With more marketing budgets than ever being invested in mobile devices, the cost of acquiring new installations is skyrocketing. Today, the biggest hurdle to success in the app space is reducing these costs. Faced with the challenge, growth professionals have learned to invest heavily in the "first user journey". From crafting the message in your ads to guiding users as you introduce them to the usability and value of your app.

The app store is ultimately the place where page views are converted into downloads. Visitors to a crowded app store decide whether to download an app after browsing the app store page for a few seconds. When publishers advertise apps, a lot of money is wasted if users go from the ad series link to the app store page and don't proceed to the download.

To optimize media spend and app marketing budgets, marketers should test app store creatives before launching campaigns. A good conversion rate can help your app become one of the hits of the store!

What should you learn from A/B testing?

First, let's define what A/B testing is.

A/B testing is the process of comparing two versions of a web page or app against each other to determine which version performs better.

It is essentially an experiment - two or more variants of a page are randomly displayed to the user. They act naturally, unaware of the test, allowing you to determine which performs better for a given conversion goal. It is well known that well-built A/B tests can increase conversion rates by up to 30%.

While A/B testing is primarily focused on improving conversion rates, the benefits go far beyond conversion optimization. It provides valuable insights into marketing campaigns, evaluates segmented audience segments and engagement/user experience, and measures the efficiency of new traffic channels. In addition, publishers can learn if they need to change their product positioning.

Planning A/B testing is an integral part of getting accurate results. Testing takes the guesswork out of optimization strategies. As a result, you can arrive at data-based decisions that shift the business conversation from "we think" to "we know".

Click "Learn More" to drive your apps & games business with ASO World app promotion service now.

iOS A/B testing and Google Play A/B testing

Also, don't use your mobile A/B test results from the Apple App Store in the Google Play Store. User behavior is different for each store, and the layout of the two stores is different. To get accurate results, you should test for each platform separately. 

Testing can be done directly in the Google Play Console. You just need to select your app and go to the left menu below. Click on Grow; then "Product Details Experiment". A/B testing is also now available in the Apple App Store, which is a brand new feature.

What app assets do you need to test?

Icons

The icon is another influential testing variable. Remember how everyone made such a big deal about Instagram icon updates? And we're seeing and designing a lot of white icons on gradient backgrounds these years.


A good icon design can grab attention, engage users, and even become an iconic brand. The icon is actually the first impression your app gives to the user. In both stores, icons are front and center in search, categories, and almost the entire storefront.

When A/B testing icons, you need to understand and test many different things to ensure that your icons distinguish your app from others, or are consistent with top competitors. For example, many mobile games have app icons with angry faces or open-mouthed characters. You can perform the same test. By getting to know your competitors, determining your app's messaging and experimenting with colors/graphics, you will be able to understand what attracts the most users to your page. This is the first step to start A/B testing. Next, design and implement! The length of the test will vary depending on the amount of traffic your page receives. One of the biggest mistakes we see developers make is that they are testing very subtle or insignificant changes. If you are going to take the time to run split tests, try to make sure that the icons are different enough so that users see two (or more) distinctly different icons.

App Screenshots

Screenshots on the Apple App Store and Google Play Store give you a great test room. You can use them for testing search results or app store listings. Here are some examples of what you can test on a screenshot.

1. Background color - light vs. dark

Contrast is the key to creating good screenshots, but contrast can go both ways - light to dark and dark to light. Intercept your screenshot and flip the foreground and background colors to make it your first experiment.

This is a very easy test to run, and although it is very simple, it can increase your downloads.

2. Background color - solid color with pattern

Another simple test you can try for the background is to use a pattern instead of a solid color.

3. Text color

The last test that involves color is the color of the captions. If your screenshot doesn't have captions, this is something you can test, but in all our experience, no captions = fewer downloads, so you can probably skip that test.

If the light background test you (definitely) run wins, you will have more flexibility in testing text color, but even if the dark background or pattern wins, you still have some options.

4. Text style

How you write your headlines can also determine if the view will translate into downloads, which really depends on your audience, so this is a good test item.

Pandora, the top-earning music streaming app, uses a hand-drawn style for their captions:

5. Fonts

The fonts you use for your captions will give them a "feel" that may also affect downloads or sliding backwards.

Rounder, thicker fonts tend to feel more friendly, while thinner, narrower fonts give a sense of authority. Try them out and see what happens.

Grubhub, uses a chunky font and bolds it to seem friendly:

6. Length of CTA

When it comes to captions, less is more, but one size doesn't fit all situations.

If your current captions are long, try shortening them, and if they're already short (because you're following best practices), try giving them more definition.

7. Text tone

Tone, or the feeling your CTA conveys, can have a direct impact on conversion.

Are your captions designed to make the reader feel good while reading them? Are they designed to instill a fear of missing out?

Try a variety of tones to see which is more effective for downloading.

8. Text placement - top vs. bottom

Where you place your text is important. While common sense says to place text about the visual it describes, this isn't always true. Try positioning the text above and below other visuals.

9. Text size

There are so many things we can try when it comes to text!

When it comes to captions, the size of the font can be important, so try a larger or smaller version of the captions. This is another very simple A/B test that you can run with only a few changes.

Robinhood, uses a large font to make the captions very easy to read.

App preview videos

App videos can improve conversions, but they don't work for everyone and they don't work for any type of video. Therefore, including videos in your app store listing is a great mobile app a/b testing idea. Should you have videos on your product pages?

When testing videos, remember that you should not duplicate content that is already in the screenshots. In addition, you should not show all the features in it. People will not watch your video for more than 20 seconds, so use it to showcase your USP.You can A/B test by changing the following.

• Features - Which feature should you show in the video?
• Color
• Period
• Have one or more videos (only available in the Apple App Store)

You can then compare the number of app downloads after watching the video. Keep in mind that videos are automatically launched in the Apple App Store, so you should not trust this metric completely.

What more you can test about?

There are more you can test about outside of app assets testing.

Pre-launch testing

If your app is not yet available, there are tools to test your product pages even before publishing the app. This means you can already test your product page with multiple hypotheses to see if it will convert.

It is a good idea to:

• Find your USP: If you're not sure which app feature is your unique selling point, you can use the pre-release phase to find that out. You can also do this with a short survey when people visit your item details. This way, you can find out which features they are most interested in.
• Find your target audience: See who is interested in your app.
• Test your app store listing: Is it converting?
• Lead Generation: You can create mailing lists. So users who log into your pre-release product details can add their email address to be notified when your app is released.
• Localization: Find out which parts of the world are more interested in your app.

Test in-app functionalities

In-app experimentation is another aspect of mobile A/B testing that is also very important. In this way, you can test in order to provide your users with the best possible app experience and keep them coming back and using your app often.

According to BuildFire, people use an average of 10 apps a day, and you want to show them why you should be one of them. For testing in-app experiments, it's best to have clear goals. These might be:

• Improve user retention: See if the number of engaged users changes when you add a push notification or change onboarding.
• Collect data from usage: Understand how users use your app and which CTAs they click on.
• Build new features: Find out what features your users really want or what features you shouldn't build.

If you're wondering what exactly in-app experiments can test, here are some tips.

• Onboarding: What's the best way to welcome new users?
• UX: Does the user's behavior meet your expectations?
• CTA: Did the user click on your CTA?
• Subscription/In-App Purchases: How do you trigger monetization in a way that gets users to pay?
• Feature Discovery: Where should you place features to make them more accessible?