How to Analyze App Store Reviews to Get Product Insights

When you want to understand how users perceive your app, your best bet is to look at your app’s average rating. However, if you want to understand what users are actually saying about your app, you need to analyze what is being said in the reviews section of your store listing. This can highlight common feature requests or help you identify any grievances your users have concerning your app.

Using AppTweak’s ASO tool, you can uncover a variety of interesting insights to help you understand and improve how users perceive your app. Calling on illustrations from real use cases, this blog will explain everything you need to know. Let’s dive right in!

App review sentiment analysis: Uncover strengths & weaknesses

In March 2020, the app Among Us! launched a new version amidst controversy that their chat system was not being moderated enough. As a result of this, they added a chat system that stopped users from being able to write their own sentences, instead showing an array of select answers that users could pick from. According to data from AppTweak, this update negatively impacted their ratings.

Discover the insights you can uncover from our improved Ratings section

However, how can we make sure that this sudden decrease in their ratings is attributed to the app’s version update? And above all, how can we be sure that users particularly didn’t like the changes made to the chat system instead of the other features added with this update? Our Sentiment Analysis section will help us pinpoint the most repeated user opinions to really understand what was said about the app update.

When you land on this section, we show a word cloud that already provides you with a lot of insights! It highlights the most repeated keywords found in the reviews fetched over the date range selected (March 2021). Here, we clearly see that “update,” “new update,” “chat,” and “quick chat” were among the most repeated keywords.

When you hover over a keyword, you will also see the number of reviews that keyword is found in and the average rating of all reviews containing that search term. In the example above, it is clear that the term “quick chat” received a lot of mixed reviews with an average rating of 2.89 (compared to an overall average rating of 4.5 at the beginning of March).

To get even more insights, the table below the word cloud shows a set of key performance indicators (KPIs) for the most repeated keywords. Unsurprisingly, the word “update” is the most repeated one (found in 4262 reviews) and we see that all the average ratings (for all reviews containing these repeated terms) are below the app’s overall average rating.

By clicking on the “Show reviews” button, you are able to read the reviews for yourself. Here, when users talk about the new update, they indicate that their frustration mainly derives from the new chat system. However, we also see that the second user in the image above “love[s] the new map”, suggesting that there was some positive feedback linked to this new version update.

The tab “Most positive keywords” allows you to see what aspects of your app users like the most. In this case, we see that the new airship map was a nice improvement for users. Thanks to this Sentiment Analysis section, we can now differentiate between the features that should be kept and those that should be removed/improved in order to maintain a good average rating.

Learn how you can monitor your app reviews over time and compare them with your competitors

Monitor & learn from how users perceive competing apps

Learning about what your users (dis)like about your app is important, but you should also check out the issues/requests your competitors are facing. This can demonstrate things that you shouldn’t do and the features that get requested the most. By implementing valuable features that your competitors don’t yet provide, you can boost your competitive advantage.

In the example above, we see that this competitor is facing an issue with the number of ads placed in its game. In October, the average rating from reviews equaled 2.31 which is really low compared to Among Us!

If we take a look at language e-learning apps, we see that the app Lingodeer gets a lot of positive reviews using the terms Korean- and Japanese-related terms. While these languages are not currently supported by Duolingo, if the app wanted to expand and attract (new) users, they have direct evidence that Korean and Japanese could be worthwhile options to develop.

Phone view: Understand what new users think of your app

When you analyze what is being said in reviews, it is always valuable to understand if new features work or if users are complaining about something in particular. However, how can you learn whether bad reviews are influencing new users when they land on your app page? Our Phone view section is all about understanding what users see when they check the Ratings section of your app page.

In the example above, the first frame shows the app’s current average rating as well as the rating distribution. This app’s score is not too bad (4.38) even though they implemented features that were not well received. In the second frame, we see the average rating of the first 50 reviews that appear when users click on the ‘see all’ buttons to browse the most helpful reviews on the app page. Once again, the score is pretty good (4.22) and we see that not too many bad reviews are shown here. In the third frame, we see the ‘featured reviews’ that are selected by Apple and Google to directly feature on the app page. These reviews are very important since most users don’t click on the ‘see all’ button and usually only consider these featured reviews when forming their opinion. In the example above, all featured reviews for this app are accompanied by 4 or 5 stars.

Basically, when a user lands on the app page for Among Us!, they typically won’t notice bad reviews and hence won’t be directly impacted by them. Nonetheless, if they start exploring different types of reviews, they may see something different:

If users switch from ‘Most helpful’ to ‘Most recent’, they will see that the average rating for these reviews drops drastically, going from 4.22 to 2.62. This section mainly displays negative reviews which may give users a totally different idea of the app.

Our Phone view section is very useful to understand how new users will perceive your app or if they will be influenced by bad reviews or not.

Get real insights about your app’s reviews & ratings

To wrap up, this blog is the last in a series of 3 blogs explaining the new features you can access in AppTweak’s upgraded Reviews and Ratings section! The focus here was more on the Sentiment Analysis and Phone view features to help you understand their potential and the kind of insights you can receive. Here is a summary of the main takeaways:

• Discover the most repeated terms used in reviews left for your app
• Understand the impact of new app updates
• Differentiate successful features from ineffective ones
• Learn from what users think about your competitors’ apps
• See if/how bad reviews impact new users’ perceptions of your app

Start receiving valuable insights from your app’s reviews and ratings to boost your conversion rate on the stores!