Wings Mob Blogs

The app rating system is going through hard times, because people have learned to manipulate them the way they want. Here 2 interests collide: developers and users themselves. App developers really need positive reviews (we’ll look at the reasons later), but ordinary people don’t need it at all. Anger usually becomes a strong enough motivator when the application is not working properly. It is quite rare to find conscious users who just want to support the service and tell that it is cool. Unless after being faced with a large number of low-quality applications, this development became a real relief. Well, there is a lot that I would like to talk about today regarding the rating system: how we are manipulated, why rating is so important, how it affects the number of downloads, where the problems are hidden here, etc.

What criteria does Google Play use to rate apps?

At the time of adding an application to the Play Market, the bot independently evaluates it according to several key criteria. The so-called pre-moderation. Over time, a real moderator can get to him, but everything will depend on the situation. In the App Store, the analysis procedure takes place mostly in manual mode, so you often have to wait for your turn for 2 months or even longer.

Important for Google:

• Duration of rendering. If the process takes longer than 16ms, that’s bad.
• The waiting time for a frozen screen. If the freeze lasts more than 700 ms, the probability of passing through the Play Market is sharply reduced.
• The number of abnormal terminations.
• Smartphone power consumption.
• The number of installations, as well as their ratio to the quick removal of the application shortly after installation.

It is clear that these factors are not as important to users as the actual efficiency of the application, the quality of its work and the benefits it brings. And the only way to determine this is through a rating based on user feedback.

How does rating and downloads compare?

The software company Apptentive has done research along these lines. The results are very interesting. I attach a picture below, and then we will discuss the main points.

What’s important here:

• 2 star rating. Unacceptable to the majority. A 1-star upgrade will improve conversions by 280%.
• Upgrading your rating from 3 to 4 stars increases the conversion rate from users by 92%. Most of the users cannot accept that the application has 3 stars, but they consider 4 out of 5 to be already acceptable.
• The difference between 4 and 5 stars does not significantly affect the number of installations. Only 4 people out of 100 can doubt and refuse the installation.

Why did this system stop working?

There are several reasons why the app rating system is now not an objective reflection of the real quality of the app, and here they are:

• People tend to be negative. To leave a review, you need some kind of incentive and very often it becomes a negative impetus. Someone out of 100 does not have the application working, and he has already left a review, and from the rest there will be 1-2 people with a positive review. That is, a fairly good software will receive 1 point and 2 by 5 points, as a result – 3.5-4 points.
• The developers have learned to use psychology. Application creators can display a message asking them to rate the quality of their work. They can display it at any time. Now the developers have started showing the proposal during periods when the person should be in high spirits. For example, a salary has just arrived and the online bank asks for an estimate. We passed a certain level in the game, talked to a friend for a long time via video link and the like. This is a kind of manipulation.
• A popular system of buying reviews. There are many sites on the web that offer the service of installing an application by the user (usually a bot or a person who does it for a small fee). Writing reviews may or may not be included in this proposal. For example, 1 installation costs only about 5 rubles, with a review – 15-20 rubles.

What we have at the end: real users often express their dissatisfaction, and developers are looking for any ways to deal with this phenomenon. Moreover, purchased reviews pull even a frankly bad application from the bottom. You’ve probably seen when some software is completely unimportant, but it has a rating of 4 or even 5 stars. This is no coincidence.

Will something change?

What can be done? Force users to write more detailed reviews – then SMMs will enter the market, who will do everything much better than ordinary people and the rating will no longer reflect reality from the word at all. Analyzing the duration of an application’s use and correlating it with a rating is technically difficult. Moreover, they are all different. Some are needed for one use, while others force the user to do what he wants for a long time, but they have no alternative.

Most likely the situation will not change, but there is one caveat. The Play Market has already upgraded the rating system, now 90% of the current rating is the average score from users, set since the last update, and only 10% is a score for the entire time. Thus, if the application was bad, but it was done well, it will go up in the rankings a lot after the update. If the utility was very decent, but after the update it does not start on some devices or some functions have disappeared, it will very quickly drop in rating. The idea is good, but it gives more room for manipulation.