n8n Cloud vs Self-Hosted: Cost, Control, and When to Switch

If you're evaluating n8n, the first decision is where to run it. n8n Cloud is managed by the n8n team - you sign up and start building. Self-hosted means you run n8n on your own server using Docker. Both give you the same workflow editor and the same nodes. The difference is who manages the infrastructure, what it costs, and how much control you have.

Cost comparison at different scales

The economics flip depending on how many workflows you run.

At low volume (under 5 active workflows, a few hundred executions per month), n8n Cloud is cheaper in total cost. The entry plan covers your needs and you spend zero time on server management. A VPS for self-hosting costs $10-20/month before you account for your time setting it up and maintaining it.

At medium volume (10-30 active workflows, thousands of executions per month), the costs converge. n8n Cloud's higher tiers add up, while a $20/month VPS still handles the load. The calculation depends on how you value your time for server maintenance versus the subscription price.

At high volume (50+ workflows, tens of thousands of executions per month), self-hosting wins decisively. n8n's self-hosted edition has no execution limits. Your cost is just the server. A $40-60/month VPS handles substantial workloads that would cost several hundred on cloud plans.

The hidden cost with self-hosting is your time. Plan for a few hours of initial setup and maybe an hour per month for updates and monitoring. If that time is worth more than the cloud subscription difference, Cloud makes more sense.

What you get with n8n Cloud that self-hosting doesn't include

Automatic updates. n8n Cloud runs the latest stable version without you doing anything. Self-hosted means pulling new images and restarting containers yourself. n8n releases frequently, so this adds up.

Managed infrastructure. No Docker, no Postgres configuration, no SSL certificates, no reverse proxy setup. n8n Cloud handles uptime, scaling, and backups.

Built-in auth and user management. n8n Cloud includes multi-user access with role-based permissions out of the box. Self-hosted Community Edition has more basic user management.

Support. Cloud plans include direct support from the n8n team.

What self-hosting gives you that Cloud doesn't

No execution limits. This is the big one. Run as many workflows as your server can handle.

Data sovereignty. Your workflow data, credentials, and execution logs stay on your infrastructure. Nothing passes through n8n's servers. This matters for HIPAA, GDPR with strict data residency, or any industry where data handling is audited.

Full network control. You choose which ports are open, which IPs can access the editor, and how the instance connects to your internal services. If your workflows need to reach databases or APIs on a private network, self-hosting avoids the complexity of setting up VPN tunnels or IP allowlists.

Custom configuration. You control the Postgres version, Node.js memory limits, execution pruning settings, and every environment variable. Cloud gives you some configuration options but not all.

Version pinning. You choose when to upgrade. If a new n8n version introduces a breaking change, you can stay on the previous version until you're ready.

When to start with Cloud and when to start self-hosted

Start with Cloud if: you're evaluating n8n for the first time, you don't have someone comfortable with Docker and Linux, you want to build workflows immediately without setup time, or your execution volume is low to moderate.

Start self-hosted if: you have data compliance requirements that mandate self-hosting, you're running high-volume automations where execution costs matter, you need to connect to internal services on a private network, or you already have server infrastructure and Docker experience.

Migrating from Cloud to self-hosted

This is a common path. Start on Cloud to validate that n8n works for your use case, then move to self-hosted when volume grows or compliance requirements kick in.

The migration process: export your workflows from n8n Cloud (they're JSON files), set up a self-hosted instance with Docker and Postgres (we have a complete guide for that), import your workflows, and re-enter your credentials (these don't export for security reasons).

The workflows themselves are fully portable. The only manual step is re-configuring credentials and webhook URLs, since your self-hosted instance will have a different domain.

Going the other direction (self-hosted to Cloud) works the same way. Export, import, reconfigure credentials.

The setup for self-hosting

If you decide to self-host, the stack is straightforward: Docker, Postgres, and a reverse proxy for HTTPS. A basic compose file with healthchecks, named volumes, and environment variables gets you running in under 30 minutes.

We maintain a production-ready docker-compose.yml with full environment variable reference that covers the complete setup including backups, upgrades, and Caddy as a reverse proxy.

The minimum server requirements: 2GB RAM, 20GB SSD, and a Linux host with Docker installed. This handles 10-20 active workflows comfortably. Scale up the VPS as your workflow count grows.

FAQ

Can I try n8n Cloud before committing?

Yes. n8n Cloud offers a free trial. You can build and test workflows without a credit card. This is the fastest way to evaluate whether n8n fits your use case before deciding on Cloud vs self-hosted.

Are there features only available on Cloud?

n8n Cloud Enterprise plans include advanced features like SAML SSO, audit logs, and source control integrations. The Community Edition (self-hosted) has a different feature set. The core workflow editor and all standard nodes are the same across both.

How hard is it to self-host n8n?

If you're comfortable with Docker and basic Linux commands, the initial setup takes 20-30 minutes. Ongoing maintenance is minimal - mainly updating the n8n image periodically and monitoring disk space. If Docker is new to you, plan for a longer learning curve or start with Cloud.

Can I run n8n self-hosted on a Raspberry Pi?

Technically yes, but it's not recommended for production. A Raspberry Pi 4 with 4GB RAM can run n8n for light workloads. For anything beyond testing, use a proper VPS or cloud server with more memory and reliable storage.