Skip to main content

Session Segments and ARPU Analysis

This guide explains how to create session segments in Google Analytics 4 to compare the impact of Fashion.AI on conversion metrics and ARPU (Average Revenue Per User).

Overview​

Creating session segments allows you to compare the behavior and performance between users who interacted with Fashion.AI recommendations and those who did not, enabling you to measure the real impact of the tool on conversions and cart additions.

Key Metrics​

  • ARPU (Average Revenue Per User): Average revenue per user
  • Conversion Rate: Percentage of sessions that result in a purchase
  • Add to Cart Rate: Percentage of sessions with a cart addition
  • Click-Through Rate: Percentage of views that generate clicks

Segment Structure​

We will create 4 segments for comparative analysis:

  1. Purchase WITH Fashion.AI: Sessions with purchase AND interaction with Fashion.AI
  2. Purchase WITHOUT Fashion.AI: Sessions with purchase BUT without interaction with Fashion.AI
  3. Add to Cart WITH Fashion.AI: Sessions with add to cart AND interaction with Fashion.AI
  4. Add to Cart WITHOUT Fashion.AI: Sessions with add to cart BUT without interaction with Fashion.AI

Creating the Segments​

Segment 1: Purchase WITH Fashion.AI​

This segment identifies users who made a purchase AND interacted with Fashion.AI.

Step by Step​

  1. Access Google Analytics 4

    • Go to Explore > Create a blank exploration

  2. Create the Segment

    • In the Variables tab, click + next to Segments
    • Select Create a custom segment
    • Choose Session segment

  3. Configure the Conditions

    • Name: Purchase WITH Fashion.AI
    • Add a condition group with AND operator between groups:

    Group 1: Had Purchase

    purchase

    Group 2: Interacted with Fashion.AI

    fashionai_click

  4. Save the Segment

    • Click Save and apply

Segment 2: Purchase WITHOUT Fashion.AI​

This segment identifies users who made a purchase BUT did NOT interact with Fashion.AI.

Step by Step​

  1. Create New Segment

    • In the Variables tab, click + next to Segments
    • Select Create a custom segment
    • Choose Session segment

  2. Configure the Conditions

    • Name: Purchase WITHOUT Fashion.AI
    • Add a condition group with AND operator between groups:

    Group 1: Had Purchase

    purchase

    Group 2: DID NOT Interact with Fashion.AI, click on "Add Exclusion Group"

    fashionai_click

  3. Save the Segment

    • Click Save and apply

Segment 3: Add to Cart WITH Fashion.AI​

This segment identifies users who added items to cart AND interacted with Fashion.AI.

Step by Step​

  1. Create New Segment

    • In the Variables tab, click + next to Segments
    • Select Create a custom segment
    • Choose Session segment

  2. Configure the Conditions

    • Name: Add to Cart WITH Fashion.AI
    • Add a condition group with AND operator between groups:

    Group 1: Had Add to Cart

    add_to_cart

    Group 2: Interacted with Fashion.AI (use OR operator between conditions below)

    fashionai_addtocart (similar products and showcases)
    OR
    fashionai_addtocart_shopthelook (shop the look button on product page)

  3. Save the Segment

    • Click Save and apply

Segment 4: Add to Cart WITHOUT Fashion.AI​

This segment identifies users who added items to cart BUT did NOT interact with Fashion.AI.

Step by Step​

  1. Create New Segment

    • In the Variables tab, click + next to Segments
    • Select Create a custom segment
    • Choose Session segment

  2. Configure the Conditions

    • Name: Add to Cart WITHOUT Fashion.AI
    • Add a condition group with AND operator between groups:

    Group 1: Had Add to Cart

    add_to_cart

    Group 2: DID NOT Interact with Fashion.AI, click on "Add Exclusion Group"

    Event name | does not exactly match | fashionai_addtocart
    OR
    Event name | does not exactly match | fashionai_addtocart_shopthelook

  3. Save the Segment

    • Click Save and apply

Setting Up Comparative Analysis​

Add Metrics​

In the Metrics section of your exploration, add:

  1. Sessions: Standard metric
  2. Active Users: To calculate ARPU
  3. Transactions: Number of purchases
  4. Revenue: Total revenue
  5. add_to_cart Events: Total cart additions
  6. Conversion Rate: Conversion percentage

Add Dimensions​

In the Dimensions section, add:

  1. Segment Name: To compare segments
  2. Date: For time-based analysis
  3. Session Source: For channel analysis

Create Comparison Table​

  1. Configure the Table

    • Format: Table
    • Rows: Segment Name
    • Values: All added metrics

  2. Add Calculated Column for ARPU

    • Click Metrics > Add calculated metric
    • Name: ARPU (Average Revenue Per User)
    • Formula: Revenue / Active Users
    • Format: Currency

  3. Add Calculated Column for Add to Cart Rate

    • Click Metrics > Add calculated metric
    • Name: Add to Cart Rate
    • Formula: add_to_cart Events / Sessions
    • Format: Percentage

  4. Add Calculated Column for Click-Through Rate (CTR)

    • Click Metrics > Add calculated metric
    • Name: Fashion.AI CTR (%)
    • Formula: fashionai_click Events / fashionai_view Events
    • Format: Percentage
    • Important: This metric shows how many users who saw the recommendations actually clicked

  5. Add Calculated Column for View Rate

    • Click Metrics > Add calculated metric
    • Name: View Rate (%)
    • Formula: fashionai_view Events / Sessions
    • Format: Percentage
    • Important: This metric shows what percentage of users scroll down to see the recommendations

Interpreting the Results​

Analysis 1: Purchase Comparison​

Compare the segments Purchase WITH Fashion.AI vs Purchase WITHOUT Fashion.AI

MetricPurchase WITH Fashion.AIPurchase WITHOUT Fashion.AIDifference
Sessions5,0008,000-
Users4,2006,500-
Transactions5,0008,000-
Revenue$ 125,000$ 120,000-
ARPU$ 29.76$ 18.46+61%
Average Ticket$ 25.00$ 15.00+67%

Expected Insights:

  • Users who purchase after interacting with Fashion.AI tend to have higher ARPU
  • Higher average ticket indicates that recommendations influence purchase value, suggesting items the customer will actually buy

Analysis 2: Add to Cart Comparison​

Compare the segments Add to Cart WITH Fashion.AI vs Add to Cart WITHOUT Fashion.AI

MetricAdd to Cart WITH Fashion.AIAdd to Cart WITHOUT Fashion.AIDifference
Sessions12,00025,000-
Users10,00020,000-
Add to Cart15,00030,000-
Transactions3,6005,000-
Conversion Rate30%20%+50%
Items per Session1.251.20+4%

Expected Insights:

  • Users who add items to cart with Fashion.AI convert more
  • Higher conversion rate indicates better recommendation quality
  • Fashion.AI increases purchase intent

Analysis 4: Recommendation Engagement (View vs Click)​

This analysis is crucial because recommendations are usually at the bottom of the page.

Important: The fashionai_view event is only triggered when recommendations are visible in the viewport (scroll tracking), not when they are just loaded on the page. This means this event already measures how many users actually saw the recommendations.

Dashboards and Reports​

Dashboard 1: Purchase Overview​

  • ARPU: WITH vs WITHOUT Fashion.AI
  • Transaction volume by segment
  • Comparative average ticket
  • Total revenue by segment

Dashboard 2: Add to Cart Analysis​

  • Add to cart rate: WITH vs WITHOUT Fashion.AI
  • Cart to purchase conversion rate
  • Average items per cart
  • Comparative cart abandonment

Dashboard 3: Conversion Funnel​

Complete funnel visualization:

Sessions
  ↓
Fashion.AI Views
  ↓
Fashion.AI Clicks
  ↓
Add to Cart
  ↓
Purchase

With conversion metrics at each stage for both groups.

Questions? Check the Google Analytics 4 or Google Tag Manager documentation.