n8n0 Mixed

GILD Review

GILD's call and the buying facts behind it.

From GILD research

Should we adopt n8n?

Where does your use of n8n cross from the free internal-use license into a paid key?

30-Second TakePick n8n if a technical team wants automations and AI agents inside its own data boundary.
Best fitTechnical operations, IT, and platform teams automating internal processes and AI agents over company data.GILD Review
Consider another workflowIf business users will build and own most automations without technical help, a hosted tool such as Zapier asks less of them.GILD Review

Drawn from 97 sources and 3 operator insights

GILD analysis

n8n Review

Choose n8n when a technical team wants its automations and AI agents inside its own data boundary, with code wherever the visual nodes stop. Billing per workflow execution keeps long workflows cheap, and the free Community edition runs on your own servers for internal use. Your team owns hosting, upgrades, the license check, and agent review, and Git, SSO, and environments need a paid license.

Workflow Automation
AI Costs

The decision

Best fit: Technical operations, IT, and platform teams automating internal processes and AI agents over company data.
Consider another workflow: If business users will build and own most automations without technical help, a hosted tool such as Zapier asks less of them.
Adoption condition: Every production workflow has a named owner and an error workflow, and the license is checked before anything reaches customers.

Basis: n8n documentation, pricing, and license terms, and practitioners' own measurements, including a 30-day run of n8n beside Windmill and Temporal on one 4 GB server. Pilot check: log executions and failed runs per workflow for a month against the plan quota.

Buying & deployment factsWhat business and product leaders should verify before rolling it out across teams.

How you buy it

€20

A month for Cloud Starter with 2,500 executions, billed annually. Self-hosted Community is free.

Source 1, n8n, Pricing: Pricing

Details

n8n Cloud, priced in euros: Starter is €20 a month for 2,500 executions and Pro starts at €50 a month for 10,000, billed annually (€24 and €60 billed monthly). Self-hosted Community is free. Self-hosted Business is €667 a month billed annually for 40,000 executions. Enterprise is quoted.

Cost drivers

Per execution

A run costs the same at any number of steps. Manual runs and error workflows do not count.

Source 4, n8n, Understand executions: Understand executions Source 1, n8n, Pricing: Pricing Source 5, docs.n8n.io, Gateway credits: Gateway credits

Details

Production executions, whatever the number of steps. Manual runs, sub-workflows, and error workflows do not count. Business overage is €4,000 per extra 300,000 executions. Model usage is billed by your AI provider, or drawn from prepaid Gateway credits on Cloud Starter and Pro. The AI Assistant draws monthly credits that do not roll over: 1,600 on Cloud Starter and up to 9,600 on Pro. Self-hosted Assistant uses your own model key.

Where it runs

Frankfurt or yours

n8n Cloud is hosted in Frankfurt, or self-host on Docker or Kubernetes.

Source 1, n8n, Pricing: Pricing Source 6, n8n, Host n8n: Host n8n

Details

n8n Cloud is hosted in Frankfurt. Self-hosted n8n runs on Docker, Docker Compose, or Kubernetes on your own infrastructure. Business is self-hosted only, and queue mode on Cloud needs Enterprise.

What to verify

Internal use only

The license bars outside users. Pricing puts SSO on Business; the security page says Enterprise.

Source 2, n8n, License FAQ: License FAQ Source 1, n8n, Pricing: Pricing Source 7, n8n.io, Security at n8n: Security at n8n

Details

Whether your use fits the Sustainable Use License, which covers internal business use and bars outside users from building or editing workflows. Git and environments need a Business key, and external secrets need Enterprise. The pricing page lists SSO and LDAP on self-hosted Business, while n8n's security page says SSO, SAML, and LDAP come with Enterprise. Cloud keeps 2,500 saved executions in 2.5 GB for 7 days on Starter and 25,000 in 25 GB for 30 days on Pro, then drops the oldest. Cloud Starter stops an execution at 5 minutes.