n8n
The only major automation platform you can fully self-host with unlimited executions, combining a visual node editor with embedded code and AI nodes — the developer's choice when Zapier's per-task pricing or cloud-only model becomes a problem. Essential in this category.
The take
What is n8n?
n8n is a source-available workflow automation platform you can self-host or run in the cloud. It connects apps, APIs, and AI models through a visual node-based editor, with the option to write custom code inside any workflow. n8n raised a $60 million Series B in 2025 as demand for self-hosted automation grew, according to the company.
The product covers 400+ integration nodes, code execution, and native AI agent nodes. It sits as a core pick in the 80/20 of automation tools we cover. n8n is the developer’s answer to Zapier: when per-task pricing or a cloud-only model becomes a problem, n8n’s self-hosting and execution-based billing solve it.
How does n8n work?
n8n works by connecting nodes on a visual canvas, where each node represents a trigger, an app action, a transformation, or a piece of code. Data flows from one node to the next, and you can branch, merge, and loop. You run it on n8n Cloud or self-host it with a single Docker command.
The node-based editor
The editor is a canvas where you drag nodes and wire them together. A trigger node starts the workflow — a webhook, a schedule, a new row in a database. Action nodes then do work: send a Slack message, update a record, call an API. Branching nodes route data based on conditions. The visual model makes complex multi-step logic readable.
Code and HTTP nodes
The code node lets developers write JavaScript or Python directly inside a workflow, transforming data however they need without leaving the editor. The generic HTTP node calls any REST API, so even apps without a dedicated n8n integration are reachable. Together these two nodes mean n8n can connect to effectively anything — the escape hatch that pure no-code tools lack.
Self-hosting and execution pricing
n8n’s defining trait is self-hosting: run it on your own server and get unlimited executions with full data control at no per-run cost. n8n Cloud, the managed option, charges per workflow execution rather than per step — a single run with 50 actions counts as one execution. This pricing model is dramatically cheaper than per-task billing at high volume.
How does n8n compare to Zapier and Make?
n8n leads on self-hosting, code flexibility, and cost at scale. Zapier leads on ease of use and template breadth for non-technical users. Make leads on visual polish without requiring self-hosting. The table below shows the trade-offs across the three.
| Attribute | n8n | Zapier | Make |
|---|---|---|---|
| Self-hosting | Yes (free, unlimited) | No | No |
| Pricing model | Per execution | Per task | Per operation |
| Code nodes | JavaScript + Python | Limited | Limited |
| Native AI nodes | Strong | Growing | Growing |
| Ease of use | Developer-focused | Easiest | Visual, medium |
| Integration count | 400+ nodes | 7,000+ apps | 1,500+ apps |
| 80/20 verdict | Pick for control + scale | Pick for non-devs | Pick for visual no-code |
“n8n is the automation tool I reach for the moment a workflow needs custom code or data has to stay on our own servers — Zapier is friendlier, but it can’t self-host and the per-task bill climbs fast at volume,” said Devon Park, Developer Tools Editor at tools8020.
Who uses n8n in 2026?
Developers and DevOps teams use n8n to self-host automations that keep sensitive data on internal infrastructure, avoiding sending records through a third-party cloud. Growth and operations teams use it to run high-volume workflows that would cost far more on Zapier’s per-task pricing. AI engineers use the native AI nodes to orchestrate LLM agents and retrieval pipelines.
The common profile is a team with at least some technical capacity that values control and cost over maximum ease of use. Companies in regulated industries — healthcare, finance — choose self-hosted n8n specifically because data never leaves their environment.
When should you skip n8n?
n8n is the wrong choice in three scenarios. Use the listed alternative instead.
- Your team is entirely non-technical. Use Zapier, whose interface and template library are built for users who never want to see code. n8n assumes some developer comfort.
- You want maximum integration breadth out of the box. Zapier connects 7,000+ apps versus n8n’s 400+ nodes. For obscure SaaS tools, Zapier likely has a ready-made connector.
- You want visual no-code without self-hosting. Use Make, which offers a polished visual builder and cloud hosting without n8n’s developer-oriented learning curve.
How much does n8n cost?
Self-hosting n8n is free with unlimited executions — you pay only for your server. The cloud Starter plan at $24 per month suits teams that prefer managed hosting. Pricing is per execution, not per step, which keeps high-volume workflows affordable.
| Plan | Price | Key inclusions |
|---|---|---|
| Self-hosted | $0 | Unlimited executions, full data control, you manage infra |
| Starter (cloud) | $24/month | 2,500 executions, managed hosting |
| Pro (cloud) | Higher tier | More executions, concurrency, admin features |
| Enterprise | Custom | SSO, scaling support, dedicated infrastructure |
Pricing verified at n8n.io/pricing on 2026-05-26. Cloud execution allowances change periodically — confirm the current execution count per tier before committing, and weigh self-hosting if your volume is high.
How we evaluated n8n
This review draws on Devon Park’s background reviewing developer tools and infrastructure, plus hands-on testing of n8n self-hosted via Docker and on n8n Cloud, building real workflows with code nodes and AI agents and benchmarking cost against Zapier. We re-verify pricing every 90 days.
See our evaluation methodology for the full criteria. n8n is a core pick in the 80/20 of automation tools, alongside Make for teams that want visual no-code without self-hosting.
Strengths & trade-offs
What earns the score
- Self-hosting means unlimited executions and full data control at no per-run cost
- Execution-based cloud pricing is far cheaper than Zapier at high volume
- Code nodes let developers drop into JavaScript or Python without leaving the flow
- 400+ nodes plus a generic HTTP node connect to effectively any API
- Strong native AI nodes for building agents and LLM pipelines
Where it falls short
- Steeper learning curve than Zapier — built for developers, not non-technical users
- Self-hosting requires you to manage updates, scaling, and security yourself
- Fewer polished pre-built templates than Zapier's app directory
- The fair-code license restricts reselling n8n as a hosted service
- Cloud plan support is lighter than enterprise-grade competitors
How it compares
| Tool | Score | Tier | From |
|---|---|---|---|
| 92 | Essential | $24/user | |
 Zapier | 91 | Essential | $19.99/user |
 Make | 68 | Situational | $9/user |
Frequently asked questions
How much does n8n cost?
n8n is free to self-host with unlimited executions. The cloud Starter plan begins at $24 per month for 2,500 executions, billed per workflow run rather than per step. Higher cloud tiers add more executions and concurrency. Self-hosting costs only your own server. Pricing verified at n8n.io on 2026-05-26.
How does n8n compare to Zapier?
n8n is built for developers and can be self-hosted; Zapier is built for non-technical users and is cloud-only. n8n charges per execution while Zapier charges per task, making n8n far cheaper at high volume. Zapier wins on ease of use and template breadth. n8n wins on cost, control, and code flexibility.
Is n8n really free?
Self-hosting n8n is free with unlimited workflow executions — you pay only for the server you run it on. The source is available under a fair-code license. The paid plans are for n8n Cloud, the managed hosting option, which removes the need to maintain your own infrastructure. The free self-hosted version is fully functional.
How does n8n compare to Make?
Both use visual node-based editors and price by operations or executions, making them cheaper than Zapier at scale. n8n adds self-hosting and code nodes that Make lacks, appealing to developers. Make has a more polished interface and more pre-built templates. Choose n8n for control and code, Make for visual polish without self-hosting.
Can n8n build AI agents?
Yes. n8n includes native AI nodes for LLM calls, agents, vector stores, and chains. You can connect models from OpenAI, Anthropic, and others, give them tools, and chain them into multi-step workflows. This makes n8n a practical orchestration layer for AI automation without writing a full application from scratch.
Is self-hosting n8n hard?
Self-hosting n8n is straightforward for developers — a single Docker command gets it running. The ongoing work is what takes effort: updates, scaling, database backups, and security patching are your responsibility. Teams without DevOps capacity often start on n8n Cloud and migrate to self-hosting once volume justifies managing infrastructure.