A consultant drafts late invoice follow-ups with AI and sends them

The task

A client was well past the payment date on a large invoice, and the author had not chased it firmly. The only follow-up was one vague email asking whether the invoice had arrived. The author says chasing money feels like begging, so there was no routine for late invoices, and drafting each message from scratch meant softening every sentence.

What they did

The author built three email templates with ChatGPT, saved them, and reuses them for every late invoice. The emails go out at fixed points after the due date and differ in tone: first friendly and assuming an oversight, then neutral and factual with the amount, the due date, and a request for a specific payment date, then direct with a short deadline. Each draft comes from a reusable prompt, which the post shows in full for the third email, that takes the client name, invoice number, amount, and due date, and asks for a short, calm, professional text with no apology for chasing. The author notes that AI tends to default to soft, apologetic wording, so the prompt rules that out. To find the slowest payers, the author also pasted a year of invoice and payment dates into a spreadsheet and asked ChatGPT to rank the clients. A phone call is the final step, and the author considers it a step AI should not take.

What the source reports

According to the author, filling in a saved template takes about four minutes per client, compared with the twenty minutes spent drafting each message from scratch before. According to the author, the average delay past payment terms across 34 invoices was 19 days. The author reports that the average delay then fell from 19 days to 6 after the invoice template gained a specific due date and a stated late fee, and that about a third as many chasing emails go out as a year earlier.

Where the approval sits

The AI only drafts. The author reads each draft, adds the client's name and history, and sends it from the author's own email address rather than from the accounting software. In the post's FAQ, the author says that letting AI chase invoices with no human sending is not yet reliable without connecting it to email and accounting software, a setup it says most small businesses lack, and that a person sending keeps the message sounding like it came from a person. The late-stage phone call is also made by the author.

Worth knowing

This is one consultant's personal account, and the page also promotes the author's AI coaching, so the figures are the author's own. The author credits the drop in average delay mainly to clearer invoice terms and calls the follow-up emails a bandage, so the figures should not be read as the effect of AI drafting alone. The author's accounting software also sends an automatic reminder on the due date, which this entry does not cover. The post gives the timing of the steps in two slightly different ways, and it is written for the UK; this entry leaves out its statements about UK late payment rules.

Prompt for your agent

More in this category

• Invoices and payment reminders

  easybill lets an AI assistant draft invoices and wait for approval

  easybill's help article describes a connector that lets an AI assistant create invoices, quotes, and credit notes in an easybill account. Each document starts as a draft, and finalizing, sending, and recording payments wait for the user's confirmation.

  Read the entry
• Invoices and payment reminders

  SimplyBills lets a connected AI draft reminders for overdue invoices

  SimplyBills describes how a freelancer connects their own ChatGPT or Claude to the invoicing tool so the assistant finds overdue invoices and drafts the reminder. Nothing is sent until the user confirms, and sending through the assistant is limited to the Max plan.

  Read the entry