How to come up with Good Test Ideas

The success of A/B testing isn’t just about setting up experiments; it’s about testing the right things. Choosing impactful, well-structured test ideas helps you learn faster and unlock meaningful improvements to your storefront.

Here’s a guide to help you generate strong, actionable test ideas that align with your business goals.

Start with a Clear Goal

Before thinking about what to test, ask yourself what you’re trying to improve:

• Are you trying to increase conversions?
• Get more visitors to add products to their cart?
• Improve average order value?
• Reduce bounce rates?

A well-defined goal will help focus your ideas and eliminate noise.

📘 Tip: Review the Test Goals article to understand which metrics you can optimize for in Shogun.

Use Your Customer Data

Strong test ideas are rooted in real shopper behavior. Consider the following data sources:

• Session recordings or heatmaps: Where are people clicking or hesitating?
• Cart abandonment rates: Are customers dropping off on specific product or checkout pages?
• Search or site analytics: Are there high-traffic pages with low conversion?
• Support tickets: Are there usability issues that could be resolved with layout or copy changes?

Look for High-Impact Pages

Focus your testing efforts on pages that matter most:

• High-traffic landing pages
• Product pages with strong potential but weak performance
• Collection pages that help drive navigation and discovery

These areas give you the most data to work with and the fastest path to results.

Identify Friction or Uncertainty

Good A/B tests often resolve a point of friction or uncertainty in the user journey. Ask questions like:

• Is the CTA clear and compelling?
• Are shoppers finding the product details they need?
• Are trust signals (like reviews or guarantees) in the right place?
• Does the page load too slowly or feel overwhelming?

If something appears to be confusing or causing hesitation, it’s worth testing.

Borrow From Proven Patterns

You don’t always have to start from scratch. Test ideas can come from:

• What’s working elsewhere on your site

  (e.g., apply a high-performing layout to a lower-performing page)

• Industry benchmarks

  (e.g., best practices around pricing display or CTA copy)

• Competitor analysis

  (e.g., how similar brands guide their customers through the purchase journey)

Just make sure to test rather than assume; what works for others won’t always work for your audience.

Make It Measurable and Testable

Once you have a test idea, refine it by asking:

• Can we isolate the change and measure its impact?
• Does the test have a clear hypothesis?
• Will the result(s) help inform future decisions, even if it fails?

🧠 Example:

“We believe that moving the ‘Add to Cart’ button higher on the product page will increase cart rate because shoppers won’t have to scroll to take action.”

That’s a testable, measurable idea, and whether it wins or loses, it teaches you something.

Keep a Running List

Testing is a long-term process. Maintain a running backlog of ideas from your team, customers, and data sources. You don’t have to test everything at once, but keeping ideas organized ensures you never lose a good one.

Still Stuck? Try These Prompts

Here are a few starter ideas to explore:

• What’s the first thing a shopper sees? Can it be more compelling?
• Are your calls-to-action clear and benefit-driven?
• Are product descriptions easy to skim?
• Could layout or image changes make the page easier to understand?
• Are you missing key trust signals?