Zapier connects apps through automated workflows called Zaps. Each Zap starts with a trigger, an event in one app, and then runs one or more actions in other apps. Zapier checks for triggers through polling or instant webhooks, and each completed action counts as one task.
For related reading, see Zapier for Nonprofits: 15% Discount, Automation Use Cases, and Finding a Certified Expert.
What Is a Zap in Zapier?
A Zap is an automated workflow that links two or more apps together. Every Zap contains one trigger and at least one action. When the trigger event happens, the Zap runs through each action in order.
For example, a Zap can trigger when a new row appears in a spreadsheet, then send a message in a team chat app. Building a Zap involves three steps. First, you choose the trigger and action apps. Second, you test each step with sample data. Third, you publish the Zap to turn it on. A published Zap only processes new data created after publishing, not data that existed before. Zapier offers integrations for more than 9,000 apps.
What Is a Trigger in Zapier?
A trigger is the event that starts a Zap. Zapier monitors the trigger app for this event, and once it occurs, the Zap runs.
Zapier uses two types of triggers. Examples include:
- Polling triggers, which check an app on a schedule
- Instant triggers, which use webhooks to send data to Zapier in real time
The trigger type depends on the app's API and is fixed for each trigger event. Examples of trigger events include a new row in a spreadsheet, a new lead in a CRM, and a new email in an inbox.
How Do Polling Intervals Work in Zapier?
A polling interval is how often Zapier checks a trigger app's API for new data. Polling intervals range from 1 minute to 15 minutes, depending on your plan.
The table below shows how the two trigger types compare.
| Trigger Type | How It Checks for Data | Speed |
|---|---|---|
| Polling | Zapier asks the app's API on a schedule | 1 to 15 minutes |
| Instant | The app sends a webhook to Zapier | Real time |
The free plan uses a 15-minute polling interval. Professional, Team, and Enterprise plans can adjust the polling interval down to 1 minute. Instant triggers do not use polling. The connected app sends data to Zapier as soon as the trigger event happens.
What Is an Action in Zapier?
An action is a step a Zap performs after the trigger event happens. Examples include sending an email, creating a record, and posting a message.
A Zap can include multiple actions. When the trigger fires, Zapier runs through each action in order. Paths let a Zap perform different sets of actions based on conditions, such as sending different follow-up emails depending on a lead's source. Paths are available on Professional plans and above.
What Is a Task in Zapier?
A task is any successful action that a Zap performs. Zapier counts one task each time an action step completes successfully.
The table below shows which steps in a Zap count as tasks.
| Step Type | Counts as a Task |
|---|---|
| Trigger | No |
| Successful action | Yes, 1 task each |
| Filter | No |
| Path | No |
| Skipped or errored step | No |
For example, a Zap can trigger when a new lead form is submitted, then add the lead to a CRM and send a Slack message to a sales team. This Zap has one trigger and two actions. Each successful run uses 2 tasks. If the Zap runs 50 times in a month, it uses 100 tasks. On the Free plan, which includes 100 tasks per month, this Zap alone would use the entire monthly allowance.
How Do You Calculate Automation Costs in Zapier?
Zapier task usage follows a simple formula. Multiply the number of action steps that run by the number of times the Zap runs.
Paid plans include a higher task limit than the Free plan, and the cost increases as the task limit increases.
A Zap can use tasks faster than expected if it creates a loop. A loop happens when a Zap's action updates the same app and resource that its trigger monitors, causing the trigger to fire again. For example, a Zap that updates a spreadsheet row, and also triggers on updates to that row, can run repeatedly and use a large number of tasks within a short time.
Flood protection is a related safeguard. If a trigger suddenly returns a large batch of new items at once, such as during a bulk import, Zapier may hold the run for review instead of processing every item immediately.
How Does Zapier Handle Errors and Zap History?
When a step in a Zap fails, Zapier may retry the step automatically for certain error types before marking the run as an error. Steps that fail and are not retried successfully do not count as tasks.
Every Zap run appears in Zap History. Each entry shows the data the trigger received, the result of each action step, and the number of tasks the run used. Zap History helps track task usage and find the cause of a failed run.
What Is a Callback URL in Zapier?
A callback URL is a unique web address that Zapier generates for certain triggers and actions. Webhooks by Zapier creates a callback URL for the Catch Hook trigger. An external system can send data to this URL to start a Zap.
Callback URLs also support actions that take time to complete. When an action calls an external service for a long-running task, the service can return a callback URL along with a status of in progress. Once the task finishes, the service sends the result to that callback URL, and Zapier resumes the Zap with the result. Zapier webhooks respond immediately, then run the rest of the Zap separately from that response.
What Other Step Types Can a Zap Include?
A Zap can include step types beyond a trigger and an action. Examples include:
- Filters, which stop a Zap if the data does not match a condition
- Paths, which branch a Zap into different actions based on conditions
- Delay, which pauses a Zap for a set amount of time before the next step
- Formatter, which transforms data such as text, numbers, and dates
- Code by Zapier, which runs custom JavaScript or Python within a Zap
Filters and Paths do not use tasks. Delay, Formatter, and Code steps each count as a task when they run successfully.
Frequently Asked Questions About How Zapier Works
What Is the Difference Between a Trigger and an Action in Zapier?
A trigger is the event that starts a Zap. An action is a step the Zap performs after the trigger occurs.
Does a Zap's Trigger Step Use a Task?
No. Trigger steps never use tasks. Only successful action steps count as tasks.
What Is the Fastest Polling Interval in Zapier?
The fastest polling interval in Zapier is 1 minute, available on Professional, Team, and Enterprise plans.
How Many Tasks Does a 3-Step Zap Use Per Run?
A Zap with 1 trigger and 2 actions uses 2 tasks per successful run, since the trigger does not count as a task.
What Is an Instant Trigger in Zapier?
An instant trigger uses a webhook. The connected app sends new data to Zapier in real time, without polling.
Do Filters and Paths in a Zap Use Tasks?
No. Filters and Paths do not use tasks, even when they run as part of a Zap.
What Happens If a Step in a Zap Fails?
Zapier may retry a failed step automatically for certain errors. If the step still fails, it does not count as a task, and the run appears as an error in Zap History.
What Is a Callback URL Used For in Zapier?
A callback URL starts a Zap from an external system, or lets an external service send a result back to Zapier after a long-running action finishes.
Can a Published Zap Run on Old Data?
No. A published Zap only runs on new data created after it is published, not on data that already existed.
Olaitan Oladipo holds a BSc in Sociology from Olabisi Onabanjo University. He is a self-taught automation builder who has spent years inside n8n doing the work that most tutorials skip: debugging OAuth errors at 2am, migrating client automations from Make.com mid-project, fighting reverse proxy misconfigurations on AWS EC2, and figuring out through trial and error what actually holds up in production versus what only looks clean in a demo.
He is not a developer by training and not a SaaS founder. He is the person in the Discord server who actually answers the question instead of linking to the docs.
His writing on n8n Automation Tutorial covers self-hosting, AI agent workflows, tool comparisons, and the security vulnerabilities the automation industry would rather not discuss. He has built AI-assisted invoice approval flows using OpenAI function calling, connected Claude via HTTP Request nodes, and holds considered opinions about Zapier, Make.com, LangChain, and CrewAI that their marketing teams would not appreciate.
He writes for people who are technical enough to follow a tutorial but experienced enough to want the honest version.