Audience Targeting

Audience targeting lets you control which visitors are included in your experiments. When creating an experiment, you can add targeting rules to reach specific segments — like mobile users only, visitors from a particular country, or traffic from a specific campaign.

By targeting effectively, you ensure your experiments reach the right audience and avoid diluting your data with irrelevant traffic.

Targeting rules can also be applied within your experiment reports to filter and segment results. This lets you see how different visitor segments responded to your test variations.

---

Available Targeting Options

Visitor Context

Device Type

Target visitors based on the device they're using.

Available Options	Rules
Desktop, Tablet, Mobile	Is one of, Is not

Visitor Type

Target visitors based on whether they've visited your store before.

Available Options	Rules
New visitors, Returning visitors	Equals, Not Equal

Country

Target visitors from specific countries. All countries worldwide are available.

Rules
Is one of, Is not

---

Traffic Attribution

Traffic Source

Target visitors based on where they came from before arriving at your store.

Available Options	Rules
Direct (typed URL or bookmark), Google (search or ads), Facebook, Instagram, TikTok, Pinterest, and others	Is one of, Is not

Entry Page

Target visitors based on the first page they landed on.

Rules
Equals, Not Equal, Contains, Does not contain, Starts with, Ends with

---

Campaign Tracking (UTM Parameters)

UTM parameters are tags added to your URLs that identify where traffic is coming from. They're commonly used in ads, emails, and social media links. You can target visitors based on any of these parameters.

utm_source

Identifies the platform or website that sent the traffic.

Common values: facebook, tiktok, google, newsletter, affiliate

utm_medium

Identifies the type or category of traffic.

Common values: social, paidsocial, cpc, email, organic, referral

utm_campaign

Identifies the specific marketing campaign.

Common values: black-friday-sale, summer-promo-2026, email-welcome-series

utm_content

Identifies specific ad creative or content variations.

Common values: video-ad, carousel-ad, banner-header, cta-button

All UTM parameters support the same rules: Equals, Not Equal, Contains, Does not contain, Starts with, Ends with

---

Page Targeting

Page URL

Target visitors based on the page URL they're viewing (includes query parameters). Useful for running experiments only on specific pages or sections of your store.

Examples: /products/best-seller, /collections/sale, /pages/landing-page

Rules: Equals, Not Equal, Contains, Does not contain, Starts with, Ends with

---

Combining Multiple Rules

You can combine targeting rules using AND or OR logic:

• AND — All conditions must be true for the visitor to be included

• OR — Any condition can be true for the visitor to be included

Example: Mobile visitors from US or Canada

Device Type is one of "Mobile"
AND
Country is one of "United States, Canada"

Example: Facebook or Instagram traffic

Traffic Source is one of "Facebook"
OR
Traffic Source is one of "Instagram"

Example: Paid traffic on mobile from a specific campaign

Device Type is one of "Mobile"
AND
utm_medium Equals "cpc"
AND
utm_campaign Equals "summer-promo-2026"

---

Rule Operators Reference

Operator	How It Works
Is one of	Matches if the value is any of the selected options (e.g., device is Mobile or Tablet)
Is not	Matches if the value is NOT any of the selected options
Equals	Matches if the value is exactly the specified text
Not Equal	Matches if the value is anything other than the specified text
Contains	Matches if the value includes the specified text anywhere (e.g., URL contains "sale")
Does not contain	Matches if the value does NOT include the specified text
Starts with	Matches if the value begins with the specified text (e.g., URL starts with "/collections")
Ends with	Matches if the value ends with the specified text

---

Tips

• Start broad, then narrow — If you're unsure about targeting, launch the experiment for all visitors first. After collecting data, use the segmentation in your reports to identify which audiences responded best — then create a personalization for that specific segment.

• Don't over-target — The more rules you add, the smaller your audience becomes, which means it takes longer to reach statistical significance. Only add rules that are essential to your hypothesis.

• Use UTM targeting for campaign-specific tests — If you're running a specific ad campaign and want to test a tailored landing experience, target by utm_source or utm_campaign to only include visitors from that campaign.