Quick Summary
n8n is an open-source workflow automation platform that connects your apps, systems, and data to automate repetitive business processes. Unlike Zapier or Make, n8n can be self-hosted (keeping your data in your environment), has no execution charges when self-hosted (you pay for the server), and supports custom JavaScript/Python for complex logic. This guide covers what n8n is, how it compares to alternatives, 5 use cases for Australian businesses, and why self-hosted is the smarter choice.
Key point: Per-execution pricing on Zapier and Make can make high-volume automation expensive at scale. Self-hosted n8n removes that cost, which is why many Australian businesses are looking at it as an alternative.
Table of Contents
- What Is n8n?
- n8n vs Zapier vs Make
- Why Choose n8n
- 5 Use Cases for Australian Businesses
- Self-Hosted Benefits (No Per-Execution Fees)
- Getting Started
- Pricing
- Frequently Asked Questions
What Is n8n?
n8n (pronounced "n-eight-n") is a workflow automation platform that connects your apps, APIs, and services to automate repetitive business processes. It uses a visual, node-based interface where you drag and drop building blocks to create workflows – without writing code (though you can add custom code when needed).
How n8n Works
| Step | What Happens | Example |
|---|---|---|
| 1. Trigger | An event starts the workflow | New invoice arrives via email |
| 2. Extract | n8n reads and processes the input | AI extracts invoice data from PDF attachment |
| 3. Transform | Data is formatted, validated, and enriched | Amounts converted, GST calculated, supplier verified |
| 4. Action | n8n takes action in your target system | Invoice created in Xero, approval notification sent |
| 5. Monitor | Workflow execution is logged and tracked | Dashboard shows success rate, processing time, errors |
Key Features
| Feature | What It Does |
|---|---|
| Visual workflow builder | Drag-and-drop nodes to create automations without coding |
| 400+ integrations | Connect to Microsoft 365, Xero, Salesforce, HubSpot, Slack, Google Sheets, and hundreds more |
| Custom code nodes | Add JavaScript or Python for complex logic that pre-built nodes cannot handle |
| AI/LLM integration | Built-in nodes for OpenAI, Anthropic, and other LLM APIs for intelligent document processing |
| Error handling | Automatic retry, error routing, and notification when workflows fail |
| Self-hosted option | Run n8n on your own infrastructure for data sovereignty and cost control |
| API access | Trigger workflows via API, expose workflow outputs as API endpoints |
n8n vs Zapier vs Make
The three most popular workflow automation platforms are n8n, Zapier, and Make (formerly Integromat). Here is how they compare:
Feature Comparison
| Feature | n8n | Zapier | Make |
|---|---|---|---|
| Pricing model | Community Edition free to self-host; cloud from €20/month billed annually for 2,500 executions (n8n) | Free plan with 100 tasks/month; paid plans from US$19.99/month, priced by task volume (Zapier) | Free and paid plans, priced by operation volume |
| Self-hosted option | Yes (full-featured, no restrictions) | No | No |
| Usage-based pricing | None when self-hosted; cloud plans priced by full workflow executions, not by step | Priced by tasks (each action step) | Priced by operations |
| Custom code | JavaScript and Python | JavaScript (Code step only) | Limited (no custom code) |
| AI/LLM nodes | Built-in (OpenAI, Anthropic, custom) | Limited (OpenAI only) | Limited (OpenAI only) |
| Integrations | 400+ | 7,000+ | 1,800+ |
| Complex workflow support | Unlimited nodes, branching, loops, parallel execution | Limited branching, no loops on lower tiers | Branching and loops, but complex workflows are expensive |
| Data sovereignty | Full control (self-hosted in Australia) | US-based servers | EU-based servers |
| Open source | Fair-code (source available, free for internal use) | No | No |
How the Pricing Models Differ at Volume
Zapier charges by tasks and Make by operations, so a multi-step workflow that runs thousands of times a month consumes a lot of plan allowance. n8n Cloud counts a whole workflow run as one execution regardless of how many steps it has (n8n), and the self-hosted Community Edition has no execution charge at all: you pay for the server and your time to run it. Price your own expected volume on each vendor's current pricing page before deciding, because plans and allowances change.
When to Use Each Platform
| Platform | Best For | Not Good For |
|---|---|---|
| n8n | Mid-market businesses with 1,000+ executions/month, data sovereignty requirements, complex logic needs, AI integration | Businesses that need 7,000+ app integrations out of the box |
| Zapier | Small businesses with simple workflows, need for maximum app integrations, non-technical users | High-volume automation (costs escalate rapidly) |
| Make | Visual workflow enthusiasts, moderate volume needs, EU data residency | Complex custom logic, self-hosted requirements, high volume (operation costs add up) |
Why Choose n8n
Compared with Zapier, Make and Power Automate, n8n stands out for four reasons:
Reason 1: No Per-Execution Fees
With Zapier or Make, usage counts against your plan, so higher volumes mean higher plans. With self-hosted n8n, the cost is the server and its upkeep, regardless of how many executions you run.
This makes automation economically viable at scale. If every execution costs money, you are disincentivised from automating high-volume processes (like email triage with 50,000 emails/month). With n8n, you automate everything without worrying about execution costs.
Reason 2: Data Sovereignty
Self-hosted n8n runs in your own infrastructure (Azure Australia East, AWS Sydney, or your own servers). Your data never leaves Australia. This is critical for:
- Government suppliers with data residency requirements
- Financial services under APRA CPS 234
- Healthcare providers handling patient data
- Any business that cannot send data to US or EU servers
Reason 3: Custom Code for Complex Logic
n8n supports JavaScript and Python code nodes, which means you can implement any logic that cannot be expressed with pre-built nodes. This is essential for:
- Complex data transformations and calculations
- Custom API integrations with proprietary systems
- AI-powered document processing with confidence scoring
- Error handling with sophisticated retry logic
Reason 4: AI/LLM Integration
n8n has built-in AI nodes for OpenAI, Anthropic, and other LLM APIs. This means you can build AI-powered automations (invoice extraction, email classification, report generation) without building custom AI infrastructure. The AI nodes are native to the platform – no separate API management, key rotation, or rate limiting to worry about.
5 Use Cases for Australian Businesses
Use Case 1: Invoice Processing
The problem: Staff receive 2,000+ supplier invoices per year via email. Each invoice requires manual data entry from PDF into Xero or MYOB. Average processing time: 15 minutes per invoice.
The n8n workflow:
Email Trigger (IMAP/Gmail) → Extract PDF Attachment → AI Document Extraction (OpenAI) →
Validate Data → Lookup Supplier in Xero → Create Bill in Xero → Send Approval Notification (Teams/Slack)
Illustrative example, assuming 2,000 invoices a year and the processing times shown:
| Metric | Before | After |
|---|---|---|
| Processing time per invoice | 15 minutes | 2 minutes (AI extracts, human reviews) |
| Staff time per year | About 500 hours | About 67 hours |
| Annual saving | About 430 hours, multiplied by your loaded hourly cost |
Use Case 2: Client Onboarding
The problem: New client onboarding involves sending welcome emails, requesting documents, verifying completeness, creating records in the CRM, and generating contracts. Average turnaround: 3 days.
The n8n workflow:
CRM New Contact Trigger → Generate Welcome Email → Create Document Upload Portal (custom web form) →
Monitor Uploads → AI Verify Completeness → Generate Contract (template merge) →
Create Records in Practice Management → Notify Account Manager
Illustrative example, assuming 200 new clients a year and the times shown:
| Metric | Before | After |
|---|---|---|
| Onboarding turnaround | 3 days | 4 hours |
| Staff time per onboarding | 2 hours | 20 minutes |
| Staff time per year | About 400 hours | About 67 hours |
| Annual saving | About 330 hours, multiplied by your loaded hourly cost |
Use Case 3: Monthly Report Generation
The problem: Finance team spends 2 days per month consolidating data from multiple systems into management reports.
The n8n workflow:
Schedule Trigger (last business day) → Pull Data from Xero (P&L, Balance Sheet) →
Pull Data from Practice Manager (revenue by practice area) →
Pull Bank Feed Data → AI Generate Narrative Commentary →
Populate Charts and Tables → Deliver Draft Report to CFO (email) →
CFO Reviews and Approves → Distribute to Management Team
Illustrative example, assuming 12 reports a year and the times shown:
| Metric | Before | After |
|---|---|---|
| Monthly close time | 5 days | 2 days |
| Staff hours per report | 16 hours | 3 hours |
| Staff time per year | 192 hours | 36 hours |
| Annual saving | 156 hours, multiplied by your loaded hourly cost |
Use Case 4: Security Incident Alerting
The problem: Security alerts from multiple sources (firewall, endpoint protection, email security) arrive in separate systems. Nobody correlates them or prioritises response.
The n8n workflow:
Security Alert Trigger (SIEM API) → Correlate with Other Alerts (cross-reference) →
AI Classify Threat Severity → If Critical: Page On-Call Engineer (PagerDuty) →
If Medium/High: Create Ticket in ITSM System →
If Low: Log and Monitor → Generate Daily Security Summary Report
| Metric | Before | After |
|---|---|---|
| Alert correlation time | Manual (hours to days) | Automated (seconds) |
| Mean time to detect critical threats | 4-8 hours | <1 minute |
| Mean time to respond | 2-4 hours | 15-30 minutes |
| Annual risk reduction | Significant – faster detection and response prevents data breaches |
Use Case 5: ESG Data Collection
The problem: Scope 1/2/3 emissions data lives in utility portals, travel booking systems, fleet management tools, and supplier invoices. Consolidating it for ESG reporting takes weeks.
The n8n workflow:
Monthly Schedule Trigger → Pull Electricity Data (utility API) →
Pull Travel Data (booking system API) → Pull Fleet Data (fleet management API) →
Pull Supplier Emissions Data (supplier portal/API) →
AI Calculate Scope 1/2/3 Emissions (using government emissions factors) →
Update ESG Dashboard → Generate Monthly Emissions Report →
Flag Anomalies (unusual spikes in consumption)
| Metric | Before | After |
|---|---|---|
| Data collection time | 2-3 days per month | 2 hours per month (automated) |
| Calculation accuracy | Manual, error-prone | Automated, verified |
| Ongoing effort | Repeated manual collection each reporting period | Review and exception handling only |
Self-Hosted Benefits (No Per-Execution Fees)
The single biggest advantage of n8n over Zapier and Make is the ability to self-host.
What Self-Hosted Means
| Aspect | Self-Hosted n8n | Cloud Platforms (Zapier, Make) |
|---|---|---|
| Where it runs | Your infrastructure (Azure, AWS, on-premise) | Vendor's cloud servers |
| Data location | Your choice (Australia) | Vendor's choice (US/EU) |
| Cost | Server and maintenance only | Platform fee that rises with usage |
| Execution limit | Limited only by your server | Capped by plan |
| Customisation | Full access to source code, custom nodes | Limited to vendor-provided features |
| Data sovereignty | Full control | Vendor dependent |
| Compliance | Can be configured to your compliance requirements (IRAP, Essential Eight) | Limited to vendor's certifications |
The Cost Math
Self-hosting swaps a usage-based subscription for a fixed server cost plus the time to patch, back up and monitor the instance. At low volumes a cloud plan is often simpler and cheap enough. As volumes grow, the self-hosted cost stays roughly flat while usage-based plans step up, so compare both at your expected volume using current vendor pricing.
Data Sovereignty for Australian Businesses
For businesses that must comply with Australian data residency requirements, self-hosted n8n is the only viable option:
| Requirement | Cloud Platforms | Self-Hosted n8n |
|---|---|---|
| Data stored in Australia | No (Zapier: US, Make: EU) | Yes (your choice of Azure Australia East or AWS Sydney) |
| APRA CPS 234 compliant | Difficult (vendor-dependent) | Yes (you control the infrastructure) |
| Essential Eight compliant | Difficult | Yes (you apply Essential Eight to your own servers) |
| IRAP assessed | No | Possible (you control the environment) |
Getting Started
Option 1: Try n8n Cloud (Free Trial)
If you want to experiment with n8n before committing to self-hosted:
- Sign up at n8n for a free trial
- Explore the workflow builder with pre-built templates
- Test a simple automation (e.g., email to Google Sheets)
- Evaluate whether n8n meets your needs
Option 2: Self-Host n8n (Recommended for Mid-Market)
For businesses with 50+ users and significant automation needs, self-hosting is usually the better fit. A production-ready Docker Compose stack (PostgreSQL, reverse proxy, backups) is available on GitHub at github.com/syncbricks/n8n.
- Provision a server (Azure Australia East or AWS Sydney, minimum 2 CPU, 4GB RAM)
- Install n8n via Docker (one command:
docker run -it --rm --name n8n -p 5678:5678 -v ~/.n8n:/home/node/.n8n n8nio/n8n), or use the ready-made self-hosted stack on GitHub - Configure integrations (API keys for Xero, Microsoft 365, OpenAI, etc.)
- Build your first workflow using the visual builder
- Monitor execution logs and refine
Option 3: Engage an n8n Consultant
For businesses that want expert implementation from day one, a typical engagement covers:
- Scoping call to identify your top 3-5 automation opportunities
- n8n installation and configuration on your infrastructure
- Workflow design, development, and testing
- Staff training on workflow monitoring and troubleshooting
- Ongoing maintenance and new workflow deployment
Pricing
n8n Plans and Pricing
Prices from n8n's pricing page as at October 2026, in euros, billed annually. Check the page for current prices and monthly billing.
| Plan | Price | Workflow executions | Best For |
|---|---|---|---|
| Starter | €20/month | 2,500 per month | Individuals and small teams testing automation |
| Pro | €50/month | 10,000 per month | Small teams running automations in production |
| Business | €667/month (self-hosted) | 40,000 per month | Companies with fewer than 100 employees needing collaboration and scale |
| Enterprise | Custom | Custom | Organisations with strict compliance and governance needs |
Self-Hosted n8n Cost
| Component | Cost |
|---|---|
| n8n Community Edition | Free to self-host |
| Server (Azure/AWS or your own) | Depends on size and provider; a small VM is enough to start |
| AI model API costs (OpenAI, etc.) | Usage-based; depends on how much AI work your workflows do |
| Your time | Patching, backups and monitoring, or a provider to do it |
Frequently Asked Questions
Is n8n truly free?
n8n is fair-code licensed, which means it is free to use for internal business purposes. You can self-host it, modify it, and use it without paying licence fees. You only pay if you want to resell n8n as a product to other companies (which is not relevant for most businesses).
How difficult is it to learn n8n?
The visual workflow builder is intuitive for anyone with basic technical skills. Most users can build simple workflows (email to spreadsheet, form to CRM) within a few hours. Complex workflows with custom code require JavaScript or Python knowledge, which a developer or consultant can provide.
Can n8n replace Zapier for my existing workflows?
Yes. n8n supports most of the same integrations as Zapier, and for integrations that n8n does not have a pre-built node for, you can use the HTTP Request node to connect to any API. Moving high-volume workflows from Zapier to n8n can cut execution costs significantly, but plan and test each migration rather than switching everything at once.
What happens if n8n goes down?
If self-hosted, n8n runs on your infrastructure, so its availability depends on your server uptime. We recommend deploying n8n on a reliable cloud provider (Azure or AWS) with automated restart monitoring. If an IT provider manages it for you, they should monitor n8n uptime and restart it automatically if it goes down.
Can n8n handle AI-powered document processing?
Yes. n8n has built-in AI nodes for OpenAI, Anthropic, and other LLM APIs. Common workflows extract data from PDFs, classify emails, generate report narratives, and analyse sentiment – all powered by AI through n8n.
Before You Automate
Most automation wins start with a clear map of the process. n8n makes workflows cheap to build, but a workflow built on a messy process just moves the mess faster. If you want help documenting and fixing a process before automating it, see our business process mapping service. Where the work is connecting your ERP to Xero, a CRM or an online store, see our ERP integration service.