Changing icons and screenshots is a crucial element in implementing App Store Optimization (ASO). This is because such changes not only affect organic downloads but also impact key performance indicators (KPIs) like conversion rates and revenue from advertisements.

Moreover, the relationship between creatives and their impact differs between iOS and Android. It's a well-known fact that in the Apple App Store, screenshots have a greater influence, while in the Google Play Store, it's more effective to change the app icon. The reason for this lies in the different screen real estate allocated to these elements in search results.

In the App Store, screenshots occupy a significantly large area in search results, consequently making a substantial impression on users.

Conversely, in the Google Play Store, the app icon is prominently featured and, in some cases, is the only element by which apps can be distinguished.

Thus, as a rule of thumb, focusing on screenshots in the App Store and icons in Google Play is essential. But how significant is the impact of these creative changes?

In this case, let's explore an example where an icon was changed in the Google Play Store. This involved conducting A/B testing of different icon versions, which appeared similar but varied in color and shape, using the Google Play Console. The most successful version, as it turned out, managed to achieve double the organic installs compared to the default icon. Below are the specific characteristics of each icon tested.

1. Features of the First Icon: The icon's edge is blue, and the egg is centered.
2. Features of the Second Icon: The icon's edge is yellow, and the egg is positioned to the right.
3. Features of the Third Icon: The icon's edge is blue, with enhanced glossiness on the egg, making it more prominent.
4. Features of the Default Icon: The icon prominently features a character.

Discussion of Results

Let's consider the reasons behind the successful results.

Firstly, icons featuring the egg yielded overall better results compared to those that solely focused on the character. Why might this be? A significant factor was the keywords. This game had a high number of installs from search keywords related to "dragon" and "breeding game." Therefore, by creating an icon that conveyed the idea of "dragon" and "breeding = egg," we achieved better results than with the default icon. In particular, making the egg golden suggested the presence of a rare character, adding to the icon's appeal.

Next, let's compare each icon. The second icon, with its yellow edge and golden egg, resulted in a blending of the character and egg, weakening the impact of both. Consequently, the first icon performed better. The third icon, despite its blue edge, slightly underperformed compared to the second icon because the character's impression was diminished.

Below, you can see the overall ranking progression immediately after the icon change. It's evident that the ranking improved significantly just from changing the icon. This was accompanied by improvements in search keyword rankings as well.

Positive results from changing an app icon can significantly impact the app's Return on Investment (ROI). This is exemplified in the following case study of a Google Ads campaign. The change in the icon is indicated by the red line. Following this change, there was a substantial improvement in ROI, quickly achieving a result of "ROI = 115%." This outcome suggests that there was not only a significant improvement in the ad's conversion rate (CR) but also a considerable increase in users likely to make in-app purchases.

Furthermore, it is believed that the success of this initiative was due to the effective appeal of the "dragon" and "egg" themes, which resonated well with what users were seeking and matched the game's content. This result stems from a strategic approach where icons were developed based on reverse-engineering from key keywords, providing a relevant and effective visual representation.

iOS A/B Testing

You may now understand that the term "keywords" encompasses a variety of usage methods. Building on this insight, we have developed an A/B testing functionality using Apple Search Ads (ASA). This approach differs from traditional A/B testing; it doesn't split all paid and organic traffic for testing. Instead, it allows A/B testing at the keyword level. Optimizing screenshots under the most impactful keywords can significantly enhance overall organic performance. In one instance, merely improving the conversion rate for a single keyword led to an increase in the product page's conversion rate (CR) from 25% to 50%.

We do not advocate for haphazard A/B testing. Meaningful and reproducible results come from a scientific approach based on solid evidence. We also share case studies like this, so if you think our insights could be of assistance, please contact us for more information.