How to come up with Good Test Ideas

The success of A/B testing isn’t just about running experiments — it’s about testing the right things. Strong, well-chosen test ideas help you learn faster, reduce redesign risk, and drive meaningful improvements to your storefront.

This guide explains how to generate high-impact test ideas and includes practical examples you can implement quickly.

Start With a Clear Goal

Before deciding what to test, be clear about what you want to improve. Common goals include increasing conversion rate, improving add-to-cart rate, raising average order value, or reducing bounce rate.

When a goal is well defined, it becomes much easier to evaluate test ideas and avoid changes that add noise without delivering insight.

Use Your Customer Data

The best test ideas are grounded in real shopper behavior. Look for signals in your existing data that suggest friction or hesitation.

This might include session recordings that show confusion, analytics that reveal high-traffic but low-conversion pages, cart abandonment patterns, or recurring questions from support tickets. These insights often point directly to what’s worth testing.

Product Page Test Ideas

Product pages are one of the highest-impact areas to test because they directly influence purchase decisions.

You can experiment with urgency messaging, such as showing low-stock indicators versus no urgency at all. If inventory is genuinely limited, testing scarcity messaging can help shoppers make decisions faster.

Imagery is another common area to test. Some stores see better results with product-only images, while others benefit from lifestyle imagery. Results can also differ between mobile and desktop, making segmentation especially useful here.

Social proof placement is another strong test candidate. Reviews or star ratings shown near the top of the page may reduce uncertainty, especially for higher-priced or more considered purchases.

Product recommendations are also worth testing. Consider whether recommended products perform better when placed near the add-to-cart button, mid-page, or lower on the page. Placement can have a meaningful impact on engagement and conversion.

Homepage Test Ideas

On the homepage, focus on what shoppers see first.

Testing the primary call to action in the hero section is a simple but powerful experiment. For some brands, a direct “Shop All” performs best, while others see stronger results with actions like quizzes, curated collections, or guided experiences.

Featured collections are another area to test. You can compare grid layouts to carousels and test them separately on mobile and desktop. Carousels, in particular, can behave very differently on mobile devices.

Collection Page Test Ideas

Collection pages play a major role in product discovery and browsing behavior.

One common test is the default sort order. You might test best sellers, featured products, or price-based sorting to see what best aligns with shopper intent.

Another idea is to test quick add-to-cart functionality directly on collection pages, rather than forcing shoppers to click through to product pages. This can reduce friction for repeat buyers, especially on mobile, but may not work as well for first-time shoppers who need more context.

Support and Contact Page Test Ideas

Support pages can influence conversion more than many teams realize, especially for higher-consideration purchases.

You can test live chat versus a contact form to see which better supports shoppers when they have questions. Even small changes, like adding reassurance microcopy near a form that sets expectations around response times, can build trust.

Another option is testing how FAQ or Help Center links are displayed. Making it easier for shoppers to find answers on their own can reduce friction and support hesitation before purchase.

Landing Page and Campaign Test Ideas

Landing pages are ideal for more targeted experiments.

Personalized headlines based on location, referral source, or campaign context can help content feel more relevant to visitors.

Exit-intent offers are another common test. You can compare discounts, free shipping, or other incentives to see which resonates most with your audience. Rather than assuming one offer works best, testing helps validate what actually drives conversions.

Make Each Idea Measurable

Every test idea should be refined into a clear hypothesis. You should be able to isolate the change you’re making and tie it to a measurable outcome.

Even tests that don’t win can provide valuable insight. A clear hypothesis ensures you learn something regardless of the result.

Keep a Running Backlog

A/B testing works best as an ongoing process. Maintain a shared list of ideas from analytics, customer feedback, support conversations, and team suggestions.

You don’t need to test everything at once, but keeping ideas organized ensures you’re never stuck wondering what to test next.

Getting Started

If you’re feeling stuck, start small. Pick one page, one goal, and one clear change. Testing even simple ideas consistently will almost always outperform relying on gut instinct alone.