Digital democracy|Digital privacy
Which apps collect the most data?
Which apps collect the most data?
Almost every app collects some data necessary for its functionality — how else are you supposed to use a GPS without disclosing your whereabouts? Some apps that collect data go a step further. They collect users' personal details and track them across different platforms or sell it to data brokers.
To find out what apps are the least privacy-sensitive, we have to ask which apps collect the most data? We studied 100 of the most popular apps on Apple's App Store and created a gamified version of our study so you can learn AND play — the app privacy checker. Select the apps you use and see which data points and how many of them get collected.
To start the game, click START, and the app privacy checker will open up in a new tab.
An overview of how the collected data is handled
Let’s see how much of the collected data is linked to you, and how much of it is used to track you across services. To get a better hang, we’ve come up with a categorization system.
Note: we’ll refer to one specific form of data as a data point.
According to Apple¹, there are three ways how the collected data point may be handled:
- It may be used to track users.
- It may be linked to users.
- It may not be linked to users.
Linking is when a service provider associates the collected data with the user’s identity. As per Apple, “data collected from an app is often linked to the user’s identity unless specific privacy protections are put in place before collection to de-identify or anonymize it.” Also, “personal information” or “personal data,” as defined in relevant privacy laws, is usually linked to the user’s identity by default.
Tracking means connecting or associating data gathered from the app about a specific user or device (like a user ID or device ID) with information from sources outside the said app (such as a third-party advertising network). Tracking is used mainly for targeted advertising or advertising measuring purposes. Also, data could be shared with data brokers (companies that collect information). They create detailed data profiles based on demographics, behavior patterns, and interests and then sell the info to various companies and institutions². Collected data may be used for advertising, market research, financial risk assessment, and more.
We analyzed 100 of the most popular apps on the App Store to answer how user data is handled. We looked into what data points each app collects based on its Apple App Store page. Then, we put each app into one of ten categories. And finally, we looked at how many and what kind of data points each category collects, and this is what we found out:
Shopping & food delivery apps are the most data-hungry app group. On average, these apps collect 21 out of 32 possible data points. Moreover, these apps link the most data to the user — 95% of collected data points are linked to the user’s identity. Also, these apps tend to use collected data to track the user the most — a third of collected data points are used for these purposes.
Wish could be named the most data-hungry app within this app category, collecting 24 out of 32 data points, linking all but one data point (Other data types) to the user's identity, and using over a third of data to track its users. Some of the data that DoorDash uses to track its users includes precise and coarse location, email address, and purchase history.
Out of the analyzed Shopping & food delivery apps, only one — Amazon — does not use data to track its users. But it collects most of the unique data about the user (25 of 32 possible data points), and also, ALL the collected data is linked to the user’s identity.
On the other side of the spectrum are AI tools. On average, they collect the least data points and link the least data to the user's identity. But when it comes to tracking, work & education and entertainment apps tend to perform best — they use the least data to track their users.
What data is used the most by the top 100 apps: an overview
To better understand the state of app privacy, we looked at how much of the 32 types of data every app collects. And this is what we got:
1523 data points are collected across 100 of the most popular apps. Statistically speaking, that's an average of 15 unique data points per app out of the 32 unique data points defined by Apple. Now, let's analyze the apps in the form of data points gathered. We'll go from the most general information gathered to the most personal.
Around 90% of the apps collect usage, diagnostic, and identifier data such as product interaction, user ID, device ID, crash and performance data. Most are essential for their app functionality.
Two-thirds of the apps collect your name and coarse location, and nearly half collect your precise location. Coarse location is a more general estimation of where you are, while precise location is more detailed and accurate. Over a third of the apps collect your contacts, and a fifth collect your emails or text messages and browsing history.
The top 10 least privacy-sensitive apps
Now let’s see what are the worst apps for privacy. We came up with a special measurement system to rank each app according to its data hunger and ways of sating it.
Apps were ranked according to the total number of data points collected, 32 being the max points an app can get. The apps that collect the same amount of data points were sorted by the amount of data they use to track the user.
Facebook and Instagram are the two most privacy-invasive apps. Both apps collect all 32 data points defined by Apple and are the only two to do so.
All collected data is linked to you, with 7 out of 32 data points used to track you. This includes name, email address, phone number, and physical address. Since Instagram and Facebook are products of Meta Platforms, Inc., it's no surprise these two apps collect and handle data the same way.
Three more apps use data points linked to you to track you. These are Wish, DoorDash and TikTok. All three apps collect 24 data points. That’s considerably more than the average of 15 across all examined apps.
Around 40% of data points collected by Wish and DoorDash are used to track you — your email address, precise location, and purchase history. The same can be said about three of TikTok's collected data points — email address, phone number, and device ID.
The top 10 most privacy-sensitive apps
Let’s see what apps are least likely to share your secrets. We ranked the apps according to the total number of data points they collected. Apps that gather the same amount were ranked by the amount of data they use to track its users.
AI Generated Art is the least data-hungry app since it doesn't collect any data points and is the only one to do so. Regarding AI-powered apps, 4 out of 10 we looked at made the least data-hungry app list.
Signal is the second least data-hungry app, collecting just 1 data point (phone number) that is not linked to you or used to track you. Signal is also the only social media and messaging app to make the top 10 most privacy-sensitive list.
Two of Signal’s main competitors, WhatsApp and Telegram³, didn’t make this list. They’re collecting 15 and 11 data points, respectively, and all the data is linked to you, although not used to track you.
How much data is too much?
It’s up to you to decide, but consider the research when doing so. We hope this study gives you an insight into which apps collect what data, and mobile privacy in general. Remember that some apps collect all data they can on you, so remain cautious when considering mobile app privacy.
Before downloading apps, we recommend checking the developer's reputation and data retention policies. Pay attention to constant permission requests to access your contact list, camera, storage, location, and microphone. We suggest limiting the app's access to your sensitive information only when the app is in use⁴.
Methodology and sources
We analyzed a total of 100 apps across 10 app categories. The apps for each category were selected from articles that appeared at the top of search engine result pages for "the most popular appCategoryX apps" keyword. Once we had the list of apps, we collected each app's data collection information from its Apple App Store page on May 17, 2023.
The App Store lists 32 unique data points that can be collected across 12 unique data point categories. We analyzed the data set according to the three layers of collected data points: unique data points collected, number of data that’s linked to the user, and data that’s used to track the user.
We re-collected privacy policy data on November 29 to see if there were some changes since then. A few apps became more data-hungry than stated in this study based on May’s collection. Detailed changes can be found in the research materials spreadsheet, which can be found by following the link below.
For the complete research material behind this study, visit here.
