App Download Estimates Explained by our Data Scientists
Thanks to its leading position in the ASO industry, AppTweak has accumulated a history of 7+ years of mobile app data from a large range of categories and countries. Leveraging this huge pool of data, our team of data scientists has created an algorithm that can estimate the downloads and revenue of any app in the App Store and the Play Store. Read below to find out more about our data scientists’ methodology and how we compute our estimates at AppTweak.
What factors influence app downloads?
Before we dig into the data science part of this article, we wanted to remind you of the two major elements that are closely related to mobile app downloads.
App downloads and category rank
The first is an app’s category ranking. We found a very close correlation between how an app ranks in its category and its daily downloads. Typically, the number of downloads an app makes in one day can have a direct impact on its category rank for several days.
Illustration of the correlation between mobile app downloads and category ranking.
Of course, each category has its own level of competition. Typically, the apps that rank in the top 10 of the Games category drive much more daily downloads than the apps that rank in the top 10 of the Utilities or Tools category.
Comparing the competitiveness of the Games and Utilities (App Store) or Games and Tools (Play Store) categories in terms of average daily downloads of top apps.
This is also true from a country perspective. For example, the US is by far the most competitive country on the App Store while India takes the lead on the Play Store. Games will require more daily downloads to rank in top positions in the Games category in these countries compared to others.
Comparing the competitiveness of several countries in the Games category on the App Store and Google Play in terms of average daily downloads of top apps.
Our data scientists have managed to calculate the competitiveness of each category and country, and therefore estimate the number of daily downloads an app needs in order to reach the top charts of a specific country and category.
The number of daily downloads an app needs to reach top positions in the Games category across several countries.
App downloads and seasonality
Another big element that influences the flow of downloads an app drives is seasonality. Seasonal events can have a huge impact on mobile app downloads. Here are a couple of examples to illustrate.
Apps in the sports industry tend to see their downloads fluctuate around popular sports events that take place in different countries. For instance, in the US, we see an increase of up to +40% in August and September, when the NFL season starts. On the other hand, in India, we see peaks of up to 57% in April, during the Indian Premier League.
Comparing the yearly seasonality of the Google Play Sports category in the United States and India
If we compare the yearly trend of the travel category across various countries, we can see that download trends are quite different from one hemisphere to another. Travel apps in the US and UK tend to see downloads increase during their summer (June to September), and travel apps in Brazil observe the same October through to February.
Comparing the yearly seasonality of the Google Play Travel & Local category across the US, the UK, and Brazil.
Other seasonality effects can be seen throughout the week. Below are a couple of examples of app categories that see different download trends on weekdays vs. weekends.
Unsurprisingly, mobile games tend to get more downloads during weekends, whereas business apps get more during the week. Another example is one of sports vs. health & fitness apps. They also follow quite different trends; the former typically drives more downloads during weekends, while the latter sees a boost during the week.
Comparing the weekly download trends of different app categories on the US App Store.
Using deep learning to enhance our data science estimates
To better understand the data science process behind these numbers, I asked our Head of Data Science – Jonathan Frisch – to explain the methodology and algorithms used by the team.
Deep learning (DL) algorithms have established themselves as a dominating force in machine learning. Both in the academic world and across the industry, these DL algorithms seem unbeatable. As a leader in the field of ASO, we decided to dip our toes in the water using DL to estimate the number of downloads an app makes every day in any country’s app store on iPhone, iPad, and Android.
One of the major advantages of using deep learning is having a model that is both (1) cross-country and (2) cross-category. This means that the model learns and understands the relationships between countries and categories, and therefore can benefit from the data of different categories in different parts of the world, to estimate the overall downloads of one specific app.
For each device and type, we feed our DL model with the data collected from the thousands of apps that are connected to AppTweak. For each app and for every day, we define a set of features related to:
- The app’s category rank
- The date (seasonality features)
- The country
We map these features to the number of daily downloads the app is making, allowing our model to learn the most relevant relationships between the different features and their related downloads.
This technique helps us to identify insights such as the impact of Christmas across different categories and countries, or which countries have similar levels of competitiveness in terms of downloads needed to reach top ranks in a given category.
Categories negatively affected by the Christmas holiday season in the US
The different impacts of the Christmas holiday season on apps in the Travel and Local category (Play Store) between the US, Brazil, and India
Once these relationships are well understood by our model, we are ready to predict the number of downloads of an app for which we don’t have data, from any category, in any country, and at any point in the past. Thanks to the many metrics we tracked during the learning process, we are confident that this new model can estimate the downloads of an app with high accuracy.
Visualizing real app data and AppTweak estimates for the downloads of apps in “All Categories” in the United States.
Specific examples of how our estimates perform
Our algorithms capture the effect of seasonality and specific events
By using deep learning algorithms, we are able to better capture the impact of seasonality and specific events on the downloads of an app. For instance, events such as the COVID-19 lockdown in 2020 or Christmas holidays might only impact some specific apps in some specific countries. Our model is able to capture the effect of such events on app downloads at a very granular level, and only when and where it is relevant.
For example, if we look at a set of apps whose activity is affected by the Christmas holidays in a country where Christmas has a high seasonal impact, we’ll see that the app ranking at #1 on December 26th will have driven many more downloads than the app that ranked #1 on November 26th in the same category.
We compared the Top Charts of the Entertainment category in the US. Apps in this category generally see downloads increase in the US around Christmas. As you can see, Netflix ranked in 3rd position on both November 26th (2019) and December 26th.
Comparing the Top Charts of the US App Store Entertainment category on November 26th and December 26th.
However, when we look at the app’s estimated downloads, we can see that they are considerably higher in December.
Download estimates for Netflix, US App Store
Another example is the case of the COVID lockdown in Q2 2020. Below are 2 examples of apps that were severely affected by the lockdown. The first is the travel app TripAdvisor, which saw its daily downloads plunge during the lockdown. The second is the business app Zoom, which witnessed an increase in downloads during the same period.
Estimates based on a large data pool
As mentioned above, our ability to provide our users with accurate download and revenue estimates is a result of the large pool of mobile app data we have managed to capture over the years. Our model uses the data from apps worldwide to predict estimates in a specific country. Typically, to predict the number of downloads a financial app makes in Vietnam, the model benefits not only from all the data we have access to in Vietnam but also from the data we have for any financial app in the world. Our pool of data grows significantly year over year as more and more users connect their data to AppTweak. This data is fed into our algorithms which are constantly improved over time.
- App downloads and category rank are closely related. The rank of an app in its category indicates the level of daily downloads the app is driving. Of course, each category and country has its own level of competitiveness to consider.
- Seasonality and specific events can significantly affect app downloads. The impact depends on various elements such as the type of app, the country, and the event itself. These subtleties are captured by our estimates model.
- Our model uses a huge pool of mobile data that has been accumulated over the past 7+ years. This dataset grows with AppTweak, which results in a higher level of accuracy for our estimates.
Check out download estimates for apps on the App Store or Google Play with AppTweak.