Some links in this article are affiliate links. Automated Wealth Lab may earn a commission at no extra cost to you. See our Affiliate Disclosure.
Disclosure: This article contains affiliate links to n8n and Kit. If you sign up through them, Automated Wealth Lab earns a commission at no extra cost to you. Stripe is not an affiliate partner — its fees are quoted because the workflow is built on it, and we’d quote them the same way if they were.
Late invoices are the tax every freelancer pays for being polite. You did the work, you sent the invoice, and then the real job starts: remembering who owes what, deciding when a “just checking in” email becomes a “per our terms” email, and doing that dance thirty times a month without souring a client relationship you still need.
This is a blueprint for handing that entire chase to a workflow — and for knowing exactly what the automation costs per invoice, so you can compare it against the cost of your own awkward follow-up time. It extends the workflow patterns in our n8n workflow blueprints collection into the least glamorous, most cash-relevant automation a solo business can run.
The workflow, node by node
The trigger: Stripe, not your memory. Stripe already knows everything the chase needs — when an invoice was created, when it’s due, when it’s paid, when a payment failed. The workflow starts from Stripe events (invoice.created, invoice.finalized, invoice.payment_failed, invoice.paid) delivered by webhook to n8n. No polling, no spreadsheet of doom. If you invoice through Stripe Invoicing, note its own fee: 0.4% per paid invoice, capped at $2, on top of standard card processing (2.9% + 30¢) — you pay it whether or not you automate, but it belongs in the cost table below.
The state store: one row per invoice. Each open invoice gets a record — a row in a Google Sheet, an Airtable base, or n8n’s own data tables — holding: client, amount, due date, touches sent, and a status flag. The workflow updates the row on every event. invoice.paid flips the flag and the chase stops mid-ladder, automatically. This is the step manual chasing always gets wrong: the reminder that goes out the morning after the client paid.
The ladder: five touches, escalating in firmness, never in rudeness.
| Touch | Timing | Tone | Sent via |
|---|---|---|---|
| 1 | 3 days before due | Friendly heads-up, payment link | |
| 2 | On due date | “Due today” + link | |
| 3 | +3 days overdue | Polite nudge, restate terms | |
| 4 | +7 days overdue | Firmer: late fee per your terms, offer to talk | |
| 5 | +14 days overdue | Final automated touch → flags YOU | Email + your task list |
Two design rules matter more than the copy. First, every touch carries the payment link — a reminder that makes the client go find the original invoice is a reminder that fails. Second, touch 5 is a handoff, not an email: after 14 days the problem is no longer forgetfulness, and the workflow’s job becomes putting it on your desk with full context (amount, days overdue, touches sent, last reply) instead of sending a sixth, harder email by itself. Automation chases forgetfulness. Humans chase problems.
The failed-payment branch. Card failures are a different species from late payers — usually an expired card, not a reluctant client. Stripe’s invoice.payment_failed event routes to a separate, shorter ladder: immediate “your card didn’t go through, here’s an update link” email, one retry nudge at +2 days, handoff at +5. Mixing failed-payment clients into the late-payer ladder is how you accidentally send a dunning tone to your best customer over an expired Visa.
Where Kit fits. If your “clients” are really customers — you sell a course, a template pack, a paid community, and invoices/payment plans run through Stripe — the reminder emails are marketing-adjacent, and sending them through Kit buys you things raw SMTP doesn’t: proper unsubscribe handling, sequences you can edit without touching the workflow, and tagging, so a customer who just got a payment-plan reminder doesn’t also get your launch sequence the same week. Kit’s free Newsletter plan covers up to 10,000 subscribers with one basic automation; the Creator plan ($39/month monthly billing at 1,000 subscribers; $33 on annual) unlocks unlimited visual automations. For B2B client invoicing — five agencies who owe you retainers — skip Kit; plain email from your own domain via the workflow is the right tool, and we’ll cost it that way. One workflow, two sending paths, chosen by invoice type.
What it costs to run
Assumptions, stated: you send 30 invoices a month, average invoice $750, clients pay by card through Stripe links. These are the model’s numbers, not industry statistics — swap yours in.
Executions. Per invoice lifecycle: 1 created event + up to 5 scheduled touches + 1 paid event + a couple of status checks ≈ 8–10 executions. Thirty invoices ≈ 270 executions/month, plus failed-payment branches. Call it 300.
| Cost line | Per month | Per invoice | Notes |
|---|---|---|---|
| n8n Cloud Starter (€20 annual / €24 monthly, 2,500 executions) | ~€20–24 | ~€0.75 | 300 executions uses 12% of the tier |
| — or self-hosted n8n | ~$5–10 VPS share | ~$0.25 | If you already self-host, marginal cost ≈ $0 |
| Stripe processing (2.9% + 30¢ on $750) | — | $22.05 | Unchanged by automation — you pay this regardless |
| Stripe Invoicing (0.4%, capped $2) | — | $2.00 | Only if you invoice through Stripe Invoicing |
| Kit (if used for customer invoices) | $0 free tier / $33–39 Creator | $0–1.30 | Only the customer-invoice path needs it |
| Email sending (own domain SMTP) | ~$0 | ~$0 | Reminder volume is trivially small |
| Automation-specific total | ~€20–24 | ~€0.75–2 | Excluding Stripe fees you’d pay anyway |
The honest comparison is not automation cost vs. zero. It’s automation cost vs. your chasing time. Planning assumption, labeled as such: manual chasing runs about 6 minutes per touch when you count finding the invoice, checking whether it was paid, and writing a message that isn’t copy-paste rude. Ten overdue invoices × 3 manual touches × 6 minutes ≈ 3 hours a month. At any billing rate above €10/hour, the €20–24 n8n tier is paid for; at a freelancer’s $75–150/hour, the workflow saves the equivalent of $225–450 of unbilled, relationship-straining time for the price of a coffee subscription. If you send five invoices a month and everyone pays on time, the manual cost is near zero and so is the case for automating — the blueprint scales with invoice count and late rate, and we’ll say plainly in the verdict who shouldn’t build it.
Set the workflow up on n8n Cloud if you don’t already run a server — 2,500 executions is roughly eight times this blueprint’s appetite, leaving room for the rest of your automations on the same tier. (If you’re deciding where n8n should live long-term, our n8n hosting comparison prices the options.)
The guardrails that make it safe to leave running
The paid-check before every send. The workflow re-reads the invoice status from Stripe immediately before each touch — not the status in your sheet, the live one. Webhooks can be missed; a stale row is how you chase a paid invoice. Belt and braces: the sheet is the schedule, Stripe is the truth.
The kill switch. One tag/column — “pause chase” — that you can set from your phone when a client emails “accounting is processing it Friday.” The workflow checks it before every touch. Any automation that a client can make you regret needs an off switch reachable in ten seconds.
Weekend and timezone manners. Touches send on business days, in the client’s timezone, mid-morning. A reminder landing Sunday at 11 p.m. reads as a collections agency, not a professional. n8n’s scheduling makes this a settings choice, not a discipline problem.
Never automate the dispute. If a client replies contesting the invoice, the ladder must stop — a human conversation has started. The workflow can’t read replies to reminder emails reliably, which is exactly why touch 5 hands off at +14 days and why the “pause chase” column exists. State this limit to yourself before you build, not after the first awkward thread.
Late fees: check your contract first. Touch 4 mentions a late fee “per your terms” — only if your signed terms actually include one. Automating a fee you never contracted for is a fast way to convert a late payer into an ex-client and a chargeback. (And on the subject of what’s actually enforceable, the same caution as our n8n fair-code license explainer: the paperwork you agreed to is the paperwork that counts.)
FAQ
How much does the invoice-chasing workflow cost to run?
On n8n Cloud Starter (~€20–24/month), about €0.75 per invoice at 30 invoices a month — 300 executions against a 2,500-execution tier. Self-hosted, the marginal cost is effectively the VPS share, a few cents per invoice. Stripe’s own fees (2.9% + 30¢, plus 0.4% capped at $2 if you use Stripe Invoicing) are unchanged by the automation.
Will automated reminders annoy my clients?
The first three touches are the emails a good bookkeeper sends anyway — heads-up, due-today, gentle nudge — on a predictable schedule, always with the payment link, never on weekends. What annoys clients is the manual failure modes this removes: the reminder after they’ve paid, and the suddenly-firm email from a freelancer who’s been silently stewing for three weeks.
What if a client pays mid-ladder?
The invoice.paid webhook flips the record’s status and the ladder stops; additionally, the workflow re-checks live status in Stripe before every scheduled touch, so even a missed webhook doesn’t produce a chase-after-payment email.
Do I need Kit for this, or is it optional?
Optional, and only for one path. If your invoices go to customers of digital products or payment plans — people on your email list — Kit’s sequences and tagging keep reminder emails coordinated with your marketing. For B2B retainer clients, send from your own domain via SMTP and skip the extra subscription.
Can the workflow handle failed card payments too?
Yes, as a separate branch: Stripe’s invoice.payment_failed event starts a shorter, friendlier ladder (update-your-card link immediately, one nudge, then a handoff to you). Never route failed payments through the late-payer wording — an expired card is not a delinquent client.
When is this NOT worth building?
Under roughly ten invoices a month with reliable payers, manual chasing costs you minutes, and the honest answer is a calendar reminder. The blueprint pays off when invoice volume, late payers, or your hourly rate make three hours of monthly chasing the most expensive admin you do.
Methodology
What we did: Designed the workflow from Stripe’s documented event model (invoice.* webhooks) and n8n’s scheduling/error-handling primitives, then costed it from October 2026 published prices: n8n Cloud tiers, Stripe’s standard processing (2.9% + 30¢) and Stripe Invoicing (0.4% per paid invoice, capped at $2) fees, and Kit’s published plan prices. Per-invoice figures are arithmetic on a stated model (30 invoices/month, $750 average, 8–10 executions per invoice lifecycle).
What we did NOT do: We did not deploy this exact workflow against a live Stripe account for this article, and we make no claim about payment-speed improvements — no “get paid X days faster” statistic appears above because we have no verified basis for one. The 6-minutes-per-manual-touch figure is a labeled planning assumption for the time comparison, not a measured average; your mileage will vary, and the break-even logic is shown so you can substitute your own. Recovery rates for automated reminders vary enormously by industry and client relationship, which is why the model prices the cost side only.
Conflict disclosure: n8n and Kit are affiliate partners; Stripe is not. The blueprint’s recommendation of Stripe as the trigger source follows from its event model, and the cost table includes Stripe’s fees in full — including the Invoicing add-on fee that an affiliate-driven article would have every incentive to hide.
AWL verdict
This is the rare automation whose ROI doesn’t need a case study: it replaces a recurring, soul-draining chore with a €20–24/month tier you’re probably already paying for, at roughly €0.75 of automation cost per invoice. Build it when you pass ~15 invoices a month or the first time you catch yourself drafting a third follow-up at midnight. Keep the guardrails — live paid-check before every send, a pause switch, a human handoff at +14 days — and let n8n be the one who never forgets, never simmers, and never sends the email you’ll regret.
Sources
- Stripe standard pricing — 2.9% + 30¢ domestic cards; add-on fee stack: https://www.activepieces.com/blog/stripe-pricing-2026-total-transaction-add-on-costs (verified October 2026)
- Stripe Invoicing fee — 0.4% per paid invoice, capped at $2; Billing 0.7%: https://checkoutpage.com/blog/stripe-processing-fees (verified October 2026)
- Stripe fee worked examples by ticket size: https://tools.dailyutilitylab.com/what-percentage-does-stripe-take.html (verified October 2026)
- n8n Cloud pricing — Starter €20/mo annual (€24 monthly), 2,500 executions: https://www.stackscout.net/articles/cc_20261014_180524.html (verified October 2026)
- Kit pricing — free Newsletter plan to 10,000 subscribers; Creator $39/mo monthly ($33 annual) at 1,000 subscribers: https://www.emailtooltester.com/en/reviews/convertkit/pricing/ (verified October 2026)
