On this page
Put Coworker to work on your stack.
Connect Salesforce, Slack, Jira and run your first agent in minutes.
Book a demoEnterprise AI
n8n Pricing 2026: Plans, Execution Limits, and Real Costs
What n8n costs in 2026: Starter, Pro, Business, Enterprise, how execution billing works, real self-host costs, and how it compares to Zapier and Make.
n8n pricing is execution-based: cloud plans are billed by monthly workflow executions rather than by seat. The Starter plan is 20 euro per month billed annually (24 euro billed monthly) for 2,500 executions, Pro is 50 euro per month billed annually (60 euro monthly) for 10,000 executions, and the self-hosted Business tier lists 667 euro per month billed annually for roughly 40,000 executions with SSO. Enterprise is custom, with 2026 buyer reports putting it around $2,000 to $3,000 per month. The self-hosted Community Edition is free software with unlimited executions, so the real cost there is whatever server you run it on.
If you are evaluating n8n in 2026, here is what each tier actually includes, why the execution count is the number that matters (one 2026 breakdown found a single Gmail polling workflow can exhaust Starter's entire 2,500-execution allowance in under 11 days), how the bill compares to Zapier and Make for the same workload, and what self-hosting really costs once you count maintenance.
n8n pricing at a glance
All plans from n8n's official pricing page, verified July 2026:
| Plan | Price | Executions/month | Notes |
|---|---|---|---|
| Community Edition (self-hosted) | Free | Unlimited | You pay only for your own server (roughly $3 to $7/month on a small VPS) |
| Starter | 20 euro/mo billed annually (24 euro monthly) | 2,500 | 1 shared project, 5 concurrent executions, 50 AI Workflow Builder credits |
| Pro | 50 euro/mo billed annually (60 euro monthly) | 10,000 | 3 shared projects, 20 concurrent executions, 150 AI credits, admin roles |
| Business (self-hosted) | 667 euro/mo billed annually | ~40,000 | SSO, LDAP, environments, and governance features |
| Enterprise | Custom | Unlimited | Dedicated infrastructure; 2026 buyer reports cite roughly $2,000 to $3,000/mo, more for on-prem with support |
n8n lists prices in euros; US-facing third-party breakdowns (InstaPods, TechJack) report the dollar equivalents as roughly $20 to $24 for Starter and $50 to $60 for Pro. Annual billing saves about 17% on every cloud tier. All plans include unlimited users, unlimited workflows, unlimited steps per workflow, and every integration. The only thing that scales the price is executions.
The Enterprise number deserves a caveat: n8n does not publish it, so treat the $2,000 to $3,000 monthly figure from buyer discussions as a starting point for negotiation, not a rate card.
How n8n's execution-based pricing works
The core rule: one workflow run counts as one execution, no matter how many steps it has. A five-step workflow and a fifty-step workflow both cost the same per run. That makes the pricing simple to reason about, but it also means the plan you need depends entirely on how often your workflows fire, not how complex they are.
April 2026 update: n8n removed active-workflow limits across every plan, so you can now run as many workflows as you want on any tier. The constraint that remains is total monthly executions, not workflow count.
How many executions will you actually use?
This is the question that decides your plan, and most teams guess wrong because triggers matter more than workflow count. Three worked examples:
Scenario 1: webhook-triggered lead router. A workflow fires when a form is submitted, enriches the lead, and posts to Slack and your CRM. At 30 submissions a day, that is about 900 executions a month. Starter's 2,500 covers it with room to spare, no matter how many steps the workflow has.
Scenario 2: inbox polling. A workflow checks an inbox every 5 minutes. That is 288 executions a day, roughly 8,640 a month, from ONE workflow that mostly finds nothing new. This blows through Starter in under 9 days and eats most of Pro. If you can switch a polling trigger to a webhook or event trigger, you should; it is the single biggest lever on an n8n bill.
Scenario 3: an ops team at scale. Ten workflows averaging 80 runs a day between them is about 24,000 executions a month. That is Business territory on cloud pricing, and exactly the point where the free self-hosted Community Edition starts looking attractive if you have anyone comfortable running a server.
The pattern: count your trigger frequency, not your workflows. Frequent pollers dominate the bill; webhook-driven workflows are nearly free.
n8n vs Zapier vs Make: the same workload, three bills
The three tools bill different units, and the difference compounds fast. n8n bills per workflow RUN. Zapier bills per task, which means every action step in a Zap consumes one task. Make bills per operation (now sold as credits), which is also per module step.
Take one workflow with 5 action steps that runs 1,000 times a month:
| Platform | Unit billed | Units consumed | Cheapest plan that covers it |
|---|---|---|---|
| n8n | Execution (per run) | 1,000 | Starter, 20 euro/mo (2,500 executions) |
| Zapier | Task (per step) | 5,000 | Professional at $19.99/mo includes only 750 tasks; you need a much larger task allowance |
| Make | Operation/credit (per step) | 5,000 | Core, $12/mo (10,000 credits) |
Three honest takeaways. First, Make's Core plan at $12 a month for 10,000 credits is the cheapest paid entry point of the three for moderate volume. Second, Zapier's per-task model is the most expensive at volume, and that is the price of its strength: the largest integration catalog and the least setup friction of the three. Third, n8n's per-run model wins specifically when workflows are long: a 50-step workflow costs the same execution as a 2-step one, which is why technical teams with complex automations gravitate to it.
Coworker
See what your AI stack really costs
Compare model and platform costs, then run it all in one place.
Open the free LLM cost calculatorWhat actually drives your n8n bill
1. Execution volume, not workflow count or complexity. Every plan allows unlimited workflows, so your bill is a function of how many times those workflows run per month.
2. Cloud vs self-hosted. n8n Cloud (Starter, Pro) is fully managed. Self-hosting the Community Edition is free software with unlimited executions, but you take on your own server, uptime, and maintenance.
3. Concurrency limits. Starter allows 5 concurrent executions, Pro allows 20. If your workflows run in parallel bursts rather than one at a time, you can hit the concurrency ceiling before the monthly cap.
4. Governance features push you to Business or Enterprise. SSO, LDAP, environments, and advanced admin controls sit on the higher tiers, not the self-serve cloud plans.
What self-hosting n8n really costs
The Community Edition's license price is zero, and for a hobbyist a $3 to $7 VPS genuinely runs it, per InstaPods' 2026 cost breakdown. For production use the honest math is different:
- Infrastructure at scale: CheckThat's 2026 analysis puts realistic production infrastructure at 80 to 300 euro a month depending on volume, once you size for reliability rather than experimentation.
- Maintenance time: ExpressTech's self-hosting analysis argues the all-in number is meaningfully higher once someone's hours for updates, backups, monitoring, and the occasional 2am outage are counted.
- The breakeven: if your execution volume would put you on Business (667 euro/mo billed annually), self-hosting pays for itself quickly even with real infrastructure spend. If you would comfortably fit in Starter or Pro, managed cloud is usually cheaper than your own time.
Hidden costs to check before you commit
- The execution cap trap. Starter's 2,500 executions can disappear fast on any polling-style workflow. Estimate your real monthly execution count before picking a plan.
- Annual vs monthly billing. Annual billing saves roughly 17% across every cloud tier.
- Concurrency, not just volume. Bursty parallel workloads hit the concurrent-execution ceiling (5 on Starter, 20 on Pro) before the monthly cap.
- Pricing model changes. n8n moved to execution-based billing in its 2025 to 2026 update and removed workflow limits in April 2026. The model has changed twice in two years, so confirm current terms on the official page rather than relying on older writeups.
n8n vs Coworker: workflow automation vs an AI coworker
n8n is a workflow automation tool you build flows in. Coworker is an AI coworker that plans and acts across your tools directly, without you first designing the workflow.
| Factor | n8n | Coworker |
|---|---|---|
| Entry price | Free (self-hosted) or 20 euro/mo (Cloud Starter, annual) | $0 free trial, then $29.99/user/mo |
| Pricing model | Per monthly execution | Per user, flat |
| Self-serve signup | Yes | Yes |
| What you build | Visual workflows you design and connect node by node | An AI agent you direct in plain language |
| Best fit | Teams that want to design and own explicit automation logic | Teams that want an agent to plan and execute multi-step work without building flows first |
| Security | Self-hosted option available; SSO on Business/Enterprise | SOC 2 Type II, GDPR, CASA Tier 2, US-hosted models |
The practical difference: n8n is infrastructure you assemble. Coworker is an agent you direct. If your team wants full control over exact workflow logic and is comfortable maintaining it, n8n's flexibility is real. If you want an assistant that reads your tools, plans the steps, and does the work with a human approving what matters, that is a different category. See AI agent use cases and agentic AI vs generative AI for the fuller distinction, and enterprise AI pricing compared for the category view.
Who n8n is a good fit for
Honestly: n8n is a strong, mature workflow automation platform, especially for technical teams that want to design exact automation logic node by node and either self-host for full control or pay for managed executions at scale. The free, unlimited-execution Community Edition is a genuinely good deal if you can run your own server, and the per-run billing model is the cheapest of the major platforms for long, complex workflows.
It is a weaker fit if you want the automation to require no building at all, if unpredictable execution counts make budgeting hard for high-frequency workflows, or if what you actually need is an agent that plans multi-step work on its own rather than a canvas you configure. For that, see AI agent workflow and no-code AI agent builder. For the broader market numbers behind this category, see workflow automation statistics.
Frequently asked questions
How much does n8n cost? n8n Cloud's Starter plan is 20 euro per month billed annually (24 euro billed monthly) for 2,500 monthly workflow executions, and Pro is 50 euro per month billed annually (60 euro monthly) for 10,000 executions. US-facing breakdowns report roughly $20 to $24 and $50 to $60. The self-hosted Community Edition is free software with unlimited executions.
Is n8n free? The self-hosted Community Edition is fully free with unlimited workflows and executions; you pay only for your own server. n8n Cloud, the managed version, starts at 20 euro per month billed annually.
How does n8n's pricing model work? n8n charges by monthly workflow executions rather than by seat or step. One workflow run, regardless of how many steps it contains, counts as one execution. All plans allow unlimited users and workflows; only execution volume is capped.
Is n8n cheaper than Zapier? For multi-step workflows, usually yes. n8n bills one execution per workflow run while Zapier bills one task per action step, so a 5-step workflow consumes 5x more billable units on Zapier per run. Zapier's entry Professional plan is $19.99 per month for 750 tasks; n8n Starter is 20 euro for 2,500 full runs. Zapier's advantages are its larger integration catalog and lower setup friction, not price at volume.
How much does n8n Enterprise cost? n8n does not publish Enterprise pricing. Buyer reports from 2026 put typical contracts around $2,000 to $3,000 per month, more for on-prem deployment with support. Treat those numbers as negotiation reference points, not a rate card.
What is the difference between n8n Cloud and self-hosted? n8n Cloud is fully managed and billed on the tiers above. Self-hosting the Community Edition is free software with unlimited executions, but you run and maintain your own server: $3 to $7 a month for a hobby VPS, realistically 80 to 300 euro a month plus maintenance time for production reliability.
Why did n8n's execution limits change in 2026? In April 2026, n8n removed active-workflow limits across every plan, so any tier now supports unlimited workflows. The constraint that remains is the monthly execution cap for your plan.
Is there an alternative to building workflows in n8n? Coworker is an AI coworker priced at $29.99 per user per month that plans and executes multi-step tasks across 50+ connected tools directly, without requiring you to design a visual workflow first.
Related reading
Ready to get started?
Put Coworker to work inside your actual stack
Connect Salesforce, Slack, Jira, whatever you use, and run your first agent in minutes.