n8n bills per workflow run, Zapier per action step. The difference determines which plan actually fits your automation.
When you compare n8n and Zapier, you inevitably encounter two different billing logics: n8n counts Executions, meaning complete workflow runs, while Zapier counts Tasks, meaning individual successful action steps within a Zap. According to n8n.io/pricing an Execution is "a single run of your entire workflow" (a single run of the entire workflow), regardless of how many steps it contains or how much data is processed. Zapier, on the other hand, according to its own pricing page, counts every successfully completed action step individually as a Task, while the trigger itself is not counted. This difference is decisive for the plan choice, because a multi-step workflow can remain mathematically cheaper with n8n than with Zapier once many actions come together in one run. As of: July 2026.
According to the definition on n8n.io/pricing every complete run of a workflow is exactly one Execution, regardless of the number of nodes, branches, or the amount of data processed. A workflow with three steps and a workflow with fifteen steps therefore each cause only one Execution for the same trigger, as long as they are counted through in one run. The plans are accordingly tiered by volume rather than by complexity: according to the pricing page, the Starter plan includes 2,500 Executions per month, the Pro plan 10,000, and the Business plan 40,000, while Enterprise customers agree on an individual quota. What the pricing page does not detail is the exact handling of sub-workflows or failed runs, which is worth checking directly with n8n for more complex automations on a case by case basis.
Zapier bills according to a different principle. Tasks are counted, according to zapier.com/pricing "when a Zap successfully moves data or completes an action", the trigger itself is explicitly not counted. Every further step and every external connector call within a Zap, on the other hand, consumes one Task, whereby the exact number can vary depending on the AI model tier, code runtime, and connector type. Some built in tools are excluded from Task counting, including Zapier Tables, Zapier Forms, Filter by Zapier, Formatter by Zapier, Path by Zapier, Delay, Looping, Sub-Zap, Digest, Zapier Manager, and Storage. If an action fails, it is also not counted.
Take a simple workflow that sends an email, updates a CRM record, and posts a message in a team channel when a new form is submitted. In n8n, this complete run, whether it contains three or eight steps, is a single Execution. If this workflow runs 1,000 times a month, it consumes 1,000 Executions, regardless of the internal complexity. In Zapier, on the other hand, each of the three successful action steps is counted individually, while the trigger itself is left out. With the same structure, the same 1,000 runs could mathematically become up to 3,000 Tasks, provided none of the excluded functions are used. How large the actual difference turns out to be depends strongly on the individual workflow, in particular on how many action steps occur per run and whether AI steps or certain connector types additionally increase Task consumption.
The plain list price of a plan says little about how quickly you will actually exhaust it. What matters is the structure of your automations: how many action steps a typical workflow has, and how often it runs per month. For short, single step automations, the difference between Execution and Task billing may turn out to be small. For multi-step workflows with many actions per run, it is worth calculating in advance, before choosing a plan, how many Executions or Tasks your actual use case causes per month, rather than relying solely on the number in the advertised price. We at NordFlux do exactly this calculation together with you before a plan decision is made, so that you experience no surprise on your bill at the end of the month. More about our approach can be found under n8n consulting from NordFlux.
A complete run of a workflow from start to finish, regardless of the number of steps or the amount of data. This is stated on n8n.io/pricing and is the central difference from the Task logic of other tools.
No. According to Zapier's own pricing page, triggers explicitly do not count toward the Task limit, only the successful action steps that follow are counted.
Among others, Zapier Tables, Zapier Forms, Filter by Zapier, Formatter by Zapier, Path by Zapier, Delay, Looping, Sub-Zap, Digest, Zapier Manager, and Storage are excluded from Task counting according to Zapier.
In n8n, generally not, since only the complete run counts. In Zapier, yes, because there every successful action step is counted individually as a Task, with the exception of the functions mentioned above.
Founder of NordFlux. Spent four years automating processes at enterprise scale at Dräger, and now brings that depth to the mid-market — pragmatic and with full data sovereignty.
Certifications
You recognize stuck executions in n8n by status indicators that won't move. Here's how to set EXECUTIONS_TIMEOUT correctly and find the cause.
Why the n8n database grows through executions and how EXECUTIONS_DATA_PRUNE, retention period, and limits automatically limit the amount of data.
Installing n8n on Hetzner: server type, Docker Compose with Caddy, and why the German server location matters for GDPR-compliant automation.
n8n counts one execution per workflow run, while tools like Zapier count every single action step as a task, and getting this wrong quickly leads to a plan that's either too small or too expensive. NordFlux calculates your actual automation volume and finds the plan size that fits your real workflow load.
