A/B Testing Overview
A/B Testing in Account Editor lets you test two versions of an upsell offer — comparing elements like discounts, layouts, or copy — so you can identify what actually improves conversion and revenue performance.
Instead of guessing what works best, A/B Testing gives you data-backed answers to refine your upsell strategy continuously.
🧠 Why A/B Testing Matters
A/B testing helps you:
- Validate new offers before rolling them out widely.
- Improve conversion rates by testing small layout or text changes.
- Discover which surfaces (Checkout, Thank You, After Checkout, or Order Status) generate the highest lift.
- Build a habit of data-driven optimization rather than assumptions.
⚙️ How It Works
When you create a new A/B test, Account Editor randomly splits your customers into two equal groups:
- Control Group (A): Sees your current offer setup.
- Variant Group (B): Sees the new version you’re testing.
The app tracks both versions automatically and compares performance metrics like:
- Revenue
- Conversion Rate
- Click-Through Rate
- Average Order Value (AOV)
After enough data is collected, the dashboard shows which version performs better — marking it as the Winner.
📍 Accessing A/B Testing
- Go to your Account Editor dashboard.
- Navigate to the Upsell Engine → A/B Testing tab.
- You’ll land on the A/B Testing Dashboard, which summarizes all active and past tests.
📊 A/B Testing Dashboard Overview
The dashboard helps you manage your tests at a glance. You’ll see:
Section | Description |
|---|---|
Active Tests | Number of currently running tests. |
Completed | Total number of finalized or stopped tests. |
Upsell Added | How many upsell items were added through your tests. |
Revenue | The total revenue influenced by your A/B tests. |
If you haven’t created any tests yet, the page will show:
“No A/B tests yet – Start testing different variations of your offers to optimize performance.”
Click “Create your first test” to begin.
🧩 Best Practices Before Creating a Test
- Always test one variable at a time (e.g., discount %, layout, or button text).
- Let your test run for at least 2 weeks to gather valid results.
- Avoid editing the same offer while it’s being tested.
- Pick a clear metric to measure success — such as conversion rate or revenue.
- Choose a surface with enough traffic to collect meaningful data.
✅ Example Use Cases
Scenario | What You’re Testing |
|---|---|
Compare two discount levels | “15% off” vs. “20% off” |
Test CTA buttons | “Add to Cart” vs. “Get This Deal” |
Test layouts | Single product block vs. bundle layout |
Test timing | Checkout vs. After Checkout surface |
💡 Tip:
A/B Testing doesn’t just tell you which version wins — it helps you understand why.
Look at click-through and add-to-cart differences between variants to uncover behavioral trends.
🧾 Next Article → Creating a New A/B Test
In the next guide, we’ll go step-by-step through:
- Setting up your test name and hypothesis
- Choosing success metrics
- Selecting surfaces and offers to test
- Understanding how traffic is split
This will help you confidently create your first A/B Test in Account Editor.
Updated on: 27/11/2025
Thank you!