Funnels Are Not One-Size-Fits-All

At first, you might think all users follow the same funnel, moving step by step toward conversion. But in reality, that’s rarely true. Different users often take different paths, and that’s why segmenting your audience is so important. By breaking down your data into smaller, meaningful groups, you can uncover valuable insights and craft strategies tailored to each group.

---

What is Cohort Analysis?

Cohorts are groups of users who share similar characteristics, like their signup date, traffic source, or even the device they’re using. Cohort analysis helps you spot patterns and behaviors unique to each group, giving you the tools to optimize their experience.

Here are some common ways to segment users:

• Signup date
• Traffic source (e.g., social, organic, paid ads)
• Country or region
• Interaction with specific features or events
• External factors (e.g., time of day, weather)
• Demographics (e.g., age, gender)
• Landing pages
• Device type or operating system

---

Exploring the Google Merchandise Store Funnel

Let’s take the Google Merchandise Store as an example. Ideally, users who begin the checkout process should complete it smoothly without dropping off. However, drop-offs happen, and our job is to figure out where and why. The goal? Make the journey as seamless and fast as possible.

---

Device Matters: Breaking Down the Funnel by Device

The user experience can vary greatly depending on whether someone is using a desktop, mobile, or tablet. Let’s break down the conversion rates by device category to see where improvements can be made.

---

1. Desktop Users

• What We See: Most users access the site on desktop, and overall funnel performance follows desktop trends.
• Conversion Rate: Desktop has the highest final conversion rate, outperforming other devices by 1-4%.
• Takeaway: The desktop experience is solid but can still be improved to push conversion rates even higher.

---

2. Mobile Users

• What We See: Mobile users make up around 10% of total traffic but show lower conversion rates, especially in earlier steps.
• Key Insights:
  • Step 1 → Step 2: The conversion rate here is 17% lower than desktop. Why? Entering detailed information on the Checkout Your Information page can be tedious on a mobile device.
  • Step 2 → Step 3: Conversion drops another 7% compared to desktop. Manually inputting credit card details likely contributes to this.
  • Step 3 → Step 4: Surprisingly, the mobile conversion rate is 13% higher than desktop, with 95.51% of users completing their purchase once they reach the Checkout Confirmation page.
• What to Do:
  • Simplify forms on mobile and use auto-fill where possible.
  • Investigate why mobile users are more likely to complete purchases at the final stage and apply those lessons to earlier steps.
  • Segment mobile users further by operating system (iOS vs. Android) to identify potential differences in behavior.

---

3. Tablet Users

• What We See: Tablet users make up a tiny 0.4% of total traffic, and their conversion rates are the lowest across all steps.
• Key Insights:
  • At every step, about half the tablet users drop off compared to the previous step.
  • The final conversion rate is just 14.29%.
• What to Do:
  • Investigate whether tablet users face specific technical or design issues not present on other devices.
  • Since tablet traffic is so small, improving this funnel may not be a high priority for now.

---

Why Segmentation Matters

When you look at your data as a whole, you’re often left with averages that don’t tell the full story. Segmenting the data by meaningful criteria—like device type—can reveal patterns and insights that would otherwise remain hidden.

---

Beware of Simpson’s Paradox

Segmenting data also helps avoid Simpson’s Paradox, a phenomenon where trends in overall data differ significantly from trends in segmented data. Without breaking things down, you might misinterpret the overall numbers and miss important insights.

---

How to Segment Funnel Data in GA4

1. Use Path Exploration Analysis
   • Navigate to Explore > Path Exploration in GA4.
2. Apply Segmentation
   • Add your segmentation criteria (e.g., device type) using the Segmentation Tool.
3. Analyze the Results
   • Compare the conversion rates across different segments to uncover opportunities for improvement.