Targeting

Targeting decides who sees a flow. Build rules with no code - from a simple page match to a nested set of conditions on user traits, events, device, and more. This page covers every condition and every operator.

How it works

On the Targeting tab you build a tree of conditions. A flow shows to a user only if the tree passes. Leave targeting empty and the flow shows to everyone (subject to frequency). A rule that depends on data you have not sent never blocks a flow - if a signal is missing, Onbixo errs toward showing.

Condition types

Page

Match the current page URL. Operators: exact (the URL matches exactly), starts with, contains, and glob (wildcard patterns like /app/*/settings).

Days since signup

Match how long since the user signed up - ideal for first-N-days onboarding. Operators: at most (lte), at least (gte), and exactly (eq). Needs a signup date, which you set by calling identify (or Onbixo uses first-seen as a fallback).

First seen / Last seen

Match on when the user was first or last active - distinct from signup date, which is when they registered. Two modes:

• Relative days: within N days, or more than N days ago. For example, "last seen more than 7 days ago" targets people you want to re-engage; "last seen within 1 day" targets active users.
• On a date: on or after / before a specific calendar date - useful for a fixed cohort, e.g. "first seen before 1 Aug" to reach users who joined earlier.

These read the server-side first/last-seen timestamps from the user's profile. If the profile is unknown (you have not called identify yet), the condition degrades to "show" so a flow is never silently hidden.

Visitor (new or returning)

Match whether this is a new visitor (first time on your site in this browser) or a returning one (been here before). This is browser-local - it uses a first-seen marker stored on the device, so it works even before identify. Ideal for a welcome flow that only shows to first-time visitors.

Session count

Match the user's Nth visit - operators at least, at most, and exactly a session number. A session is a visit after 30+ minutes of inactivity (the standard analytics definition), counted per browser. Use it to hold a flow until, say, someone's 3rd session, once they are past the first look.

Seen / completed a flow

Match on whether the user has interacted with a flow. Operators: seen, not seen, completed, not completed. The flow condition targets another specific flow's state (has seen / has not seen / has completed / has not completed).

User attribute / Company attribute

Match any trait you send with identify - a user trait (plan, role) or a company trait (size, tier). These use the full operator set below.

Event

Match on a behavioral event you send with track. Operators: did / did not (do it at all), count at least (count_gte) / count at most (count_lte), and within days / not within days (how recently). For example: "did created_project at least 1 time within 7 days."

Device / Browser / OS

Match the device type, browser, or operating system. Operators: is, is not, is one of, is not one of. These values are derived on the server, so they cannot be spoofed from the client.

Country / Locale

Match where the user is (country) or their language (locale). Country/device use the server-derived value; locale adds a starts with operator (e.g. all en locales).

Audience

Reference a saved audience by name. Operators: is in / is not in. Combine it with extra inline rules to narrow further.

Operators (for attributes)

Attribute conditions support a rich operator set. What "value" means depends on the operator:

Numeric operators coerce values to numbers; a non-numeric or missing value simply does not block the flow.

AND / OR groups

Combine conditions into groups. A group requires all of its conditions (AND) or any of them (OR), and groups can nest. This lets you express things like:

On the billing page
AND (plan is one of [pro, team]
     OR company size greater than 50)

Preview simulator

The targeting editor includes a simulator: enter sample attributes, events, and device/geo, and it renders a pass/fail result for your whole rule tree - so you can check the logic before publishing.