Zapier Zap Stopped Running
Check whether the Zap is off, on hold, or on and simply not triggering. Those are three different problems with three different fixes.
Open Zap history and filter to the last seven days. What you find there splits this immediately: errored runs, held runs, or no runs at all.
Zapier's statuses do not all mean what they sound like
| Status | Meaning | Turns the Zap off? |
|---|---|---|
| Errored | The run failed | Yes, if it repeats |
| Safely halted | Stopped deliberately, usually a search found nothing | No, and this is not a fault |
| Handled error | A step failed and your error path ran | No |
| On hold | Disconnected app, task limit, or flood protection | No, but nothing is running |
| Scheduled | Errored, and Autoreplay will retry | No |
| (no runs at all) | Never triggered | n/a |
On hold is the status people misread most. The Zap is on, it is not erroring, and it is doing nothing. It stays that way until the underlying condition is cleared, and it will not resume on its own if the cause is a task limit.
Safely halted is not a failure. A search step that found no match halting the run is the designed behaviour, and hundreds of them in a row usually mean your filter is wrong rather than your Zap is broken.
If the Zap was turned off
Zapier turns a Zap off after repeated consecutive errors. The notification goes to the Zap owner, which in an agency or a team that has changed shape is frequently someone who no longer reads that inbox.
Check My Apps for a Reconnect prompt before switching it back on. An expired OAuth connection is the most common underlying cause, and re-enabling without fixing it produces another streak of errors and another shutdown within the hour.
If it is on with no runs
Three causes cover nearly all of these.
The first is deduplication. Zapier remembers which records it has already seen for a trigger, and it will not fire twice on the same one. Testing with a record you used before produces no run and looks exactly like a broken Zap.
Create a genuinely new record to test with. Editing an existing one usually does not count as new either, depending on the trigger.
The second is that Zapier only picks up items created after the Zap was turned on. Turning a Zap on does not backfill anything that happened while it was off, so the work from your outage window has to be handled manually.
The third is the polling interval. Polling triggers check on a schedule that depends on your plan, and on lower tiers that can be up to 15 minutes. A Zap that looks dead for ten minutes may simply not have polled yet.
Check the trigger app, not just Zapier
If the source application changed a field name, deprecated an API version, or revoked the connection, the trigger returns nothing and Zapier has nothing to act on. That produces no runs and no errors, because no run was ever attempted.
Check the source app's own status page and its changelog for API changes around the date the Zap stopped. Matching those dates is often the fastest route to the answer.
Task limits are worth checking early
Hitting your monthly task allowance puts runs on hold rather than erroring them. The banner appears in the Zapier dashboard and is easy to scroll past, and everything looks fine inside the Zap itself.
Why nothing warned you
A Zap that stops triggering produces no runs, no errors and no history entries, because nothing happened to record. Zapier cannot alert on the absence of something it never saw.
The only thing that catches it is an external monitor expecting a regular ping from the Zap's final step, which is what SensaCat does and what the setup in how to monitor Zapier automations describes.