYMENT STATUS, DISPUTE STATUS AND ACCOUNT OWNER]
   
   First, identify which records are eligible for a reminder using the due date, payment status and dispute status. Exclude paid, disputed, cancelled, missing-contact and ambiguous records, and list each exclusion with its reason. Then produce:
   1. A proposed workflow for selecting eligible invoices, checking them, obtaining approval and sending reminders.
   2. A reminder message for each eligible invoice, using only confirmed fields from the table.
   3. A plain-text approval list showing the invoice number, customer, recipient, amount, due date and proposed message.
   4. Exception rules for bounced emails, replies, partial payments, disputes, missing data and duplicate sends.
   5. A short test plan using sample records before any real message is sent.
   Do not send anything, claim that a message was sent, or give legal advice. Mark every item that a human must approve or verify.

   Open it prefilled in ChatGPT or Claude, or copy it into Gemini, which takes no prefill link.