n8n pricing for a monthly operations workflow

You are renewing the automation budget, and the workflow now handles more orders than it did at launch. Start an n8n pricing review with the production runs in a normal month and the people who maintain them. A long workflow and a frequently triggered workflow create different budgeting questions.

For the illustrative 6,000-run monthly workload below, evaluate Cloud Pro if you want managed hosting and its access features fit. Consider Community when your team can operate the installation and accept its collaboration restrictions. Keep an existing setup that meets the workload and access requirements unless the expected savings cover migration and maintenance. This assessment uses official documentation checked October 23, 2026; the workflow has not been tested firsthand.

Count the runs your operations process creates

The n8n pricing page bills by workflow executions, with one execution representing an entire workflow run rather than each step. Its listed plans include unlimited users and workflows. Put execution volume, permissions, and operational responsibility on separate lines in the purchase request.

Suppose an order-review workflow runs 200 times a day for a 30-day month. That produces 6,000 planned executions. Assume each run checks one order against customer and shipment records, then prepares an exception entry for a reviewer. The budget assumes three maintainers and no extra scheduled workflows; it is an example to replace with your own counts.

A daily batch would have a different run count from a workflow triggered for every order. Choose the schedule around the required response time, then estimate usage. Batching can also change failure recovery: if a run stops after some records have been processed, the restart must identify which records still need attention.

Use a representative month of execution history, including the busiest period. Ask how failed runs, retries, sub-workflows, and manual tests apply to your contract before adding a contingency. Do not assume that every entry in a debug log is a billable production execution or that a retry is free.

Price 6,000 monthly runs over one year

These calculations use the euro prices displayed by n8n, annual billing, and twelve identical months. Taxes, implementation, external service subscriptions, and AI model usage are excluded. Keep the currency fixed when comparing a quote rather than combining a dollar hosting estimate with a euro license price.

These are advertised annual-rate calculations, not month-to-month offers. Pro leaves 4,000 executions of room against the stated 6,000-run assumption, but that does not establish capacity for a burst of simultaneous orders. Obtain the applicable concurrency, duration, and overage terms for the selected tier before signing.

For a hypothetical Community budget, assume €40 a month for infrastructure and four maintenance hours at an internal planning rate of €75 per hour. That is €480 + €3,600 = €4,080 a year before initial setup. These invented budget inputs illustrate the calculation; they are neither a hosting quote nor a claim that every team needs four hours.

Apply those same assumed operations costs to Business and the example becomes €8,004 + €4,080 = €12,084. A managed service can still require workflow maintenance, so include that labor on both sides. Attribute only server administration that the managed option removes to a hosting saving.

Check who can inspect and change the workflow

The Community edition comparison excludes shared projects, workflow and credential sharing, SSO, and Git version control. It says workflows and credentials are accessible to their creator and the instance owner. A free license therefore needs a separate review when several employees must maintain the same production process.

For the three-maintainer example, specify who can edit logic, use credentials, inspect execution data, and publish changes. Give the backup maintainer a real account during evaluation. If the process depends on borrowing an owner login, the proposed operating model has failed its access requirement.

The pricing matrix places Cloud SSO and audit logging in Enterprise; Business advertises SSO for self-hosting. If either control is mandatory, price that requirement before approving a lower tier. Ask whether the evidence retained by the chosen plan meets your investigation needs.

A workflow that uses a shared service credential also needs a policy for which business records it can read. A successful connection test does not prove that every recipient should see its output. The enterprise app development article covers the broader ownership decision when an automation becomes a maintained internal application.

Try the same order-review fixture in each deployment

Create a small synthetic input set with OrderID, CustomerID, OrderedAt, ExpectedShipDate, and SourceRecordURL. Include an ordinary order, a missing customer match, and a repeated OrderID. Use a separate shipment file with its own update timestamp, then define when an order should enter the exception register.

In both the cloud and self-hosted candidate, configure the trigger, load the same input, apply the same matching rule, and write to a test-only destination. The expected result is a traceable exception for the unresolved customer and one retained identity for the repeated order. A reviewer should be able to see which source record and rule produced each result.

Disconnect the shipment source and repeat the run. Require a visible failure or stale-input status rather than a clean-looking report based on missing evidence. Then restore the connection and replay the same input to check for duplicate output. These are proposed acceptance checks, not claims about observed n8n behavior.

For self-hosting, add a restart and recovery exercise. The hosting documentation describes deployment on your infrastructure; your budget still needs an owner for updates, backups, monitoring, and recovery. Record the recovery steps somewhere the backup maintainer can access without the primary maintainer.

If you are comparing execution billing with per-module usage, the Make pricing calculation shows how processing several line items changes the estimate. Use the same inputs and expected output before comparing the two meters.

Where Permute fits: evidence behind the exception

Permute is the connected business data and custom application software we sell. When the CRM customer and ERP order use different identifiers, we can help establish matching rules while retaining source evidence. Uncertain matches should remain visible for review rather than becoming an assumed customer relationship.

We can apply a declared definition of an order exception to that connected data and make the resulting view available under current permissions. An operations owner can inspect the source and freshness behind a flagged order. That work addresses the reliability of the input and result; it does not establish which automation subscription you need.

Confirm the delivery boundary before buying

We do not provide an n8n license, administer your installation through this offer, or promise a native n8n connector. Source access and any implementation work need confirmation. This article does not promise external write-back or certify that a deployed workflow enforces your business permissions.

Move credentials and triggers with a rollback plan

The n8n export documentation supports workflow JSON transfer and warns that exports contain credential names and IDs; HTTP nodes can also contain authentication headers. Inspect an export before sharing it. Importing workflow logic is not evidence that the replacement has working credentials or equivalent access.

Inventory callbacks, schedules, environment settings, destination tables, and error notifications. Reconnect each dependency in the target environment and run the fixture before changing the live trigger. Keep one live owner of each trigger during cutover so the same order does not launch duplicate work in two deployments.

Reconcile output counts and unresolved exceptions, then test rollback with the last accepted input. The ERP data migration checklist extends that acceptance work when the business systems are moving too. Retain the old environment until the agreed recovery window closes.

Renew against the workload and operating responsibility

For the example workload, Pro covers the planned execution count at a €600 annual subscription calculation. Self-hosting needs an infrastructure and maintenance budget, and access requirements can change the edition you need. Approve the deployment after the same input and recovery checks pass. Keep the current system when its cost and behavior still meet those requirements.

Questions about budgeting n8n