Skip to main content
Guardrails are eligibility rules on a treatment that determine which users are allowed to receive it. A user must satisfy all guardrail conditions to be eligible for a given treatment. Guardrails let you control targeting precisely — ensuring the right message reaches the right audience without over-messaging or serving irrelevant content.

​
How Guardrails Work

Guardrails are defined using Data Fields. Each guardrail is a condition expressed as a comparison: When a treatment has multiple guardrail conditions, all conditions must be satisfied for the user to be eligible.

​
Guardrails vs. Goals

Guardrails control who can receive a treatment. Goals define what outcome the treatment is trying to drive.

​
Treatment Selection with Guardrails

When a user makes an API request for treatments, Auxia first filters all Live treatments down to those the user is eligible for (based on guardrails). It then applies its decisioning algorithms to select the optimal treatment from the eligible set. This means guardrails define the candidate pool — Auxia’s algorithms make the final selection within that pool.

​
Frequency Caps and Suppression

In addition to attribute-based guardrails, Auxia also supports:
  • Frequency caps — Limit how often a user can receive a specific treatment (e.g., no more than once every 3 days).
  • Post-interaction suppression — Automatically suppress a treatment for a user after they have clicked or dismissed it (e.g., do not re-show a dismissed treatment for 3 days).
These are configured at the treatment level and work alongside guardrail conditions.
  • Treatments — guardrails are set on individual treatments
  • Data Fields — the data used to evaluate guardrail conditions
  • Journeys — journeys group treatments; guardrails apply per treatment
  • Goals — the outcome that eligible treatments optimize toward
How-to guides: Managing Treatments