In recent years, Google has been applying machine learning technology to their services.  Read more previous blogs about games or app marketing method, keyword install or anything related to ASO on asobooster.com, the best agency that will help your app outrank your competitors. Conatact us today.

The turning point for Google Play has also opened brand new doors, as they have decided to work on recommending better games and apps to users based on internal metrics (such as app performance, in-app purchases, app deletion, etc.). But many developers think that this adjustment by Google is excellent for those who make products properly, because it is the bright hope.

But in the process of moving forward, it is inevitable that there will be a little twist, because just since June 19th, many developers are complaining that there is a serious decline in new users installing apps in Goolg play, mainly for natural traffic. For some developers, this figure reaches 90% of the number of new users per day. This problem affects not only Russian companies, but also foreign ones. This problem has been discussed in many forums, for example, Reddit.

For example, the problem that affects natural users mainly comes from search and recommendation systems. We all know that search is the most important source of getting natural traffic, accounting for more than 50% of all app installs. But the other day we would see the first 16-20 apps cycling through the display after searching for keywords, and the remaining 230+ apps in each search query disappeared from the search.

Also we found similar apps in which irrelevant products appeared.

With both issues, there was a dip in app exposure and a drop in install conversion rate from 15.5% to 11.8%.

Chi Huai-hsin, Google’s lead AI researcher, said – “Google has implemented computer-based learning technology on Google Play and has successfully used machine learning techniques to create an accurate personalized situational app recommendation system that has increased Google app install rates by 3.3 percent.

Currently, Google play has over a million apps available for download for users of 2 billion active Android devices as they found over 82 billion app downloads in 2017. However these numbers are sure to continue to expand in the future and the Chi Huai-hsin research team shows that it is also a very important task for them to improve the app install rate.

Chi said that Google’s recommendation system has three main principles to follow in order to provide better recommendation quality to users. First, the recommended application content must be diverse and meet personalized needs. Secondly, the interface of all application products must be continuously optimized with the help of machine learning. Third, Google’s recommendation system must be applicable to all users, not specific users. In order to increase app downloads from Google Play, user preferences must first be understood and personalized recommendation systems must be built to meet the needs of different identities and age groups, and then contextual elements must be added to such systems to more accurately recommend apps to users. For example, tablet users typically prefer entertainment video apps, and smartphone users prefer to download apps related to everyday life, such as lifestyle tools, email, calendars and notebooks. In addition, news apps typically see more downloads during the day, while gaming apps are more popular at night, according to Chi.

More robust algorithm tuning could address 3 current issues.
1. better help apps will get more users, which will spur further growth and growth across the mobile app ecosystem.
2. The revenue level between giant apps and other apps will be blurred. Current statistics show that 1 percent of apps get 95 percent of the revenue. With the new recommendation algorithm based on individual user needs, the share of “giants” should decrease slightly.
3. The disappearance of the gray industry, in which cheating based on the number of installs and rating reviews in the hope that the app is found in the overall list and classification, as well as in a particular keyword search. Because Google Play found that there are many developers of low quality apps that like to go to some extreme ways to make revenue. The presence of this initiative is good both for the users who download the apps and for the high quality apps. Because of this, developers will not spend their budgets on gray promotions to keep their position, but rather try to develop high quality volume apps and hopefully gain a high quality audience to further profit. This way as well, many high quality developers, can spend their budgets on formal and measurable sources of paid channels.

Google’s ranking mechanism continues to be improved, showing more determination for user protection, hoping to show better applications to users, and hoping that developers can develop better applications.

Share on facebook
Facebook
Share on twitter
Twitter
Share on linkedin
LinkedIn
Share on reddit
Reddit
Share on whatsapp
WhatsApp