The Difference between Triggers and Filters
When building a segment-triggered campaign, what’s the difference between a trigger and a filter?
Triggers determine WHEN a campaign begins. Think about the trigger as starting the clock for a campaign, since people begin entering your campaign the first time they match your trigger conditions.
Filters decide WHO gets your messages. Think about filters as reflecting the state of a user. These can be evaluated right before sending any message. So filters also work as a confirmation step to make sure the right people stay in your campaign workflow, because things are always changing.
That means, by default, people who stop matching either your trigger or filter segments can exit the campaign (after a grace period), unless you’ve set your campaign exit condition to ensure that “People will move through the entire campaign.”
Filters are evaluated:
- right before entering the workflow
Once a Person has entered the workflow, the Exit Conditions take over. If the campaign cares about continued filtering match (i.e., when your Exit Conditions require it), filters will ALSO get evaluated every time the journey continues from a pause. This includes:
- right after exiting any delay action item in the workflow that actually causes the Person to wait (e.g., a Time Window that a Person does not pause in will not trigger a new evaluation of the filters)
- right after every Attribute Update Action (because those cause a temporary pause while we process the attribute update)
For event-triggered campaigns, People must meet the trigger (perform the event as configured) AND meet the filter conditions before getting a journey for the campaign.
For segment-triggered campaigns, we start a journey for People when they meet the trigger conditions (enter/leave segments) - even if they do not meet the filter conditions. When they do not meet the filter conditions they enter a grace period but their journeys will almost always end with the Person falling out before the first message due to not meeting the filters.
You might be asking, “why can’t I just combine everything into trigger?” Here’s one example of why this isn’t a good idea:
Let’s say you want to send a campaign to people who have viewed the pricing page at least twice within 10 days of signing up. Your goal is to convince them to upgrade with a special “newbie” offer. Here’s how to set it up:
- Your trigger segment is “Signed up”
- Your filter segment is “has viewed pricing page at least once”
- You’d start the workflow with a 10-day delay.
This means: as soon as someone signs up, they enter the campaign and the 10-day clock starts ticking. After 10 days, we’d only send messages if they viewed the pricing page at least two times. After this 10-day post-signup window, nobody will get this newbie offer.
If you had combined both conditions into the trigger, the 10-day campaign clock would not start ticking until people had both signed up AND viewed the pricing page twice. This combination of conditions could happen at any time, no matter when they signed up. For example, this setup could send a user your newbie offer a year later.
Using a filter over a trigger may not matter in cases when there’s no delay at the beginning of your campaign.
Still, we recommend always thinking about triggers and filters separately to help differentiate the condition that triggers a campaign versus the state of the user going through it.
Summing up: triggers define the when for a campaign. Ask yourself, “when do I want the clock to start ticking on this campaign for the user?” Filters define who the campaign goes to. Ask yourself: “who should and shouldn’t be getting this campaign?”
Have any questions about the trigger and filter setup of a particular campaign? Get in touch!