Short answer
Pick Make if you want a visual builder with nothing to maintain: it is cloud only, the free plan includes 1,000 credits a month, and paid plans start at $9 a month. Pick n8n if you have someone technical: it can run on your own server, it has code steps in JavaScript and Python, and it counts one execution per workflow run however long the workflow is. Both are canvases of connected steps, so the choice is between ease and control.
n8n vs Make at a glance
| n8n | Make | |
|---|---|---|
| Built for | Technical teams that want control | Operations people who want no servers |
| Free option | Community Edition, self-hosted; a trial on the cloud | Free cloud plan: 1,000 credits a month |
| Paid from | €20 a month billed annually, 2,500 executions | $9 a month, 10,000 credits |
| What is counted | Executions: one per complete run of a workflow, whatever the number of steps | Credits: most steps use 1; some AI features use more |
| Hosting | Your own server, or n8n Cloud | Cloud only |
| Builder | A canvas of nodes with branches, merges and loops | A canvas of modules with routers, iterators and aggregators |
| Integrations | 1,000+, plus community nodes and an HTTP request node | 3,000+ apps, plus custom apps for any public API |
| Code and APIs | JavaScript and Python in the Code node | HTTP modules and custom apps for any public API |
| Seats | Unlimited users on every plan | One team below the Teams plan; Teams adds more teams and team roles |
| AI agents | AI Agent and LangChain nodes; local models through Ollama | Make AI Agents, built and debugged in the scenario builder |
| MCP | An MCP trigger turns a workflow into an MCP server | Make MCP server |
| Seeing everything you run | Insights: 7 days on Pro, 30 on Business, 365 on Enterprise | Make Grid, a live map of your automations, on paid plans |
Pricing
n8n
- Community EditionFreeSelf-hosted, so you pay for your own server. Business and Enterprise features need a license key.
- Starter€20/moBilled annually, hosted by n8n. 2,500 executions a month, 5 at a time, 1 shared project.
- Pro€50/moBilled annually, hosted by n8n. 10,000 executions a month, 3 shared projects, 7 days of insights.
- Business€667/moBilled annually, self-hosted only. 40,000 executions, 6 shared projects, SSO.
- EnterpriseCustomHosted or self-hosted. 200+ concurrent executions, log streaming, audit logs.
Checked Nov 11, 2026 on n8n.io
Make
- Free$01,000 credits a month, with no time limit.
- CoreFrom $9/moMake's starting price for 10,000 credits a month. Annual billing is cheaper than monthly.
- ProFrom $16/moStarting price for 10,000 credits a month.
- TeamsFrom $29/moStarting price for 10,000 credits a month.
- EnterpriseCustomAdds Enterprise apps and priority support.
Checked Nov 11, 2026 on make.com
The units pull in opposite directions. A workflow with 20 steps that runs 1,000 times a month is 1,000 n8n executions, inside the €20 Starter plan. In Make it is about 20,000 credits, more than the 10,000 the Core plan starts with. A two-step automation that runs 4,000 times flips it: over n8n Starter's 2,500 executions, and about 8,000 Make credits. Self-hosted n8n has no per-run price at all. Make's starting prices are the ones its own site lists.
Cloud only, or your own server
This is the first fork. Make is a cloud service on every plan. You sign up and build, and Make runs it. Its site lists SOC 2 Type II and ISO 27001 certification.
n8n runs in two places. n8n Cloud is hosted for you, starting at €20 a month billed annually. The Community Edition is free to run on your own server, which keeps the data inside your network and removes the price per run. You take on the updates, backups and uptime. SSO, log streaming and other Business and Enterprise features need a license key, and the Business plan is self-hosted only.
n8n's source is public under the Sustainable Use License. It is not a standard open source license, so read it before you offer n8n to your own customers.
Executions against credits
n8n counts executions. One execution is one complete run of a workflow, and the number of steps does not matter. Starter includes 2,500 a month and Pro 10,000.
Make counts credits. Most steps use one credit each time they run: reading, searching, creating or updating data, and transforming or iterating over it. Some AI features use more. Routers and error handlers use none. The free plan has 1,000 credits a month, and paid plans start at 10,000.
So the cheaper tool depends on the shape of your work. Long workflows that run rarely favor n8n. Short automations that run often favor Make.
Two canvases
Both builders are visual, which sets them apart from a step list.
Make's canvas is the friendlier of the two. Modules connect in a line, a router fans the flow out, an iterator walks through a list and an aggregator gathers the results. People who think in diagrams tend to like it quickly.
n8n's canvas is closer to programming. Nodes pass data as JSON, a Code node runs JavaScript or Python, and an HTTP request node reaches any API. You can do more without leaving the tool, and you need to understand more to start.
Make says it connects 3,000+ apps, and n8n lists 1,000+ integrations plus community nodes.
AI agents and MCP
- n8n builds agents inside workflows with an AI Agent node and LangChain nodes, has built-in evaluations, and can use local models through Ollama. An MCP trigger turns a workflow into an MCP server.
- Make has Make AI Agents that you build, run and debug in the scenario builder. Agents and scenarios can be called through Make's MCP server.
If a model has to run on your own hardware, n8n is the one that allows it.
Teams and oversight
n8n includes unlimited users and workflows on every plan, and limits shared projects: 1 on Starter, 3 on Pro and 6 on Business. Insights go back 7 days on Pro and 365 on Enterprise.
Make has Make Grid on paid plans, which draws a live map of every automation in the organization and how they connect. Extra credits come in bundles of 1,000 or 10,000, and Core, Pro and Teams can buy them automatically.
Which to pick
Pick n8n if
- Your workflows are long: many steps per run.
- The data has to stay on your own server.
- You want code steps, or models that run locally.
- You want unlimited users on every plan.
Pick Make if
- Nobody on the team wants to run a server.
- You want a free cloud plan to start with.
- Your automations are short and run often.
- You want a map of every automation in the organization.
Also consider
- BriefkinWe make thisWhatever runs the automation, its result tends to land in a channel nobody reads. A workflow can post its result as a card on a Briefkin board with one HTTP request, next to your coding agents' sessions, so the weekly report and the agent that is waiting on you sit on the same screen.See how it works
- ZapierThe most apps and the gentlest start. See n8n vs Zapier.
- A scheduled AI agentFor jobs that need judgment, such as a weekly summary, a prompt on a schedule can replace a long workflow. See our Claude Code routines guide.
- More n8n alternativesActivepieces, Windmill, Pipedream and scheduled AI agents, compared in one list.
Questions
Is n8n cheaper than Make?
It depends on the shape of your workflows. n8n counts one execution per workflow run, so long workflows are cheap. Make counts a credit for most steps, and its plans start lower: $9 a month for 10,000 credits against €20 a month for 2,500 executions. Self-hosting n8n's Community Edition costs only the server.
Does n8n have a free plan like Make?
Not on the cloud, where n8n has a trial. The free route is the Community Edition on your own server. Make has a free cloud plan with 1,000 credits a month and no time limit.
Can I self-host Make?
No. Make is a cloud service on every plan. n8n can run on your own server or on n8n Cloud.
Which is better for AI agents?
Both build them. n8n has an AI Agent node, LangChain nodes and support for local models through Ollama, inside a workflow. Make has Make AI Agents that you build and debug in the scenario builder. If the model must run on your own hardware, n8n is the one that allows it.