How text optimization helped the Vivaldi browser improve visibility and expand the semantics twice as much
Here’s how we ran the first ASO iteration for the Vivaldi browser on Google Play, and how we managed to improve visibility in each selected country.
Vivaldi is a fast, private and secure browser with built-in ad and tracker blocker. The browser is not only designed for Android devices, it also works on PCs.
We started with an audit to understand the strengths and weaknesses of the app, so we could develop hypotheses to improve the app ranking in the app store.
ASO audit results on Google Play
Since Google Play indexes metadata of all localizations, for the audit and the first iteration we chose countries that could influence the growth of traffic from several regions where local languages are used: Russia (RU), United States (US), Brazil (BR), Spain (ES).
The text was optimized by regular translation without taking into account the cultural background of the region. This is quite common for the initial stage of app development, but the metadata did not include keywords.
The audit revealed a rather low Search Visibility Index of 48% (RU), 36% (US), 33% (BR), 30% (ES). SVI is a score of an app visibility for search traffic, which shows how easy it is to find an app using a search query on Google Play. Such a low score means that the app is not indexed for all relevant keywords and it ranks low in search results.
The app drove the majority of its traffic from branded keywords. The rest of the general niche keywords were out of visibility and mostly ranked among the top 50/100.
The visual ASO (icon, screenshots) of the browser was performed quite well, but we also gave some tips on some little things to improve to boost conversion rates.
The main optimization issue could be called low SVI and poor traffic from Search on Google Play. In addition, such competitors as Chrome, Opera, DuckDuckGo and Yandex.Browser are products with a long history.
- Enhance the app visibility and increase the number of keywords in the top;
- Boost traffic from Search by using non-branded keywords.
What we’ve optimized for Google Play
Having identified 4 target countries (RU, US, BR, ES), which can make the greatest impact, we started experimenting with text ASO to find the best options with the highest number of relevant keywords.
During such global changes, it’s important to strike a balance, so that the app home page is not reindexed. Reindexing can cause a loss of ranking positions and a steady flow of organic traffic for an indefinite period.
Our goal was to update the texts in several locales at once, including the main locale. In an environment where even a small change can pessimize the ranking positions, our work was very risky.
We have developed several text versions in Russian, English, Portuguese and Spanish to enhance indexing and ranking for keywords. New text elements have been optimized based on existing metadata.
To avoid re-indexing and pessimization, we decided to gradually test short and long descriptions.
- First, running experiments is a well-known life hack to improve indexing.
- Secondly, it helps to evaluate descriptions in terms of conversion rates.
- Third, this way we tried to help the Google Play algorithm “get used” to the new texts.
We carried out a series of tests, as a result of which the most converting descriptions were published in the target locales. The other versions, which also showed an increase in conversion rate, were published in additional locales. For example, versions in Russian were published, respectively, in the Russian locale, as well as in Azerbaijani, Belarusian, Georgian, etc. At the same time as applying the results of the experiments, we updated the titles and repeated the experiment with all the texts for the target countries.
Selecting the best texts through a series of experiments helped the algorithms get used to it, improve ranking and boost conversion rates. As an example, you can see a screenshot of the result of the experiment with descriptions for Russia.
We revised the semantics for 4 countries, involving not only the target locales, but also additional ones. This led to an almost twofold increase in the total number of indexed keywords in each country.
Our experiments with metadata allowed the app to become more visible for general keywords, for example, for the super-relevant query “браузер/browser” we managed to go up by 20 (RU) and by 60 (US) positions. For the first iteration, given the popularity of the competitors, this is a good result.
The results we achieved are only the initial stage of ASO. It was a pleasure to deal with such a high quality and engaging app, as well as with the team of professionals at Vivaldi. The product team continues to improve performance and shared a comment on the work done:
We’d like to thank the Checkaso team for their work on improving the visibility of our app on Google Play.
Thanks to our joint efforts and progress in increasing the app visibility, we’ve been inspired to continue our experiments in exciting ways to grow key ranking metrics.
We were able to improve the app rating from 4.0 to 4.2 because the value props now match what users see in the actual product.
Now, we are adapting the value props in headlines and images in all locales and collecting new feedback for the next iterations.
We realize that ASO is a marathon, not a sprint, so we will continue with our careful iterative work. Our immediate plans include working on visuals, onboarding, and conducting regular quantitative and qualitative user research.
If you want your app to reach its users and become really visible on the App Store or Google Play, email us at [email protected] to get started.