n8n Pricing: Workflow Executions Are Not LLM Tokens

Count executions at the orchestration layer

n8n’s pricing page states that the charge-driving unit on its hosted plans is monthly workflow executions, with unlimited workflow steps on listed tiers. Under the displayed annual-billing mode, Starter is €20 per month for 2,500 executions, Pro is €50 per month for 10,000 and Business €667 per month for 40,000. Business is shown as self-hosted only on that plan comparison. These are annual-commitment equivalents, not ad-hoc month-to-month invoices; country and tax conditions still matter.

AI spend remains its own ledger

An n8n workflow might contain an LLM call, web retrieval, a browser action, a media generation step and an email send. The workflow execution counter is not the same as the tokens billed by the LLM provider, or the minutes billed by a transcription provider. Included assistant credits in n8n may apply to specific platform assistance features, not automatically to arbitrary external APIs. Store the platform execution meter and every third-party provider charge in separate fields.

High step count can still matter operationally

The stated unlimited-step pricing removes a simple per-step bill for eligible workflow executions, but complex workflows can consume more compute, queue time, external requests and support attention. Self-hosted business deployments may incur infrastructure, observability, backups, security and maintenance beyond the n8n license or tier fee. A reader comparing hosted and self-hosted editions should count all direct operational expenses and staff time rather than just one advertised subscription number.

Workload scenario with retries

A small agency forecasting 6,000 completed eligible executions a month would exceed a 2,500-execution Starter allowance if all were billable under that plan. That does not automatically say which plan is cheapest: queue behaviour, concurrency, billing terms, permission limits and overage policy need evaluation. Track retries, scheduled jobs, webhook runs and AI API traffic independently, then calculate actual monthly total as platform fee plus allowable overage, model tokens, media units, storage and taxes.

Before choosing a paid tier

Obtain the billing quote for the exact n8n product, account geography, tier, model or subscription, currency and commitment length. If the provider offers several kinds of meters, forecast each independently rather than summing quantities in incompatible units. The relevant captured meters here include monthly_workflow_executions, per_month_equivalent_annual_billing. Build a low, expected and high workload from your own request logs or editorial schedule; a pricing page is not a credible estimate of how many retries or approvals your team will need.

For governance, add an approval threshold for paid overages, reconcile invoices with metered activity, and schedule a quarterly or release-triggered source audit. The absence of an identified enterprise rate, promotional expiry or regional tax line is not evidence that the amount is zero. This guide does not represent a product trial, capability benchmark, procurement offer or warranty of uninterrupted access.

Historical availability and boundaries

Preserve the precise source version, checked date and any stated effective date before reusing these figures. A model launch announcement may confirm an introductory token rate without proving which private offers remain available; a mutable plan page can describe a later product state. If a source changes after this cutoff, publish a new snapshot instead of silently editing the historical record. Specialised product evaluations belong to the relevant AI 02–09 editorial teams, while this article documents the billing model and its limitations.