When you launch your app, you want everything to appeal to your users as much as possible. Every screenshot should perfectly display your app’s features and points of pride in a polished and visually engaging manner – but how can you be sure your creative set does that? In order to ensure your screenshots are as effective as possible, you’ll want to A/B test them. Here are some important points to remember when running your A/B test.

Research Trends

 Before you begin designing your screenshots, you should have an understanding of what creative elements appeal best to users. This requires research.

Look at your top competitors and examine their screenshots. What aspects are they focusing on? How are they framing their screenshots? For mobile games, do they include more characters, cutscenes or gameplay? Analyzing their screenshots for common attributes helps you find what styles or imagery work best.

Once you understand what aspects of a screenshot perform well, you can begin designing your creatives accordingly. Utilize the design styles and choices that you see to test them for your own creatives and determine if they’ll work well for you. This is where your testing can begin.

Know What to Test

 When you A/B test screenshots, you can’t simply toss multiple elements into a test and see what performs best overall. The idea of the test is to determine which individual aspects work best.

 You can test which color choices perform best, or what messaging appeals to your core audience. You can test if a screenshot performs better when it showcases characters facing the screen or shows them in action. The important thing is that you test incrementally.

Incremental testing allows you to have more control over the A/B tests. Each change you test is another step in the right direction, providing more information on what will result in the most conversions and the best results.

Each A/B test should experiment with a different aspect, whether it’s the callout text, designs or color choices. Changing more than one thing at a time will make it difficult to tell what changes are driving the improvements.

For instance, if an app tests a screenshot variant that changes both the image shown on the phone and different text, it will not be clear which part of the variant is driving the changes in conversions. It could be that the callout text highlights a feature that users find more interesting, or the image on the phone could be more visually appealing.

Instead, running two separate tests makes it clear which change is responsible for the improvements. It could be that the image drives more conversions, but the changed text was holding it back, or the callout text works best with the initial image. Without running the tests separately, it will be hard to tell.

Touch Nothing but the Test

 As you’ll be testing individual elements, it’s important to only change one thing at a time during the tests. As with any experiment, it’s important to have a “control group” – the base creative you’ll be testing the changes against.

When you run your tests, you’ll want to ensure everything outside of the test stays the same. In addition to maintaining all other aspects of the screenshots, the metadata should also remain the same during the tests. If the app’s keywords change between tests, the results will be skewed based on the changes to its visibility.

The control group of creatives needs to stay the same so every variant has a consistent baseline to compare itself to. You’ll want to be able to isolate the changes you’re making. This enables a clean A/B test, so you can gleam the best insights from your results.


A/B testing takes time, but the results are well worth it. Each test provides new information about how your screenshots are performing and where you have room to improve. With the information from your tests, you can design screenshots that generate more conversions and increase your click-through-rate.

 Before you test, it’s important to do your research and find what aspects you want to integrate into your creatives. During the test, you’ll want to know what to focus on and run tests for each change individually. Doing so will provide you with the cleanest results and the best insights, so your final designs will be the best you can create.