The short answer
n8n vs Make vs Zapier is the wrong fight for most companies of 10 to 100 people, because the three tools overlap more than the comparison posts admit. What separates them is maintenance, data residency and the way each one counts your usage.
Zapier is the fastest to start. Make is the cheapest per step once a workflow gets long. n8n is what you pick when you want to own the system outright.
We work on all three, with a page for each: n8n, Make and Zapier. All three run AI steps now, so the model you want is no longer a reason to prefer one of them.
Price, read on the vendor pages on 12 September 2026
Prices move, so treat these as the shape rather than as a quote. All three came off the vendors own pricing pages the same morning.
- Zapier. Free for 100 tasks a month. Professional starts at $29.99 a month for 750 tasks, or $19.99 billed annually. Team starts at $103.50 a month for 2,000 tasks, or $69 annually.
- Make. Free up to 1,000 credits a month. Core is $9 a month for 10,000 credits, Pro $16 and Teams $29, with around 15 percent off on annual billing.
- n8n. Cloud Starter is $24 a month for 2,500 workflow executions, or $20 billed annually. Pro is $60 a month for 10,000 executions. The self hosted Community Edition is free to run on your own server.
The difference that shows up on the invoice
Zapier counts a task for every successful action. Make counts a credit for each module that runs. n8n counts one execution for the whole workflow, however many steps it contains.
That single rule decides the bill on long workflows. A twelve step process running 2,000 times a month is 24,000 tasks on Zapier and 2,000 executions on n8n. Make lands close to the Zapier figure.
The platform is rarely the expensive part. The expensive part is who fixes the workflow on the morning it breaks.
What the pricing pages leave out
The subscription is the small line. Two costs sit underneath it, and together they usually decide the real bill.
First is rebuild time when a vendor changes an app integration, which lands more often on the managed platforms. Second is the AI usage inside your workflows, billed by the model provider rather than by the automation tool.
A workflow that reads fifty documents a day through a language model can cost more in tokens than the platform subscription does. Ask for that line separately in any quote you receive.
Who keeps it running
Zapier and Make are managed for you. The vendor patches the platform, so the only maintenance you own is the logic you built on top.
n8n cloud works the same way. Self hosted n8n does not: somebody has to update the container, watch the disk and restore a backup when a release goes wrong.
This is where companies of 10 to 100 people choose wrong most often. Self hosting looks free until you price the hour of the person who owns it, and then managed maintenance starts to look like the cheaper answer.
Where your data sits
For a European company this is often the line that decides it, and the three answers are genuinely different.
- n8n. Cloud data is stored in the EU, on servers in Frankfurt. Self hosted, it sits wherever you put it, including a server you already rent.
- Make. You pick a zone when the account is created. The EU zone runs on European infrastructure, and moving an existing account across zones is a support request rather than a setting.
- Zapier. The platform runs in the United States and relies on the EU to US Data Privacy Framework for transfers. There is no European region you can select in the product.
None of that makes Zapier unusable in Europe. It does mean your legal team may want the data processing agreement before you connect anything holding staff or patient records.
When each one is the right call
- Pick Zapier when you need three simple connections working this afternoon and nobody in the company wants to learn a new tool.
- Pick Make when your workflows branch and loop over awkward data. The visual canvas is the one an operations lead can read without training.
- Pick n8n when the logic is really software, when the data cannot leave Europe, or when the monthly bill on the other two has passed the cost of a server.
One rule cuts through the rest. If an automation touches more than two systems and runs thousands of times a month, per step pricing gets expensive faster than anyone expects. Run the numbers on your busiest workflow before you sign anything annual, because that is the one that sets your tier.
When to move, and when to stay
Migrating platforms costs weeks and buys nothing by itself. Move when the bill has doubled without the work doubling, or when a compliance answer forces it.
Staying put and rewriting three bloated workflows often saves more than the migration would. That is the boring answer, so nobody sells it.
The question that settles it faster than a table
Ask who in your company would open the tool six months from now to change one condition. If the honest answer is nobody, pick the managed platform with the friendliest canvas, whatever the per step price says.
Where somebody technical enjoys this kind of work, n8n pays for itself quickly and leaves you a system nobody can take away. That one answer predicts satisfaction better than any feature list.
What we would say on a call
The platform question usually arrives dressed as a strategy question, and it is not one. Pick the tool your team can maintain, then spend the argument you saved on choosing the right process.
Want the ranked list for your own company? That is what the audit produces. Already know the process, and the systems page explains how a build runs.
Five things you can do tomorrow.
- Count the steps in your longest workflow, then multiply by how many times it runs a month.
- Open your current bill and check whether you are paying per action or per run.
- Ask your legal contact whether any connected system holds data that cannot leave Europe.
- Name the person who would patch a self hosted server, and price their hour honestly.
- Before migrating anything, rewrite your three heaviest workflows and measure again.
