12 Real n8n Automation Examples for Business (2026)

n8n automation examples you can build

If you want concrete n8n automation examples, here are 12 real workflows businesses run every day, grouped by function: sales, marketing, operations, support, finance, and data. Each one lists the trigger that starts it, what it does, and the tools it connects. Skip to the function you care about, then jump to the section at the end on how to decide which to build first.

I build workflows like these for clients, and the pattern that separates a useful automation from a fragile one is almost always the same: a clear trigger, a single job done well, and real error handling. The examples below are the ones that earn their keep.

A quick note on why n8n specifically. It's source-available under a fair-code license (not OSI open source, but free to self-host for internal business use), and it connects to roughly anything with an API. The project crossed 123,000 GitHub stars in 2025 and raised a $180M Series C in October 2025 that valued it at $2.5 billion, so the ecosystem and node library are deep and getting deeper (n8n's funding announcement). If you're still deciding between platforms, I wrote a separate breakdown of n8n vs Zapier for business.

Sales automations

Sales is where automation pays back fastest, because every hour a rep spends on admin is an hour not spent talking to a buyer.

1. Lead capture, enrichment, and routing

Trigger: A new form submission (Typeform, your website form, or a landing page).

What it does: The lead comes in, gets enriched with company size, job title, and LinkedIn URL, then an AI node scores how qualified it is. High-fit leads get pushed straight into the CRM and a Slack message lands in the rep's DMs with the score and a link to the record, so they see a qualified lead before they even open the CRM. Low-fit leads go to a nurture list instead of wasting a rep's time.

Tools it connects: Webhook or Typeform, Clearbit or Hunter.io for enrichment, an OpenAI node for scoring, HubSpot or Pipedrive, and Slack.

The whole thing runs in under a minute per lead. A basic version takes a few hours to build; the full system with scoring and routing is a day or two. What happens to the leads sent to nurture is its own build, and I wrote up my lead nurture system in n8n, with source-aware welcome emails and a drip that stops when a lead books a call.

2. Two-way CRM sync

Trigger: A record changes in either system (a deal stage updates in your CRM, or a row changes in the spreadsheet your ops team actually uses).

What it does: Keeps two systems in agreement without anyone copy-pasting. It deduplicates on email so you don't end up with three records for the same person, and it writes changes both directions on a schedule.

Tools it connects: HubSpot or Salesforce, Google Sheets or Airtable, with a dedup step in the middle.

This is the unglamorous one nobody demos, and it's also the one that saves the most quiet frustration.

3. Meeting-to-CRM follow-up

Trigger: A calendar event ends, or a call recording finishes processing.

What it does: Pulls the transcript, has an AI node summarize what was discussed and the next steps, logs it as a note on the deal, and drafts a follow-up email for the rep to review and send.

Tools it connects: Google Calendar, a transcription service, OpenAI, and your CRM.

The email half of that follow-up, including the nodes, deliverability and a lead-form flow, is covered in my n8n email automation guide.

Marketing automations

Marketing automations are mostly about distribution and consistency: doing the repetitive publishing and reporting work the same way every time.

4. Content repurposing pipeline

Trigger: A new YouTube video is published, or a new row is added to a content queue.

What it does: Takes one piece of long-form content and turns it into several short ones. It transcribes the video, generates a LinkedIn post, an X thread, and a short newsletter blurb, then drops them into a review folder or schedules them directly. You approve, it ships.

Tools it connects: YouTube or RSS, OpenAI, and your scheduler (Buffer, LinkedIn, or a Google Doc for review).

5. Weekly marketing report

Trigger: A schedule (every Monday at 8am).

What it does: Pulls last week's numbers from your ad platforms and analytics, assembles them into a clean summary with week-over-week changes, and posts it to a Slack channel or emails it to the team. No one logs into five dashboards to build the same slide every week.

Tools it connects: Google Analytics, Meta or Google Ads, and Slack or Gmail.

6. Review and mention monitoring

Trigger: A new review appears, or a brand mention shows up in a monitored feed.

What it does: Catches new reviews and mentions, runs sentiment on them, and routes anything negative to the right person immediately while logging the rest. Nobody finds out about an angry one-star review three weeks late.

Tools it connects: Google Business Profile or a review API, an OpenAI node for sentiment, and Slack.

Operations automations

Operations is the function where automation quietly compounds. Delivery Hero automated its account recovery process with n8n and reported saving more than 200 hours a month (n8n). Vodafone put 33 workflows into daily operations and estimated close to 5,000 person-days saved a year (n8n).

7. Client or employee onboarding

Trigger: A deal is marked "won," or a new hire is added to your HR system.

What it does: Fires off the whole onboarding sequence the moment it's needed. It creates the accounts, sets up the project folders, assigns the right team members, sends the welcome email, and schedules the kickoff call. Every client gets the same professional experience instead of whatever the account manager remembered to do that day.

Tools it connects: Your CRM or HR tool, Google Workspace or Slack, a project tool like Asana or ClickUp, and a calendar.

8. Document and approval routing

Trigger: A file lands in a watched folder, or a request is submitted through a form.

What it does: Reads the document, pulls out the key fields, routes it to the right approver, and chases the approval if it sits too long. The thing that used to live in someone's inbox now moves on its own.

Tools it connects: Google Drive or Dropbox, an OCR or AI extraction node, and Slack or email for approvals.

Customer support automations

Support automation works best when it handles triage and the boring first response, and hands the human a head start rather than replacing them.

9. Ticket triage and routing

Trigger: A new support ticket or inbound email arrives.

What it does: Reads the ticket, classifies the intent, urgency, and sentiment, tags it, routes it to the right queue, and sends an instant acknowledgment so the customer knows they're heard. Urgent issues and VIP customers get flagged so they don't sit in a general queue. Teams running this kind of multi-channel triage have reported response times dropping by up to 80% (n8n).

Tools it connects: Gmail, Zendesk, or a help-desk inbox, an OpenAI node for classification, and Jira or your ticketing system.

If you want to see how far this can go, I put together a deeper look at how I approach support automation specifically on the AI automation consultant page.

10. AI draft replies from your knowledge base

Trigger: A ticket is tagged as a common question type.

What it does: Searches your help docs and past answers, drafts a reply grounded in that content, and leaves it for an agent to approve or edit. The agent sends a good answer in seconds instead of writing it from scratch. Crucially, a human still hits send, so you don't ship a confidently wrong answer to a customer.

Tools it connects: Your help desk, a vector store or your docs, and an OpenAI node.

Finance automations

Finance teams lose a lot of hours to manual data entry and chasing. One pattern I see often: invoice processing dropping from around 10 hours a week to under 2 once the matching and reminders are automated.

11. Invoice and overdue-payment reminders

Trigger: A schedule (every morning at 9am).

What it does: Pulls open invoices, filters the ones that are overdue, drafts a polite, personalized reminder for each, and sends it. It logs what went out so nobody double-chases a client who already paid.

Tools it connects: QuickBooks, Xero, or Stripe, an OpenAI node for the message, and Gmail or your email tool.

12. Expense and receipt capture

Trigger: A receipt is emailed in or dropped in a folder.

What it does: Reads the receipt, extracts the vendor, amount, and date, categorizes the expense, and writes a clean row into your accounting system or a spreadsheet. The shoebox of receipts becomes a structured ledger.

Tools it connects: Gmail or Drive, an OCR or AI extraction node, and Google Sheets, QuickBooks, or Xero.

Bonus: data automations

Most of the examples above quietly depend on a data layer. A couple of workflows worth running on their own:

• Scheduled ETL sync. On a schedule, pull from an API or database, clean and reshape the data, and load it into your warehouse or a Google Sheet that feeds your dashboards. This is the plumbing behind every "live" report.
• Daily data-quality check. Each morning, scan a key table for missing fields, duplicates, or values that look wrong, and post a short alert if something's off. You hear about bad data before your CEO does.

If you want to see live versions of workflows like these running against real data, I keep a set of interactive demos on the site.

How to decide which n8n automation to build first

Twelve examples is a lot. You don't build twelve. You build one, prove it works, then add the next. Here's the order I use with clients.

Start with frequency times pain. List your candidate automations and score each on two things: how often the task happens, and how much it hurts when it's done late or wrong. A task that happens fifty times a day beats one that happens twice a month, even if the monthly one feels more impressive. Lead routing and ticket triage usually win on this math.

Then check the data. The best first automation is one where the trigger and the tools both have clean, reliable APIs. If the system you need to touch has a flaky integration or lives in a PDF nobody can parse, that's a harder first project. Pick something with solid plumbing so your first win is fast.

Then weigh the blast radius. Early on, favor workflows where a mistake is cheap and visible. An internal Slack report that's wrong is a shrug. An automated payment that's wrong is a phone call you don't want. Build confidence on low-risk internal workflows before you automate anything that touches money or customers without a human in the loop.

Put those three together and the first build picks itself: high frequency, clean data, low blast radius. For most businesses that's lead capture, ticket triage, or a recurring report. Ship that one, watch it for a week, then move down the list.

One last thing. The examples here are deliberately practical, but the gap between "this works in a demo" and "this runs unattended for a year" is real. That gap is deduplication, retries, and someone getting paged when a workflow fails quietly. If you'd rather not learn that the hard way, that's exactly the kind of work I do as an n8n consultant, and you can book a 30-minute call to talk through which automation would move the needle for your team.