n8n Pricing in 2026: What Starter, Pro and Business Cost, and the Trigger That Decides Your Plan
As of October 6, 2026, n8n Cloud Starter is €20 a month billed yearly (€24 monthly) for 2,500 executions, Pro €50 (€60 monthly) for 10,000, and self-hosted Business €667 (€800 monthly) for 40,000. The Community Edition is free. n8n bills whole workflow runs, so how often your schedule triggers fire can matter more than your volume.
How much does n8n cost? The short answer
As of October 6, 2026, n8n Cloud Starter costs €20 a month billed yearly (€24 billed monthly) for 2,500 workflow executions, Pro costs €50 a month billed yearly (€60 monthly) for 10,000, and the self-hosted Business plan costs €667 a month billed yearly (€800 monthly) for 40,000. The Community Edition is free to self-host. n8n bills per workflow run, not per step, so the number that decides your plan is how often your workflows start, and a schedule trigger can use more of it than every lead you get.
n8n prices in euros. Every figure below comes from n8n's own pricing page and its documentation on how executions are counted, both read October 6, 2026, including the plan data the pricing page loads for its tier selector.
The plans side by side
| Plan | Billed yearly | Billed monthly | Executions a month | Where it runs | Limits that matter |
|---|---|---|---|---|---|
| Community Edition | €0 | €0 | No plan quota | Your own server | You run updates, backups and security yourself |
| Starter | €20/mo (€240/yr) | €24 | 2,500 | n8n Cloud | 5 concurrent runs, 5-minute max run time, 1 shared project, 7 days of execution logs |
| Pro | €50/mo (€600/yr) | €60 | 10,000 | n8n Cloud | 20 concurrent runs, 40-minute max run time, 3 shared projects, admin roles, workflow history |
| Pro, larger tier | about €121/mo (€1,450/yr) | €145 | 50,000 | n8n Cloud | Up to 50 concurrent runs, up to 9,600 Assistant credits |
| Business | about €667/mo (€8,000/yr) | €800 | 40,000 | Self-hosted only | SSO and LDAP, Git version control, multiple environments, queue mode |
| Enterprise | Contact sales | Contact sales | Custom | Cloud or self-hosted | 200+ concurrent runs, 365 days of insights, dedicated support with an SLA |
Every plan includes unlimited users, unlimited active workflows and every integration. Yearly billing is labeled "Save 17%," and the math checks out on each tier: €240 against €288 for Starter, €600 against €720 for Pro. Companies under 20 employees can apply for n8n's Start-up Plan, which takes 50% off Business.
Business sits above Pro on the page, but it is not simply a bigger Pro. It is only available self-hosted for now, it is self-serve with forum support, and its price buys governance features. Per run it is the most expensive plan on the page: €800 for 40,000 executions is 2 cents each, against 0.6 cents on Pro's 10,000 tier billed monthly.
Free playbook
What to Automate First
The SMB playbook for picking your first (and next) AI automation — the same prioritization we use with clients. PDF, emailed instantly.
Also subscribes you to The Service Business AI Brief. No spam, unsubscribe anytime.
The Business ladder past 40,000 executions
The page shows the entry price only. The tier selector carries the rest:
| Executions a month | Billed monthly | Billed yearly (per year) |
|---|---|---|
| 40,000 | €800 | €8,000 |
| 50,000 | €985 | €9,852 |
| 75,000 | €1,456 | €14,555 |
| 100,000 | €1,726 | €17,260 |
| 200,000 | €2,639 | €26,390 |
| 500,000 | €5,245 | €52,450 |
Go past your Business or Enterprise quota and n8n keeps your workflows running. Its FAQ says overage charges "may apply if you do not upgrade," sales reaches out first, and unresolved overages are invoiced 45 days later. On Business the overage price is €4,000 per extra bucket of 300,000 executions. The pricing page does not describe what happens when a Starter or Pro account runs out, so check your account settings before you count on it.
What counts as an execution, and what does not
An execution is one run of a whole workflow. A ten-step workflow that creates a contact, calls an AI model and sends three messages is still one execution. That is the whole pitch of n8n's pricing, and it holds. The details that change your count are in the docs:
- Only production runs count. Running a workflow from the editor while you build or test it is free.
- Schedule triggers count every time they fire, "regardless of outcome." A check that finds nothing still uses one execution.
- Polling triggers count only when they find new data. A Google Drive trigger that checks and finds no new file uses nothing.
- Webhook triggers count once per inbound request, including requests with an empty body. Malformed requests that fail before the workflow starts do not count.
- Sub-workflows and error workflows are free. When one workflow calls another, only the parent run counts, and runs of an error workflow do not count.
The second rule is the one that surprises people. n8n's own estimating guide puts a workflow scheduled every 5 minutes at about 8,600 to 8,900 executions a month. That one workflow is more than three times the Starter plan, even if it never finds anything to do.
A worked example: 300 leads a month
Take a home-services company that gets 300 web leads and about 150 missed calls a month, and automates four things in n8n:
- Lead intake. The website form posts to a webhook. One run scores the lead with an AI model, creates the CRM contact, texts the lead, books a follow-up task and posts to the team chat. 300 executions.
- Missed-call text-back. The phone system posts each missed call to a webhook, and the workflow sends a text. 150 executions.
- Daily summary. A schedule trigger once a day. 30 executions.
- Follow-up chaser. A schedule trigger checks the CRM for leads nobody has called back. Its frequency is the variable.
| Chaser runs | Chaser executions | Total a month | Plan that fits | Cost (yearly / monthly billing) |
|---|---|---|---|---|
| Every 5 minutes | 8,640 | 9,120 | Pro 10,000 | €50 / €60 |
| Every 15 minutes | 2,880 | 3,360 | Pro 10,000 | €50 / €60 |
| Every hour | 720 | 1,200 | Starter 2,500 | €20 / €24 |
| Hourly, 8 a.m. to 6 p.m. on 22 workdays | 220 | 700 | Starter 2,500 | €20 / €24 |
The 300 leads and 150 calls use 480 executions between them. The chaser's clock uses anywhere from 220 to 8,640. Whether this business pays €20 or €50 a month is decided by one schedule setting, not by how much work the automation does. And at hourly, the company could quadruple its leads before Starter runs out.
The same intake run looks very different on a per-step tool. Zapier bills each successful step, and its rates page lists an AI step at 1, 3 or 5 tasks by model tier, so this lead costs roughly 5 to 10 tasks there instead of one execution. Our breakdown of how much Zapier costs runs that same 300-lead workflow through Zapier's task ladder, and our Make.com pricing guide does it in Make's credits, where every polling check is billed.
Cloud or self-hosted: what the free version really costs
The Community Edition is free on GitHub, and n8n does not meter its executions. What you pay instead is the server, and the time to keep it patched, backed up and running. If a workflow handles customer data, that time is not optional.
Two more things to know before you choose. n8n Cloud stores data in the EU, on servers in Frankfurt; self-hosting puts it wherever you run it. And Business features on a self-hosted instance work through a license key that pings n8n's license server daily, including your production execution count, so Business is metered even when you host it.
For most small businesses running a few lead and follow-up workflows, Starter or Pro is cheaper than an hour a month of server upkeep. Self-hosting starts to pay when your volume runs to hundreds of thousands of executions, or when your data cannot leave your own infrastructure.
Is n8n worth it for a small business?
On price per run, yes, once your workflows have more than a few steps. At 10,000 executions for €60 a month, each run costs 0.6 cents however many steps it has, and AI steps do not multiply the bill the way they do on per-step tools. You still pay your AI model provider for tokens.
It is less of a fit if nobody on the team will build and maintain the workflows. n8n's editor is more technical than Zapier's, the Starter and Business cards list forum support, and dedicated support with an SLA is Enterprise only. The software is cheap. The building is where the cost goes.
Who can build an n8n automation for my business?
If you want the workflow built rather than learning n8n yourself, you can hire a freelancer, an automation agency or a studio. Our ranking of the best AI automation agencies compares who does this work and how they price it, and our guide to AI automation pricing covers what the market charges per project and per month.
SuperDupr is one of those options. We are an Austin studio that builds AI systems for service businesses, and our published price for AI workflow automation is $10,000 to $25,000 for a connected build, with $200 to $2,000 a month ongoing depending on volume. Most builds go live in 2 to 4 weeks, the system belongs to your business, and we pick the platform, n8n or otherwise, by what it costs at your volume. The first step is a free 30-minute strategy session, booked from that page.
How to keep an n8n bill on the smallest plan
- Prefer webhooks to schedules. A webhook runs only when something happens. A schedule runs whether or not anything did.
- Slow the schedules you keep. Ask how fast a follow-up really needs to be. Hourly during business hours is 220 runs a month; every 5 minutes is 8,640.
- Use polling triggers where the app offers them. Empty polls are free on n8n.
- Split big workflows into sub-workflows freely. Only the parent run counts.
- Check Insights mid-month. n8n shows production execution counts on the Insights dashboard.
Sources
- n8n pricing page, n8n.io/pricing, read October 6, 2026: plan cards, comparison table, FAQ, and the tier data loaded by the page's plan selector.
- n8n documentation, "Understand executions," docs.n8n.io/workflows/executions, read October 6, 2026: how executions count toward quotas by trigger type, and what does not count.
- Zapier task usage rates, zapier.com/pricing/rates, read October 6, 2026: tasks per step and per AI step by model tier.
Frequently Asked Questions
-
As of October 6, 2026, n8n Cloud Starter is €20 a month billed yearly, or €24 billed monthly, for 2,500 workflow executions. Pro is €50 a month billed yearly, or €60 monthly, for 10,000, with a 50,000-execution tier at €145 monthly. The self-hosted Business plan starts at €800 a month, or €8,000 a year, for 40,000 executions. Enterprise is quoted.
-
The self-hosted Community Edition is free on GitHub and has no execution quota, but you pay for the server and the time to maintain it. The paid Cloud plans start with Starter, and new Cloud accounts get a free trial capped at 1,000 executions.
-
One production run of a whole workflow, however many steps it has. Manual test runs, sub-workflow runs and error-workflow runs do not count. Schedule triggers count every time they fire, even if they find nothing, while polling triggers count only when they find new data and webhooks count once per inbound request.
-
n8n lists its plan prices in euros on its pricing page, so a card billed in dollars will see the converted amount. For the annual Business plan paid by invoice, n8n offers a contract form for dollar payment.
-
On Business and Enterprise, workflows keep running, n8n's sales team reaches out, and unresolved overages are invoiced 45 days later; on Business the overage price is €4,000 per extra 300,000 executions. The pricing page does not describe the Starter or Pro behavior, so check your account before relying on it.
-
Usually, once workflows have several steps. n8n charges one execution per run, while Zapier charges per successful step and 1, 3 or 5 tasks per AI step depending on the model tier. A lead workflow with an AI step is one n8n execution and roughly 5 to 10 Zapier tasks. Frequent schedule triggers are the main way an n8n bill grows faster than expected.