Automation and AI

Trigger

Also called Event, entry condition, starting step

The event that starts an automation — a form submitted, a payment received, a stage changed, a scheduled time reached.

Quick facts: Trigger

Category
Automation and AI
Also called
Event, entry condition, starting step
Level
Beginner
Affects
Who enters an automation, timing, duplicate messages, tag firing
Where to see it
n8n, Make, Zapier, HubSpot workflows, Google Tag Manager
In this article4
  1. How a trigger works
  2. Why triggers matter
  3. Where triggers go wrong
  4. How to act on it

How a trigger works

Nothing in an automated system runs continuously. Each workflow sits idle until something tells it to begin, and that something is the trigger. It carries the event and usually the data attached to it — which form, which contact, which order — so the steps that follow have something to work with.

Triggers come in a few recognisable kinds. Event triggers fire the moment something happens: a form submitted, a payment received, a deal marked won. Schedule triggers fire at a set time, which suits reports and reminders. Condition triggers watch for a record changing into a particular state, such as a contact reaching a new lifecycle stage. Incoming triggers wait for another system to call them, which is what a webhook does.

The word carries a second meaning in tracking. In Google Tag Manager a trigger is the rule that decides when a tag fires — on a page view, a click, a form submission — and the principle is identical: an event, a condition, then an action.

Why triggers matter

The trigger decides who enters an automation and when, which makes it the single riskiest line in the whole build. Every message sent to the wrong person, every duplicate alert, every sequence that started a day too late traces back to a trigger that was too broad, too narrow or watching the wrong thing.

It also decides how the automation behaves when something changes. A trigger tied to a specific form breaks the day someone builds a new form. A trigger tied to a field value survives that, but breaks when the field is renamed. Choosing what to watch is a decision about how fragile the system will be later.

Where triggers go wrong

Repeat firing is the classic. A trigger set on “record updated” fires again every time anyone edits the record, and the contact receives the same welcome message repeatedly. The fix is a condition that only allows a contact through once, or a trigger tied to the change itself rather than to the record.

Loops are the more damaging version: automation A updates a record, which triggers automation B, which updates the record again and re-triggers A. Then there is silent breakage — a renamed field, a form rebuilt, an integration disconnected — after which the trigger simply never fires. No error appears, because nothing happened. Timing errors are the last: firing on a page view rather than a completed submission counts intentions instead of actions.

How to act on it

Pick the narrowest trigger that captures what you actually mean, then add conditions rather than widening it. State in one sentence what should start the workflow, and check the trigger matches those words exactly. Decide deliberately whether a contact may enter more than once, and set the rule rather than discovering it from a complaint.

Test with a real submission from a real device, not with the platform’s preview alone, and watch what the workflow received as well as whether it ran. Then build a way to notice absence: a weekly check of run counts, or an alert when a workflow has not fired in a period where it always used to. Keeping an inventory of triggers is a core part of any workflow automation setup, because the failure you cannot see is the one that costs money.

Do and do not

Do

  • Choose the narrowest trigger, then add conditions
  • Decide whether a contact may enter more than once
  • Alert yourself when a workflow stops firing

Do not

  • Trigger on record updated when you mean record created
  • Let two workflows update fields that trigger each other
  • Trust a preview instead of a real submission

Questions people ask about this

Can one automation have more than one trigger?

In many tools, yes — a workflow can start from several events, such as a form submission or a stage change. It keeps related logic in one place, but it also makes debugging harder, because you must work out which route a contact took. If the follow-up steps genuinely differ by entry point, separate workflows are usually clearer.

Why did my automation run twice for the same person?

Usually because the trigger watches a record rather than an event, so every later edit re-qualifies the contact. Sometimes two workflows share overlapping conditions, or a form was submitted twice. Add an entry condition that excludes contacts who already went through, and check whether another workflow is quietly updating the same record.

Is a Tag Manager trigger the same as an automation trigger?

The idea is the same — an event plus conditions that cause an action — but the scope differs. A Tag Manager trigger decides when a tracking tag fires in the visitor's browser during a single visit. An automation trigger starts a workflow that may run for days and act on records across several systems.

Related terms

Found this useful?

Share it, or ask an AI to summarise it

Back to the glossary

Knowing the term is the easy part

Applying it to your own site and budget is the work. Book a call and I will tell you what actually applies to you.