n8n, Make or Zapier: what to choose for automation

For a simple connection between two apps — a new order in the e-shop writes a row into a spreadsheet, a form sends a notification to Slack — any of the three will do. The difference between them lies in whether you can run the tool on your own infrastructure, and in exactly how you pay for using it. For an ordinary connection of two or three systems, the difference is more about price and about who has access to the data than about features.

Can any of them be hosted on your own infrastructure?

Yes, n8n. The base version is freely available on GitHub under a so-called fair-code licence and can be deployed on your own server — so the data doesn’t leave your infrastructure. The company behind n8n also offers its own cloud and paid enterprise tiers with support, SSO login and other features for teams.

Make and Zapier work only as a cloud service run by the vendor. Self-hosting isn’t possible with them — the data passes through Make’s or Zapier’s infrastructure, regardless of which server in the company the scenario is triggered from.

When the requirement is “data must not leave our environment”, self-hosted n8n is the only one of the three that meets it directly. For more sensitive processes, this often matters more than price.

What exactly do you pay for?

Each of the tools counts usage differently, and that affects when each one pays off:

• n8n, in its cloud version, charges by the number of runs of the whole workflow (executions) — it doesn’t matter how many steps it has inside or how much data it processes. When self-hosted, you don’t limit executions at all; you pay only for the server it runs on.
• Make charges credits for the individual steps of a scenario — each action (for example, writing a row into a spreadsheet) usually uses one credit, more complex AI steps more.
• Zapier charges tasks for each successfully executed automation step, including calls via AI. A free plan with a limited number of tasks per month exists for all three tools.

Exact prices and limits change with the vendors, so we deliberately don’t give them here — check the current price list directly with the vendor before deciding. The principle matters more: with n8n, a workflow with ten steps costs you the same as one with a single step; with Make and Zapier you pay for each step separately. For a process with many steps and high volume, this can make a noticeable difference.

When is an off-the-shelf tool enough, and when do you need a custom solution?

All three cover connecting two systems by a simple rule — “when X comes in, do Y”. Nothing more is needed for that.

You can tell where an off-the-shelf tool stops being enough by whether the process requires decision-making, not just moving data. When text from a document needs to be understood and a decision made about which system it belongs in and in what form, or when an agent has to handle exceptions that can’t be written out as a finite list of rules, a tool like n8n or Make usually becomes just one part of the solution (the trigger or the integration), not the whole solution. What an AI agent is and how it differs from rule-based automation is explained in the article What is an AI agent and how does it differ from a chatbot.

We handle connecting systems via n8n, Make or a custom API as a separate service — system integration from €1,900 (3 days to 6 weeks), depending on the number and complexity of the integrations.

How do you choose?

If a simple connection is enough for you and you don’t want to deal with infrastructure, Zapier has the simplest interface and the most ready-made connections to common apps. Make pays off when you need more complex logic in a scenario — branching, iterations, processing larger batches of data. Choose n8n when you want to keep the data and the running of the automation in your own hands, or when you’re planning a large volume of executions, where paying per step with Make or Zapier can become significantly more expensive.

When extracting data from documents — for example in invoice and document processing — both n8n and Make are commonly used as the link between the OCR/LLM step and the accounting system, but the actual extraction of data from the document is handled by another tool.

Before you decide, it’s worth describing the process step by step first and only then choosing the tool — not the other way round. That’s what we do in a process analysis from €490 (3 days to 1 week).