Getting paid on time shouldn't feel like a part-time job — but for most small business owners, chasing invoices is exactly that. Setting up automated invoice reminders AI can handle end-to-end is one of the highest-ROI automations you can build in an afternoon. We're talking about a system that watches your invoices, knows when they're overdue, writes a personalised follow-up message, and sends it without you lifting a finger. This guide walks you through exactly how to build that system — with real settings, real examples, and zero fluff.
Why Manual Invoice Follow-Up Is Costing You More Than Time
Before we get into the setup, let's be honest about what's actually happening when you chase invoices manually.
You spot an overdue invoice. You open your email. You stare at a blank screen wondering how to be firm without being rude. You write something, second-guess it, soften it too much, and finally send a message that's so apologetic it barely registers. The client ignores it. You wait another week.
The problem isn't just the time — it's the inconsistency. Manual follow-up is emotionally draining, so you put it off. You're less firm with your biggest clients because you don't want to damage the relationship. You forget to follow up on smaller invoices because they don't feel worth the hassle.
An automated invoice reminders AI system solves all of this. It's consistent, it's timely, and it doesn't get anxious before hitting send. A well-built reminder sequence can reduce your average payment time by 30–50% without a single awkward conversation.
What You'll Need Before You Start
This tutorial uses Make.com as the automation platform and OpenAI (via API) for AI-generated message personalisation. You can adapt this for Zapier if that's what you're already using — the logic is identical, the interface just differs.
Here's what to have ready:
- A Make.com account (free tier works for testing; Starter plan for production use)
- An invoicing tool with API or integration support — QuickBooks, FreshBooks, Xero, or Wave all work
- An OpenAI API key (GPT-4o or GPT-4o mini is fine for this — cost per reminder is fractions of a cent)
- An email account connected to Make (Gmail, Outlook, or any SMTP setup)
- Your standard payment terms clearly defined (Net 7, Net 14, Net 30, etc.)
If you're on Zapier, the same structure applies — swap Make's "Modules" for Zapier's "Actions."
Step 1 — Create the Invoice Trigger in Make.com
Open Make.com and create a new Scenario. Your first module is the trigger — the thing that watches for overdue invoices.
If you're using QuickBooks:
- Search for "QuickBooks Online" in the module library
- Select Watch Invoices as your trigger
- Authenticate your QuickBooks account
- Set the filter: Status = Overdue (or "Past Due" depending on your QB version)
- Set the polling interval to every 1 hour (daily is fine if you're not in a cash crunch)
If you're using Xero:
- Use the Watch Invoices trigger in the Xero module
- Set filter: Invoice Status = AUTHORISED + Due Date < Today
If your invoicing tool isn't in Make's library: Use a webhook trigger and set your invoicing software to push data to Make when an invoice becomes overdue. Most modern invoicing tools support outgoing webhooks in their settings under "Integrations" or "Notifications."
At this point, your trigger is live. Every time it fires, it will pass along the invoice data — client name, amount, due date, days overdue, and any other fields you've mapped.
Step 2 — Filter by Days Overdue (Build a Sequence, Not a One-Shot)
The most effective payment reminder systems don't send one message — they send a sequence. A common structure:
- Day 1 overdue: Gentle nudge — "Just checking in" tone
- Day 7 overdue: More direct — reference the amount and terms
- Day 14 overdue: Firm — outline next steps if payment isn't received
- Day 30 overdue: Final notice — mention collections or late fees if applicable
To build this in Make, add a Filter module (the little diamond icon) after your trigger.
For the first reminder, set the condition: Days Overdue >= 1 AND Days Overdue < 7
Duplicate the entire sequence path and change the filter for each stage. Make's visual interface lets you branch these routes cleanly — you'll end up with four parallel paths feeding into your AI message generator.
Step 3 — Set Up the AI Message Generator
This is where the magic happens. Instead of sending a generic template, you're going to use AI to write a contextually aware, personalised message every time.
Add an OpenAI (GPT) module to your scenario and select Create a Completion (or Create a Chat Completion in newer module versions).
Model: GPT-4o mini (cheaper, fast, more than capable for this task)
System prompt: ` You are a professional accounts receivable assistant for a small business. Your job is to write payment reminder emails that are firm but polite, match the appropriate urgency level, and maintain a good client relationship. Never be aggressive. Always be clear about the amount owed and the original due date. `
User prompt (dynamic — built from your invoice data): ` Write a payment reminder email for the following invoice:
Client name: {{ClientName}} Invoice number: {{InvoiceNumber}} Amount due: {{Amount}} Original due date: {{DueDate}} Days overdue: {{DaysOverdue}} Reminder stage: {{ReminderStage}} (e.g. "first gentle reminder", "second reminder", "firm final notice")
The email should be 3–4 short paragraphs. Include a clear subject line at the top, starting with "SUBJECT:" on its own line. Sign off as [Your Business Name]. `
The {{variables}} pull directly from your invoice trigger data. Make.com lets you map these by clicking into the field and selecting the relevant data point from the trigger output.
Run a test. You'll see GPT generate a complete, contextually appropriate email — different tone for a first reminder versus a final notice, with real invoice details woven in naturally.
Step 4 — Parse the AI Output and Send the Email
The AI module returns the full text as a single string. You need to split the subject line from the email body before you can send it.
Add a Text Parser module (use the "Match Pattern" function) with this regex: ` SUBJECT: (.+)\n([\s\S]+) `
This captures:
- Group 1: The subject line
- Group 2: The full email body
Now add your Gmail (or Outlook, or SMTP) module:
- To: Map the client email address from the invoice trigger
- Subject: Map to Group 1 from your text parser
- Body: Map to Group 2 — set to HTML if you want basic formatting, or plain text works fine
- From name: Your business name
- Reply-to: Your actual email address (so replies come back to you, not a no-reply)
Before going live, turn on Make's Error Handler on this module. If an email fails to send (bad address, server issue), route it to a Google Sheet or Slack notification so you can follow up manually. Don't let failed sends disappear silently.
Step 5 — Log Every Reminder in a Tracking Sheet
Automated doesn't mean invisible. You need to know what's been sent, when, and to whom.
Add a Google Sheets module at the end of each path — select Add a Row and log:
| Column | Data |
|---|---|
| Timestamp | Make's {{now}} variable |
| Client Name | From invoice trigger |
| Invoice Number | From invoice trigger |
| Amount Due | From invoice trigger |
| Days Overdue | Calculated field |
| Reminder Stage | Your stage label |
| Email Sent | Yes/No |
This becomes your accounts receivable dashboard. At a glance, you can see every outstanding invoice, where it is in the sequence, and whether anything has gone sideways.
Step 6 — Add a Stop Condition
A critical step most tutorials skip: you need to stop sending reminders once an invoice is paid.
Add a second trigger or scheduled check that monitors for paid invoices. In QuickBooks, this fires when status changes to PAID. In Xero, look for VOIDED or PAID status changes.
When this triggers, use a Google Sheets Search module to find the row for that invoice and mark it "PAID" — then add a filter at the start of your reminder paths that skips any invoice number already marked as paid in your sheet.
Alternatively, some invoicing tools let you pass a "paid" flag directly in the webhook payload, which makes this even simpler.
This is not optional. Nothing burns client relationships faster than getting a "your invoice is overdue" message when they paid three days ago.
Real-World Example: What a Day 7 Reminder Looks Like
Here's an actual GPT output from this system (client name changed):
SUBJECT: Invoice #1042 — Payment Due ($2,400)
>
Hi Sarah,
>
I hope you're doing well. I'm following up on Invoice #1042 for $2,400, which was due on 3rd January. It looks like this one may have slipped through the cracks — these things happen.
>
Could you take a moment to process this or let me know if there's anything you need from my end to get it sorted? Payment details are included on the original invoice.
>
Thanks so much — I appreciate working with you and want to make sure everything stays on track.
>
Warm regards,
[Business Name]
Professional. Specific. Not robotic. Not aggressive. This is the kind of message that gets results without damaging the relationship — and your system is generating it automatically, every time, without you writing a word.
Frequently Asked Questions
Q: Can I set up automated invoice reminders AI without knowing how to code? A: Absolutely. Make.com is entirely visual — you connect modules by dragging and dropping. The only "technical" step is writing the OpenAI prompt, which is plain English. If you can write an email, you can build this automation.
Q: What if my invoicing software isn't supported by Make.com? A: Check Make's integration library first — it supports over 1,500 apps including most major invoicing tools. If yours isn't listed, use the generic Webhook module to receive data from any system that supports outgoing webhooks, which most modern software does.
Q: Will AI-generated reminders sound generic or obviously automated? A: Not if you build the prompt correctly. When your prompt includes the client's name, invoice number, exact amount, and days overdue — and you specify the tone for each reminder stage — the output reads like a human wrote it. Most clients will never know it was AI-generated.
Q: How much does this cost to run? A: Very little. Make.com's Starter plan covers most small businesses at around $9/month. OpenAI API costs for invoice reminders are typically under $1/month even if you're sending dozens of reminders. It's one of the cheapest high-value automations you can run.
Q: What happens if the AI writes something inappropriate or wrong? A: Two safeguards: first, a well-crafted system prompt keeps the AI on-brand and professional. Second, you can add a review step — route new reminders to a Slack or email notification for a quick human approval before they send, at least for the first few weeks until you trust the output.
You've Just Automated One of the Most Painful Parts of Running a Business
Let's be clear about what you've built here: a system that monitors every invoice you issue, knows exactly where each one is in the payment cycle, writes a contextually appropriate follow-up message, sends it at the right time, and logs everything so nothing falls through the cracks. All without you touching it.
The automated invoice reminders AI workflow in this guide typically takes 2–3 hours to set up the first time. After that, it runs in the background indefinitely, consistently, without getting uncomfortable before hitting send.
The next logical step is either connecting this to a broader client communication system — or comparing tools to make sure you're on the right platform for your business.
Ready to go deeper? Check out our full comparison of Make.com vs Zapier to see which automation platform is the better fit for your business size and budget.
Recommended Tool
Looking for a great tool to help with this? Try Getgenie — AI SEO content writer.
Want the Full AI Playbook?
If you're serious about building a lean, AI-powered business, grab the free guide that thousands of creators are using to do exactly that.
👉 Download "The Lean AI-Powered Business Playbook for Creators" — Free