Skip to main content
Whether an app survey is shown to a given user comes down to a matching trigger plus a few independent eligibility gates. When a trigger fires, the survey appears only if it passes all of the gates. This page builds that model up one layer at a time, from the simplest case to the most advanced. Formbricks decides a survey’s eligibility from three gates, evaluated in this order:
  1. Recontact Options — how often may this survey be shown to the same user?
  2. Cooldown Period — has enough time passed since the user last saw any survey?
  3. Targeting — is the user in the survey’s audience?
A Trigger (a matching action) is the event that then shows an already-eligible survey — it is not a fourth gate, it is what starts a display attempt. The gates are computed up front (on load and whenever the user’s data syncs); the trigger is checked against that result.

​
Level 0 — Always show

Take a survey with no targeting, no Cooldown Period, and Recontact set to “Keep showing while conditions match”. Nothing gates it, so it shows every time its trigger fires. Simple, but usually too aggressive for anything but a persistent feedback button.

​
Level 1 — Recontact Options (per survey)

Recontact Options control how often a single survey may reappear. They count this survey’s own displays and responses, so they are genuinely per survey: Example: an onboarding survey set to Ask until they submit a response keeps appearing on each trigger until the user answers, then stops. See Recontact Options for setup.

​
Level 2 — Cooldown Period between surveys

On top of per-survey recontact, a Cooldown Period enforces a minimum gap between surveys. Two settings feed it, but they share one clock:
  • Workspace Cooldown Period (default 7 days): after a user sees any survey, no survey shows again for the Cooldown Period.
  • Per-survey override (Ignore Cooldown Period / Set custom Cooldown Period): replaces the workspace number for that survey only.
The Cooldown Period is measured from the last time the user saw any survey — not from the last time they saw this one. So a per-survey “wait 90 days” really means “90 days since the user saw any survey,” which frequently-shown surveys keep resetting. The per-survey number changes the value, not the clock. To gate one survey on its own schedule, use interaction-based segments (Level 4).
Example: with the default 7-day Cooldown Period, showing any short survey today blocks every survey — including a quarterly NPS — for the next 7 days. See Cooldown Period for setup.

​
Level 3 — Targeting with Segments

Targeting is part of Attribute-based Targeting, included in the Enterprise Edition.
Targeting restricts who is eligible. A survey can be assigned a Segment built from Attributes, other Segments, and Devices. Only users who match the Segment pass this layer; everyone else never sees the survey, regardless of triggers or the Cooldown Period. Example: a Segment plan = "pro" means only Pro users are eligible for the survey. See Attribute-based Targeting for setup.

​
Level 4 — Interaction-based Segments

Interaction-based segments add a fourth Segment filter type: target users by how they interacted with your surveys — seen, started, or completed within a recent window. Because these look at a specific survey’s own displays and responses, they do what the shared Cooldown Period clock cannot. A filter reads as one sentence:
<operator> <any survey | specific surveys> within <number> <days | weeks | months>

​
Interaction operators

“have started responding to” vs “have not completed” are not opposites. Have started requires the user to have engaged at least once. Have not completed is the negation of have completed, so it also matches everyone who never saw or started the survey — use it as an exclusion, not stand-alone targeting. To target drop-offs, combine both with AND: have started responding to A AND have not completed A.

​
Primary use case: a true per-survey Cooldown Period

This is what interaction segments unlock over the built-in Cooldown Period. To show an NPS survey on its own ~3-month cadence, driven by that survey’s own history rather than the shared cross-survey clock:
  • Recontact: Keep showing while conditions match.
  • Cooldown Period: set a small custom value (e.g. 1 day) — a short guard so it can’t repeat within a session (see the note below). Do not use Ignore Cooldown Period.
  • Targeting Segment: have not seen NPS within 3 months (specific survey = NPS).
The Segment provides the real 3-month per-survey cadence; the small Cooldown Period is only a short-term guard.

​
Other use cases

  • Follow-up funnel: a detail survey targets have completed NPS within 30 days.
  • Re-engage drop-offs: have started responding to Onboarding AND have not completed Onboarding within 14 days.
Interaction membership refreshes on sync, not instantly. Recontact and the Cooldown Period react the moment a survey is shown; interaction-segment membership is recalculated when the user’s data syncs (on load, on identify/setAttributes, and periodically). This lag is irrelevant for long windows like the 3-month cadence above — the short Cooldown Period guard covers it — but don’t rely on interaction filters for instant re-display suppression, and don’t pair one with Show only once, which already blocks re-display permanently.

​
Putting it all together

Still have questions or seeing unexpected behavior? Join us in GitHub Discussions and we’ll be glad to help.