
Zapier counts a task each time an action in a Zap completes, so a Zap with 10 action steps that runs 10,000 times a month uses 100,000 tasks. n8n counts one execution per run of the whole workflow, so the same workflow is 10,000 executions, and you can host n8n on your own servers. Both definitions are on Zapier's pricing page and n8n's pricing page. Your team has outgrown Zapier when the task bill grows faster than the work it automates, when workflow data has to stay in your own cloud or in the EU, or when your workflows need more code, branching or agent steps than Zapier's limits allow.
For short Zaps that people outside engineering build and own, Zapier stays the quicker tool. Zapier runs every Zap on its own servers, so there is nothing to host, and every paid plan runs multi-step Zaps. The "Zapier tax" is what the per-task model costs once a Zap with a long chain of action steps runs at volume.
Zapier's pricing page says "A task is counted whenever Zapier successfully completes a unit of work for you." Each action step in a Zap that completes is a task, and "Failed actions are not counted." Triggers, polling for new data and Zapier's built-in tools, such as Filter, Paths, Formatter and Delay, count no task, while some actions count more than one, such as a Zapier MCP tool call, which counts 2. A Zap that enriches a lead, updates the CRM record, posts to Slack and writes a row to a database uses four tasks every time it runs.
On Zapier's pricing page on 20 November 2026, the Professional plan starts at $19.99 a month billed annually for 750 tasks, and the Team plan at $69.00 a month for 2,000 tasks. 100,000 tasks a month costs $489.00 on Professional or $599.00 on Team, billed annually, and above 2,000,000 tasks a month the price goes through Zapier's sales team. With pay-per-task turned on, tasks over your plan bill at 1.25 times the base rate on annual billing and 2.5 times on monthly billing, and your Zaps stop at three times your task subscription until the next cycle. With pay-per-task off, they stop at the limit.
n8n prices a different unit. Its pricing page says "An execution is a single run of your entire workflow. It doesn't matter how many steps are in the workflow or how much data it processes." On that page on 20 November 2026, the Starter plan on n8n Cloud is 20 euros a month billed annually for 2,500 executions, Pro is 50 euros a month billed annually for 10,000 executions, and the self-hosted Business plan is 667 euros a month billed annually for 40,000 executions. Only production runs count, according to n8n's documentation, so manual test runs, sub-workflows called from a parent workflow and error workflows count no execution. Self-hosted without a license key, n8n runs as the Community edition, "the free edition", with no execution fee.
Take one workflow with 10 action steps that runs 10,000 times a month. On Zapier that is 100,000 tasks, $489.00 a month on Professional billed annually. On n8n Cloud it is 10,000 executions, inside the Pro plan at 50 euros a month billed annually. On a self-hosted Community edition it carries no execution fee, and you pay for the server instead. Make bills per step as well. Its pricing page says "Every time a module in a scenario performs an action, it consumes credits."
Paths, Zapier's branching tool, allows "up to 10 path branches within the same path group", and "A nested Paths step lets you create up to 3 levels of path groups within a Zap." Code by Zapier lets a Zap "run small snippets of Python or JavaScript", with 30 seconds of runtime per step on Professional and Team, 2 minutes on Enterprise, and "an I/O limit of 6 MB". A workflow that needs more branches, a longer script or a larger payload has to be split across several Zaps or moved off Zapier.
In n8n, the Switch node "supports multiple output routes", and the Merge node combines "data from multiple streams, once data of all streams is available", so branches rejoin inside the same workflow. The Code node runs "custom JavaScript or Python" as a step in the workflow. Self-hosted, it can import external npm modules; on n8n Cloud it cannot.
Zapier's data privacy page says "Zapier hosts data in AWS servers located in the United States". To the question "Is there an option to have my data stored only within the EU?" it answers "Zapier does not support this option." Zapier holds SOC 2 Type 2 and SOC 3 reports, and its Team and Enterprise plans add SAML single sign-on and an audit log kept for 6 months on Team and 1 year on Enterprise.
n8n's pricing page says n8n Cloud stores data "within the EU, on servers located in Frankfurt, Germany", and that for self-hosted plans the data sits "wherever you decide to host n8n". Self-hosted, the Business plan adds SSO and LDAP sign-in, environments and version control with Git, and Enterprise adds an external secret store, log streaming and audit logging. A team whose security review requires workflow data, credentials and execution logs to stay in its own cloud account can meet that with self-hosted n8n, and cannot with Zapier.
| Zapier | n8n | |
|---|---|---|
| Unit | a task each time an action in a Zap completes | one execution per run of the whole workflow |
| Entry plans | $19.99 a month billed annually for 750 tasks | 20 euros a month billed annually for 2,500 executions |
| Code | Python or JavaScript as a step in a Zap | custom JavaScript or Python |
| Branching | up to 10 path branches within the same path group | supports multiple output routes |
| Data | AWS servers located in the United States | wherever you decide to host n8n |
| Sign-on | SAML single sign-on | SSO and LDAP sign-in |
| Agents | agents as a separate product | a step inside the workflow |
Self-hosting moves the running of the platform to your team, and n8n's guide to choosing how to use n8n lists maintenance of a self-hosted instance as "Your responsibility". For production volume, n8n runs in queue mode, where "a main instance receives triggers and webhooks, Redis queues the executions, and you scale by adding or removing workers". Queue mode is included in the free Community edition, and n8n recommends PostgreSQL for production. Multi-main mode, which runs more than one main process for high availability, needs the Enterprise plan. On every instance Ovidius builds or takes over, n8n runs in Docker in your own cloud account, with Postgres and Redis workers.
A move from Zapier does not need a single cutover day. List every Zap first, with its trigger, its steps and the systems it writes to; for a large Zapier setup, an AI Audit produces that list before anyone quotes the work. Rebuild the highest-value workflows in n8n first, with error handling, while the Zaps keep running. Run both side by side until the outputs match. Cut over once the outputs match, and switch off the Zap.
The parallel run matters most for workflows that feed other systems, such as CRM updates, database writes and billing triggers, where a failed step changes records downstream. It also gives the people who will own the workflows time to learn the n8n editor before it becomes their main tool.
Zapier sells agents as a separate product. Zapier Agents "connect to your business data and perform tasks across 9,000+ apps", and their activities do not draw from your Zaps' task allowance. In n8n, the AI Agent node is a step inside the workflow: "Connect a chat model and one or more tools, and the agent decides which tools to call to complete a task." The steps around it can read the record before the agent runs and write the result after, and n8n lets you "require human approval before an AI Agent executes a specific tool", with the approval request sent to Slack, email or n8n's own chat. For an agent that updates a CRM or messages a customer, that approval is where a person signs off before anything changes.
Move a workflow off Zapier when one of these holds for it:
One signal on one workflow is reason enough to move that workflow alone. Zapier can keep the short Zaps your teams build for themselves while n8n runs the high-volume and regulated ones. Ovidius AI is an official n8n Select Partner. We rebuild your zaps in n8n, add error handling, and run old and new side by side until the outputs match. How long that takes depends on how many workflows there are and how they depend on each other, so we tell you before you commit.
It is the way a Zapier bill grows with the number of steps in a workflow, not only with the number of runs. Zapier counts a task each time an action in a Zap completes, according to its pricing page, so a Zap with 10 action steps that runs 10,000 times a month uses 100,000 tasks. n8n counts one execution per run, according to its pricing page, so the same workflow is 10,000 executions.
No. Zapier's pricing page says triggers, polling for new data and its built-in tools, including Filter, Paths, Formatter and Delay, count no task, and failed actions are not counted either. Each action that completes counts one task, and some actions count more, such as a Zapier MCP tool call, which counts 2.
It depends on how many steps each run has. The entry plans cost about the same: $19.99 a month for 750 tasks on Zapier and 20 euros a month for 2,500 executions on n8n, both billed annually, on the two pricing pages on 20 November 2026. For 10 action steps run 10,000 times a month, Zapier's pricing page lists 100,000 tasks at $489.00 a month on Professional, and n8n's pricing page covers 10,000 executions with Pro at 50 euros a month.
Yes. n8n's pricing page says n8n Cloud stores data within the EU, on servers in Frankfurt, Germany, and self-hosted n8n keeps it wherever you host it, including your own cloud account. Zapier's data privacy page says Zapier hosts data on AWS servers in the United States and does not offer storage only within the EU.
Yes. Code by Zapier runs small snippets of Python or JavaScript as a step in a Zap, with 30 seconds of runtime per step on Professional and Team, 2 minutes on Enterprise and a 6 MB limit on input and output, according to Zapier's pricing page and Help Center. n8n's Code node runs JavaScript or Python too, and a self-hosted instance can import external npm modules.
No. List every Zap, rebuild the highest-value workflows in n8n first with error handling, and run both side by side until the outputs match before you switch a Zap off. Short Zaps that your teams build for themselves can stay on Zapier while n8n runs the high-volume and regulated workflows.
Ready to move off Zapier?
Plan the migration with us. The Ovidius n8n page sets out how we rebuild zaps in n8n, set up self-hosted n8n in your own cloud account and keep it running after launch.
Start with an audit. If you are not yet sure which workflows to move, an AI Audit maps how data moves between your systems and ends in a ranked roadmap of what to build first.