n8n Plans & Pricing
n8n has one of the most compelling pricing models in automation software: if you can self-host on your own server using Docker, it's completely free with no limits on workflows, executions, integrations, or users. The managed cloud version starts at $20/month for teams that prefer infrastructure managed by n8n rather than their own server. This creates a pricing model that genuinely rewards teams with technical capability — the more self-sufficient your infrastructure management, the less you pay, down to zero. For teams without server management capability, the cloud pricing is competitive with alternatives that offer a fraction of n8n's power.
| Plan | Price | Best For |
|---|---|---|
| Self-hosted (Free) | Free | Individuals & light usage |
| Starter | $20/mo | Solo creators & freelancers |
| Pro Most Popular | $50/mo | Teams & power users |
| Enterprise | Custom | Enterprise & custom needs |
Is n8n Worth the Price?
The comparison against Zapier illustrates the pricing advantage most clearly. Zapier's Starter plan at $19.99/month allows 750 tasks — a 'task' being one action step in one workflow execution. A five-step workflow that runs 200 times per month costs 1,000 Zapier tasks, pushing you to the Professional plan at $49/month. n8n cloud's Starter at $20/month allows 2,500 complete workflow executions with unlimited steps — for complex automations with many steps, this is 10–15× more execution value per dollar. And self-hosted n8n is literally unlimited. For any team with workflows that involve more than a few steps or run frequently, n8n's economics are dramatically better than Zapier at any comparable price point.
Self-hosted: Free forever, unlimited workflows, unlimited executions, unlimited integrations. This is the right choice for any team with a developer who can set up and maintain a Docker container on a VPS. Cloud Starter ($20/month): 2,500 executions, 5 active workflows — appropriate for teams starting with automation who want managed infrastructure. Cloud Pro ($50/month): 10,000 executions, unlimited active workflows, priority support — for teams running automation at operational scale. Enterprise: custom pricing with SSO, advanced permissions, SLA guarantees, dedicated support — for large organizations with compliance and governance requirements.
n8n Free Trial — What's Included?
The self-hosted version of n8n is not a trial — it's permanently free. You can download, install, and run n8n on a $5/month DigitalOcean, Hetzner, or Linode VPS and operate it indefinitely at no additional cost beyond the server. For teams evaluating the cloud version, n8n offers a trial period on cloud plans — check n8n.io for current trial terms, as these are periodically updated. Most teams evaluating n8n start with the self-hosted version since it's free and functionally identical to the cloud version.
Frequently Asked Questions
Self-hosted n8n is free (beyond server costs of $5–20/month for a basic VPS). Cloud plans: Starter ($20/month, 2,500 executions, 5 active workflows), Pro ($50/month, 10,000 executions, unlimited workflows). Enterprise pricing is custom. For most teams with technical capability, self-hosted is the recommended starting point — you get unlimited everything at server cost only.
Yes. If you run n8n on your own server using Docker, it's completely free with no execution limits, workflow limits, or integration restrictions. A basic Hetzner or DigitalOcean VPS costs $5–6/month and handles the workflow automation needs of most small-to-medium teams comfortably. The n8n documentation includes step-by-step guides for setting up on common VPS providers — it's a 30-minute setup for anyone comfortable with the command line.
An execution is one complete run of a workflow — regardless of how many steps (nodes) that workflow contains. If your workflow has 30 nodes and runs once, that's 1 execution. This is dramatically more generous than Zapier's task-based billing where each step costs 1 task — a 30-step Zapier workflow costs 30 tasks per run, while the equivalent n8n workflow costs 1 execution per run. For complex multi-step automations, n8n's execution model provides far more value per dollar.
No. n8n workflows can have unlimited nodes (steps). There is no per-step limit or charge. This is a fundamental architectural difference from Zapier, which limits step counts or charges per step. In n8n, you can build workflows with dozens or hundreds of nodes without any additional cost — critical for complex automation patterns that require many processing, routing, and action steps.
Make's free plan includes 1,000 operations per month (comparable to tasks in Zapier). Make's Core plan at $9/month includes 10,000 operations. n8n cloud's Starter at $20/month includes 2,500 executions, but each execution covers all steps in a workflow rather than counting each step as an operation — a 10-step workflow in n8n uses 1 execution, while in Make it uses 10 operations. For complex multi-step workflows, n8n's execution model provides more value. Make is cheaper for simple low-step automations.
n8n cloud does not have a permanent free plan. Self-hosted n8n is the free option — it's functionally identical to the cloud version. n8n cloud provides a trial period for evaluation. For teams without server management capability who want to try n8n cloud before paying, the trial period is the evaluation path. Check n8n.io for current trial terms.
For a small team with at least one person comfortable with basic Linux server commands, self-hosted on a $5–10/month VPS is the better choice — identical functionality at server cost only, with full data sovereignty. For teams with no technical infrastructure capability at all, n8n cloud at $20/month is the right path. The cloud version provides all the same workflow capabilities without any server management overhead. The decision comes down to whether your team has someone willing to spend 30 minutes setting up and maintaining a Docker container — if yes, self-hosted is clearly more economical.
n8n cloud billing is monthly, charging the flat plan rate plus any overages if execution volume exceeds your plan's allocation. Executions are counted per complete workflow run, regardless of how many nodes (steps) that workflow contains. The Starter plan at $20/month includes 2,500 executions monthly; Pro at $50/month includes 10,000. If you regularly exceed your plan's execution allocation, upgrade to the next tier rather than paying recurring overage charges. Annual billing options may provide a cost reduction compared to monthly billing.
Non-technical operations teams can build useful workflows in n8n without writing code — the visual builder and 400+ pre-built integrations handle most common automation needs through configuration rather than programming. Connecting apps, setting up webhook triggers, filtering data, and routing to different actions are all achievable without code. Where technical knowledge helps: debugging complex workflows, writing code node JavaScript for custom logic, configuring self-hosted infrastructure, and building sophisticated AI agent workflows. For simple to intermediate automation, n8n is accessible to motivated non-developers.
Yes. n8n has native integrations with email providers (Gmail, Outlook, SMTP), messaging platforms (Slack, Teams, Discord, Telegram), and notification services. Setting up a workflow that sends an email notification when a specific condition is met is one of the simplest n8n workflows to configure. More sophisticated patterns include: escalating alerts that try Slack first then email if the first isn't acknowledged, personalized email sequences triggered by CRM events, and multi-channel notifications that inform different people about different parts of a workflow.
Yes. n8n has a template library at n8n.io/workflows with hundreds of community-contributed workflow templates for common use cases. Templates can be imported and customized, providing a starting point for standard automation patterns without building from scratch. The template library is browsable by category (CRM, marketing, DevOps, AI, data sync) and integration, making it easy to find relevant starting points. The community forum also has workflows shared by users as examples and learning resources.
Running n8n self-hosted on a basic VPS costs $5–20/month depending on your server provider and workflow volume. A Hetzner CX21 (€4.35/month) or DigitalOcean Droplet ($6/month) comfortably handles automation for teams with moderate workflow volume. For production environments with high execution frequency (thousands of daily executions), you may need a larger server at $20–40/month plus database costs. Adding a managed database (PostgreSQL on Supabase free tier or PlanetScale) adds $0–25/month depending on volume. All-in, a production-grade self-hosted n8n setup typically runs $10–50/month total — a fraction of Zapier or Make at comparable capability. This is the full picture: n8n is free software, infrastructure is the only real cost.
One n8n execution is one run of one workflow from start to finish, regardless of how many nodes (steps) that workflow contains. A workflow with 20 nodes that runs once consumes one execution. A simple 2-node workflow that runs once also consumes one execution. This is dramatically more favorable than Zapier's task-based counting, where each node action counts as a separate task — a 20-node workflow that runs once consumes 20 Zapier tasks. For complex automations with many steps, n8n's execution-based counting can be 5–20x more economical than Zapier's task-based model at equivalent plan pricing.
Yes — n8n Cloud and self-hosted n8n run the same software and include identical features. AI agent nodes, LangChain integration, OpenAI and Anthropic connections, and all other capabilities are available on both cloud and self-hosted. The differences are purely operational: cloud is managed by n8n (infrastructure, updates, uptime); self-hosted gives you full control and data sovereignty. There are no AI features exclusive to one deployment method. If you start on cloud and later want to move to self-hosted, your workflows export and import cleanly — the migration is straightforward since both environments run the same n8n core.