The average mobile app consumes about 11.4 megabytes of data per minute. For many of us, this might not sound significant, especially given the enormous data allowances and the widespread availability of free Wi-Fi most are used to in the Global North. Scrolling through weather apps, playing mobile games and streaming videos on the train rarely gets a second thought.
However, there’s a hidden environmental cost to every megabyte of data we use. Data traffic translates directly to energy consumption. It powers the vast network of servers that store our data. These servers require constant cooling. Even the production of these servers contributes to emissions. So, while our data usage might feel limitless, its environmental impact is very real.
As part of the “Mobilsicher” project, the German Institute for Technology and Journalism was able to find out that a modern nuclear power plant would have to run for around one hour a day just for the energy consumption of our apps. However, as the current electricity mix does not only consist of nuclear power or renewable energies, our installed mobile phone apps certainly contribute to our digital CO2 footprint. However, if you want to live an environmentally conscious life, you can limit the data hunger of your smartphone apps. But first, we need to understand how this data and emissions are generated in the first place.
From mobile phones into the atmosphere – how are our “app gases” produced?
Most applications on our smartphones present us with up-to-date information, data and content from the internet. Weather apps no longer connect to a weather station on the roof; they access current weather data from servers. Mobile games allow us to compete against other players and obtain their moves and high scores from servers. Music, films, series and photos also find their way onto our devices from the cloud or streaming servers.
Tip: The “Priva score” for particularly data protection-friendly apps
IT security expert Dr Felix Sühlmann-Faul has compiled a list of apps on his website that are particularly privacy-friendly. As these applications also contain fewer trackers and adverts, they also tend to be more energy-efficient than apps with similar or the same functions.
The “Priva Score” currently covers categories ranging from messenger apps to dating apps. The service is free!
This information and content, which is the main focus of the applications, accounts for around 77 percent of the data traffic of our smartphone apps on average. However, just under 23 percent of the transmitted data is used for tracking usage data, displaying advertising and other analyses. This was the result of a study conducted by the Institute for Technology and Journalism using a test procedure for smartphones with Android operating systems. The following results, therefore, only include Android mobile phones; however, most applications are also available for Apple iPhones.
“Our study actually grew out of the topic of data protection,” explains Miriam Ruhenstroth from the Institute for Technology and Journalism in an interview. “Originally, we wanted to see exactly where apps are transmitting to. Many manufacturers don’t even know what the apps do because they use an ecosystem that many people don’t necessarily know.” Miriam Ruhenstroth explained that most mobile phone apps are made up of different components. And although these are used by programmers, they usually don’t have a precise insight into how they work.
Energy measurement via hacking methods
“This then led us to look at how high the energy consumption of apps is.” Miriam Ruhenstroth and her team used a study by the Federal Environment Agency, which “for the first time provides a reliable model to calculate what happens in the transmission grid and what it costs.” In short: thanks to this study, the researchers were able to understand how much energy is used in data traffic from the mobile phone to the respective servers that “spark” the respective apps.
The analyses had to be as realistic as possible, explains Miriam Ruhenstroth: “We installed apps on real devices or emulators. These were then used correctly for a certain period of time and the data collected using a machine-in-the-middle method.” In addition to the amount of data, the team was also able to track the addresses to which it was transferred in detail. According to Miriam Ruhenstroth, this was the key to being able to precisely analyse various apps.
This is because the allocation of data traffic ultimately also made it possible to check which data was required for the actual operation of the application and how much data had to be used for advertising purposes, tracking or analysis. In the study, which was last updated in 2024, this is summarised as “overhead” and thus as “data traffic [ … ] that would not be necessary for the actual function of the app”.
Is it even worth saving data with apps?
The amount of data required and the proportion of advertising and analytics data varies greatly depending on the type of application. In the study, the Mobilsicher team differentiated between categories such as “streaming”, “shopping”, “games”, “information” and “weather”. While news apps such as “Focus Online” only transfer a small amount of data, the data consumption of streaming providers such as Netflix or Amazon Prime is naturally much higher. This is because high-resolution videos are inherently larger than texts in which a few images are embedded.
This also explains why the percentage of advertising and analysis data for information apps is significantly higher (76 percent) than for streaming apps. Here, the ‘overhead’ only accounts for 4 per cent, as the required data traffic itself is significantly higher. And yet it is striking that information apps such as news apps or digital newspapers generate almost as much data traffic as gaming apps due to the use of trackers and adverts. Miriam Ruhenstroth and her team attribute this to the way in which the apps generate revenue, among other things. The study states:
“Most of the mobile phone apps in these categories [editor’s note: information, health and weather] are financed by advertising revenue – the high proportion of corresponding third-party providers is therefore not surprising. The proportion in the “Games” category, which are also often free, is surprisingly low. This could be due to the fact that game providers often generate revenue from in-app purchases.”
Green coding has been lost as a trend
According to Miriam Ruhenstroth, the fact that manufacturers can offer apps free of charge via advertising and tracking is by no means the main problem. “You can’t equate the two – apps that have a lot of trackers don’t necessarily generate a lot of data volume.” Although it is not entirely clear why some apps require more or less data, Miriam Ruhenstroth suspects that the programming plays a part in this.
“Game providers in particular used to optimise their apps for a low data volume, as this was still more expensive back then. Now that more and more people are using flat rates, optimisation for a low data volume has almost completely disappeared.” According to Miriam Ruhenstroth, we are seeing rebound effects here that can be explained by technologies that are actually more economical, such as 5G or cheaper mobile phone contracts.
Even if there are variations in what apps transmit data for, the study emphasises one thing very clearly: “Data protection is climate protection” and the superfluous power consumption of popular apps is currently too high. The application “My Talking Tom”, a mobile game with around one billion downloads in the Google Play Store, would consume almost 103 megawatt hours of electricity if every user used it for just one minute.
According to study results, this is enough energy to drive an electric car around the equator 12 times. The graphic shows how significant the proportion of advertising and tracking is in this app.
How can we reduce our app data consumption?
Miriam Ruhenstroth also emphasised in the interview that her study delivered comparatively conservative results. In other words, other studies with a different methodological approach had come up with higher results. We should therefore take apps seriously as power guzzlers. After all, they not only consume the energy that our smartphones already draw from the socket. And according to Miriam Ruhenstroth, they will require significantly more energy in the future as more and more manufacturers integrate AI functions.
However, consumers can take action themselves to minimise the energy consumption of their smartphone apps. “Google Maps, YouTube, Spotify and the like offer offline modes that allow you to avoid downloads via mobile networks. For data-intensive applications, content can therefore be downloaded via good networks.”

5G, WLAN, 3G – what are “good networks”?
The energy consumption of transmitted data on a smartphone depends heavily on the network used. Basically, you can remember the following rule: If you are using a Wi-Fi network, you are usually using the most energy-efficient network. Even if DSL connections, for example, require more energy than fibre optics, these technologies still use less power than the mobile network.
In the mobile phone network, the new 5G mobile phone standard is considered to be particularly energy-efficient. One of the reasons for this is that mobile phone masts can target specific devices using beamforming. 5G is around three times more energy-efficient than LTE – and LTE in turn is around three times more energy-efficient than the old UMTS network.
You can also do a lot with the app selection,” explains Miriam Ruhenstroth. “There are offline apps and apps without adverts for many functions, but unfortunately, there are no functions in app stores to find them reliably.”
In addition to the app selection, mobile operating systems offer various functions to keep mobile data volumes low. These are intended to prevent you from exceeding the monthly quota of the mobile phone contract you are using, but they are also suitable for saving energy.
Apple’s iPhone operating system, iOS, allows mobile data to be blocked specifically for installed applications. A corresponding menu can be found in the iOS settings in the “mobile data” category. It is a little more difficult to provide general instructions for Android mobile phones. This is because most manufacturers use customised Android versions that offer different options, especially with regard to system settings. However, anyone using an Android mobile phone has a special option with regard to tracking and advertising.
Google-free operating systems come without any trackers
Unlike Apple’s operating system, Android is an open-source system. This gives independent developers the opportunity to develop alternative operating systems for Android mobile phones. The Berlin-based organisation Topio sells Google-free smartphones with pre-installed operating systems such as LineageOS or KaiOS. Michael Wirths from Topio told us the following when asked how the use of trackers on Android mobile phones could be avoided;
“A good way to prevent tracking on a large scale is to change your own DNS settings with just a few clicks on your smartphone.” This would mean that many known servers, including those of advertising companies, would no longer be contacted, resulting in less data traffic and therefore energy savings. Anyone wishing to change their DNS settings can find step-by-step instructions at Topio.
However, using a mobile phone with a free operating system also has other advantages. By nature, they come with fewer pre-installed apps, which in turn means fewer distractions and less data traffic. Free operating systems also offer extended usability, making it possible to use older smartphones.
But even with mobile phones without tracking, it is important to handle your mobile data wisely. Taking care not to download certain data on a mobile network or keeping the streaming quality on YouTube low, for example, saves plenty of data.