Skip to content
Pricing guide

n8n Pricing (2026): Cloud Plans, Free Option, Real Costs

n8n pricing for 2026 explained: Cloud plans from €24/month, what an execution is, whether n8n is free, self-hosting costs and three worked cost examples.

The short answer
  • n8n Cloud costs €24/month for Starter (2,500 executions) and €60/month for Pro (10,000 executions) billed monthly, or €20 and €50 billed annually (November 2026).
  • One execution is one full workflow run, no matter how many steps it has. That makes n8n cheap for long, multi-step workflows.
  • n8n is free if you self-host the Community edition. You still pay for a server, roughly $5 to $25 a month for a small business setup, plus your time.
  • AI model usage (OpenAI, Claude and others) is billed separately by the model provider.

n8n pricing in short: n8n Cloud starts at €24 a month (Starter, 2,500 executions) or €60 a month (Pro, 10,000 executions) when billed monthly, and €20 or €50 a month when billed annually, according to n8n’s pricing page as of November 2026. If you self-host the Community edition, the software is free and you only pay for a server.

The number that decides your bill is executions, and n8n counts them differently from Zapier and Make. Below: every plan, what an execution is, what self-hosting really costs, and three worked examples.

n8n pricing plans at a glance (November 2026)

Plan Monthly billing Annual billing Executions/month Hosting
Starter €24/mo €20/mo 2,500 n8n Cloud
Pro €60/mo €50/mo 10,000 (50,000 tier: €145/mo) n8n Cloud
Business €800/mo €667/mo 40,000 Self-hosted
Enterprise Custom Custom Custom Cloud or self-hosted
Community Free Free No cap Self-hosted

All plans list unlimited active workflows and unlimited users per n8n’s page. Prices are in euros; n8n doesn’t publish separate US-dollar prices.

Bar chart of n8n plan prices billed monthly: Starter €24 for 2,500 executions, Pro €60 for 10,000, Business €800 for 40,000 Source: n8n.io/pricing, checked November 2026.

What each n8n Cloud plan includes

Starter is for one person or a small team getting going: 2,500 executions, 5 concurrent executions, AI Agent and MCP nodes, one shared project, forum support and one day of workflow history. It also includes 1,600 AI Assistant credits a month (the built-in helper that writes and fixes workflows for you).

Pro adds 10,000 executions, more concurrent executions, three shared projects, admin roles, five days of workflow history, seven days of insights and up to 9,600 AI Assistant credits. For most small service businesses running several workflows, this is the plan you’ll end up on.

Business is aimed at larger teams: 40,000 executions, SSO and LDAP sign-in, Git version control, separate environments (dev, staging, production) and queue mode for scaling. As of November 2026 n8n sells it as self-hosted, with Cloud on a waitlist. n8n’s startup program gives companies under 20 employees 50% off Business.

Enterprise adds unlimited projects, log streaming, external secret stores, longer data retention and a support SLA. Pricing is by quote.

n8n Cloud pricing vs self-hosting

n8n Cloud means n8n runs the servers, updates and backups. You pay for the plan and nothing else (apart from AI model usage).

Self-hosting means you install n8n on your own server. There’s no license fee for the Community edition and no execution cap, but you pay for hosting and someone has to maintain it. Our what is n8n guide covers the setup side in more detail.

What self-hosting n8n actually costs

Cost Typical range Notes
Server (VPS) About $5–$25/month 2 vCPU and 2–4 GB RAM is a comfortable start
Database $0–$15/month PostgreSQL on the same server, or a managed one
Backups $1–$5/month Snapshots or off-site database dumps
Domain and HTTPS ~$1/month Free certificates via Let’s Encrypt
Your time 1–3 hours/month Updates, monitoring, fixing breakages

The server range comes from two 2026 hosting comparisons: Livemy’s n8n hosting roundup estimates $6 to $26 a month on a self-managed VPS, and Geekchamp’s guide lists a 2 vCPU, 4 GB Hetzner server at about $7 a month. Check the provider’s own page before you buy, since VPS prices change often.

The hidden cost is time. If your hourly rate is $75 and you spend two hours a month on upkeep, self-hosting “costs” $150 a month, far more than Pro. Self-hosting makes sense when you already have a technical person, need data to stay in your own environment, or run tens of thousands of executions.

n8n installation: what’s involved

If you’re pricing out the self-hosted route, here’s the work: rent a Linux server, install Docker, run n8n with Docker Compose alongside PostgreSQL, put it behind a reverse proxy with HTTPS (Caddy or Nginx), set the encryption key and back it up, and schedule updates. The n8n hosting docs walk through each option. A developer who has done it before needs an hour or two; a first-timer should budget a day.

What counts as an execution in n8n

An execution is one complete run of a workflow, from trigger to last node. n8n’s pricing page says the number of steps and the amount of data don’t change the count.

That’s the biggest pricing difference between n8n and its rivals:

  • Zapier charges one task per successful action step.
  • Make charges one credit per module run (most modules).
  • n8n charges one execution per workflow run.

The same six-step workflow counted three ways: 1 execution in n8n, 5 tasks in Zapier, 6 credits in Make

So a 12-step lead workflow that runs 1,000 times a month uses 1,000 executions on n8n, but up to 11,000 tasks on Zapier (one per action step). If your workflows are long, n8n is usually the cheapest of the three. If they’re two steps each, the gap is small. Our Zapier pricing breakdown shows how tasks add up on that side.

What does and doesn’t use executions

According to n8n’s pricing page (November 2026):

  • Scheduled workflows count every run. A daily schedule is 30 or 31 executions a month; one that runs every 5 minutes is about 8,600 to 8,900, even if it finds nothing to do. This is the most common way people burn through Starter.
  • Each AI agent turn counts as one execution, so a busy chatbot uses executions per message, not per conversation.
  • Manual test runs, sub-workflows and error workflows don’t count. Building and testing in the editor is free, and splitting a big process into sub-workflows doesn’t multiply your bill.
  • Webhooks hit by bots or spam each start a production run. Add a quick check at the top of the workflow and stop early.

Run-time limits also differ: a single execution can run up to 5 minutes on Starter and 40 minutes on Pro before n8n stops it. Waiting time (a Wait node, a human approval) doesn’t count toward that.

Is n8n free?

Yes, in two ways, with limits:

  1. Self-hosted Community edition: free to use and modify for your own business. You pay hosting. Some features (SSO, Git environments, log streaming) need a paid license.
  2. Cloud free trial: n8n’s page says no credit card is needed, and the trial includes Pro features and 1,000 executions. After the trial, you pick a paid plan.

There is no permanent free Cloud plan. If you need “free forever”, self-hosting is the only route.

n8n pricing examples for small businesses

These are worked examples with round numbers, not client cases. Prices are n8n’s November 2026 list prices.

Example 1: a two-truck plumbing company. Workflows: missed-call text-back (300 runs/month), new-lead to CRM (150), review requests after each job (200), daily schedule summary (30). Total about 680 executions. Starter at €24/month (or €20 billed annually) covers it with lots of room.

Example 2: a dental practice with two locations. Appointment reminders (2,400/month), new-patient form processing (200), insurance-document intake with an AI step (300), recall campaigns (1,500). Total about 4,400 executions. That’s over Starter’s 2,500, so Pro at €60/month (€50 annually). The AI step adds model costs, likely a few dollars a month at this volume.

Example 3: a property manager with 600 units. Maintenance request triage with an AI agent (1,800/month), rent reminders (1,800), vendor dispatch (600), owner reports (100), plus one internal sync scheduled every 15 minutes (about 2,900 runs, since every scheduled run counts). Total about 7,200 executions on Pro. If it grew past 10,000, the choices would be Pro’s 50,000-execution tier (€145/month billed monthly) or self-hosting on a small server.

For comparison, the same Example 2 workflows average several actions each and would land in Zapier’s 10,000- or 20,000-task tiers. See the side-by-side in our Zapier vs Make vs n8n comparison.

Costs that aren’t on the n8n pricing page

  • AI model usage. When an AI Agent node calls OpenAI, Anthropic or Google, you pay that provider for tokens with your own API key. For short emails and form messages this is usually cents per hundred runs; for long documents it can be more.
  • Paid apps you connect. Twilio texts, a CRM seat, a document parser: these bill separately.
  • Build and maintenance time. Whoever builds the workflows (you, a staff member or an agency) is the real cost for most small businesses. Templates help: browse our n8n workflow examples for starting points.

How to keep your n8n bill down

  1. Bill annually once you’re sure: €20 vs €24 and €50 vs €60 is about 17% off.
  2. Filter early. Put an IF node right after the trigger so junk submissions stop before doing work.
  3. Batch scheduled jobs. One workflow that processes 50 rows every hour is 24 executions a day; one that triggers per row is 50 or more.
  4. Merge tiny workflows that always run together into one.
  5. Watch the executions list for workflows erroring on repeat, which waste runs.
  6. Consider self-hosting past roughly 20,000 executions a month if you have someone technical.

Is n8n worth the price?

For service businesses with several multi-step workflows, n8n Cloud Pro at €60 a month is usually cheaper than equivalent Zapier usage, and it handles AI agents natively. It’s less worth it if your automations are a handful of simple two-step jobs, where Zapier’s free or entry plan may be enough, or if no one on your team wants to learn it.

If you want the workflows without the learning curve, we build and maintain n8n automations for small businesses; read how an AI automation agency engagement typically works.

Want to see this working for your business?

We build a free demo on your own services and hours.

Get a demo

FAQ

Is n8n free?

The self-hosted Community edition is free to use for your own business under n8n's Sustainable Use License. n8n Cloud is paid, with a free trial that doesn't need a credit card.

How much does n8n cost per month?

As of November 2026, n8n Cloud Starter is €24/month and Pro is €60/month billed monthly, or €20 and €50 per month billed annually. Business is €800/month (€667 annually) and is self-hosted.

Does n8n charge in dollars?

n8n publishes its prices in euros. If you pay in US or Canadian dollars, your card or bank converts the charge, so the exact amount moves a little with exchange rates.

What happens if I go over my n8n execution limit?

n8n's pricing page lists overage on Starter and Pro as capped, so you won't get an open-ended bill, but you should upgrade before you hit the limit. Business and Enterprise keep running and may incur overage charges.

Do test runs count as n8n executions?

No. n8n's pricing page says manual test runs, sub-workflows and error workflows don't count. Production runs of your active workflows do, including scheduled runs that find nothing to process.

opsmigo team

We build AI receptionists and WhatsApp agents for service businesses, and write up what we learn.

LinkedIn →

Stop sending customers to voicemail.

Book a 20-min call

Keep reading

Pricing guide

AI Agent Development Services: What They Cost (2026)

Pricing guide

Answering Service Cost: 2026 Prices From 9 Providers

Pricing guide

AI Chatbot Development Services: Costs and Process (2026)