Losing more than calories: The privacy trade-off in using fitness apps

Spring is in the air and, with the marathon season coming up, so are the starting pistols ushering in the start of running season. Competitors in these races have already been preparing for months, but now’s the time regular folk take to the streets, pavements, and trails to start or restart their fitness journeys.

In a time when there’s an app for everything, it’s not surprising that people are reaching for fitness apps to help achieve their health and fitness goals. But what’s the trade-off that comes with using these apps? Whether free, paid, or subscription-based, apps like these help themselves to users’ personal data.

Many of the same reasons health and fitness data is so sensitive are also what makes it so valuable to data brokers. Insurance companies, advertisers, and governmental bodies¹ all have their uses for such information, and data brokers are only too happy to oblige. An app that tracks users’ eating and exercise habits, location, and more is a treasure trove of personal data.
Incogni’s researchers gathered information on the leading fitness and sports apps by revenue² and examined what data they collect and share by consulting their developers’ data collection disclosure statements. The results reveal some concerning data collection practices while also suggesting a path forward for those looking for a leg-up on their way to a healthier lifestyle.

Key insights

• The top US fitness apps collect an average of 15.2 personal data points each about their users, with exercise tracker Fitbit collecting the most at 21 data points. 
• Wellness platform BetterMe shares 17 data points with third parties, including photos, videos, and health and fitness information.
• Two fitness apps—Peloton and Yazio—share location data with third parties, while 4 apps share user IDs for “advertising or marketing purposes.” 
• Seven out of nine apps studied collect users’ photos and videos, while 5 collect their precise location.
• The WeightWatchers: Weight Health app collects information on users’ race and ethnicity, while the YAZIO Food & Calorie Counter app collects information regarding users’ sexual orientations. In the first case, providing this information is optional, but in the second, it’s required to use the app.
• The vast majority (72.9%) of collected data points are non-optional, meaning that using the app is synonymous with agreeing to the collection (and possibly sharing) of these data points.

Shared data ends up in the hands of third-party companies, which can include data brokers, and can subsequently be sold to hedge funds, insurance companies, advertisers, and government agencies.

The personal data these apps collect and share

There are two things a fitness app can be set up to do with users’ personal data: collect it from the user’s account information, device, and interactions with the app, and share it with third parties, including by selling or siphoning it through to data brokers and other companies. An app may collect data without sharing it.

The apps we investigated collect an average of 15.2 data points each. Fitbit stands out as the most data-hungry app with 21 data points collected, which is 32% above the average. Justfit, on the other hand, collects the least, at 5 data points (2/3 of the average).

The kinds of personal data these apps collect are also a cause for concern. We found 4 apps collecting precise location data (MyFitnessPal, Strava, Peloton, and Fitbit), 3 of which also collect approximate location data (Strava, Peloton, and Fitbit). So these apps’ developers know where their users train and possibly which other locales they frequent.

Many of the developers of the studied fitness apps have their users’ contact details, with 8 apps collecting email addresses and 7 of those apps also collecting users’ names. Two apps (Fitbit and WeightWatchers: Weight Health) collect users’ phone numbers.

Worryingly, at least one app’s developers can keep tabs on where their users live, with the YAZIO Food & Calorie Counter app collecting users’ street addresses.

Some of the fitness apps we studied collect personal information that seems particularly difficult to justify. The WeightWatchers: Weight Health app, for example, collects information on users’ race and ethnicity, while YAZIO Food & Calorie Counter collects users’ sexual orientations. 

Moreover, according to the YAZIO app’s data collection disclosure on the Google Play Store, sexual orientation is not an optional data point. This means that, in order to use the app, users have to provide this information. It’s certainly difficult to make a case for knowing a user’s sexual orientation before providing dietary-tracking functionality.

All 9 studied apps collect the vaguely defined subcategory of personal information known only as “other info.” This virtual catch-all comprises “[a]ny other personal information such as date of birth, gender identity, veteran status, etc.”³ The data points listed are concerning enough—what hides behind that “et cetera” all the more so.

Thankfully, we found that the fitness apps we looked at don’t share the data they collect with third parties as much as apps in other categories we’ve investigated. Still, the BetterMe app, for example, shares 17 out of the 18 data points it collects with third parties.

Developers are obliged to state reasons for collecting and sharing personal data through their apps.³ Although open to manipulation, this requirement on the Google Play Store gives at least some insight into the purposes behind some of this data collection and sharing.

For example, Peloton shares users’ approximate location for the purpose of advertising, while BetterMe shares email addresses, app interactions, and in-app search histories for the same purpose.

The stated purposes behind all this data collection and sharing

Given that the Google Play Store requires app developers to state the purposes behind the collection and sharing of data through their apps, our researchers were able to dig deeper into this aspect of their operations.

One line of thought that motivated our researchers was the idea that the more purposes that are given for collecting or sharing data, the more is likely being done with it. This, in turn, would mean more people and companies seeing and handling that data, making it an important metric for assessing privacy risks. 

BetterMe gives the highest number of purposes for both collecting and sharing their users’ personal data. For each data point collected, BetterMe provides an average of 4.3 purposes; for each data point shared, it provides an average of 3.7 purposes.

Strava, having tied for second place in terms of how much data it collects (19 data points), provides the second most purposes per data point collected, at 3.5. However, it provides only one purpose each for the 3 data points it both collects and shares.

The WeightWatchers app, tied with Strava in terms of data points collected, provides an average of 3.4 purposes for each data point collected. This app’s developers claim it doesn’t share any data with third parties. MyFitnessPal provides 2.5 purposes for each data point shared, which is the second highest in the sample (although it claims to share only 2 data points).

For collected data, “app functionality” and “analytics” each account for around a quarter of all purposes provided, while “analytics” is the most popular purpose given for sharing data, by a significant margin.

The app Calorie Counter by Lose It!, which shares two data points, shares them both exclusively for advertising or marketing purposes. Peloton, which shares 3 data points, shares them all for two purposes, “advertising or marketing” and “analytics.” YAZIO, which also shares 3 data points, shares two of those for advertising purposes.

Collected data points are used for marketing less frequently, with 11.7% of all purposes given for collecting data being marketing-related. Meanwhile, 15.2% of all purposes given for sharing data are marketing-related.

Optional data

Not all the data points these app developers collect and share are required to use their apps. The question is whether users know they can skip over some input fields when signing up for and setting up these apps.

Among data points that apps request but that users don’t have to give up, race and ethnicity information is asked for by the WeightWatchers app but is entirely optional. Phone numbers (collected by the WeightWatchers and Fitbit apps) are optional in both cases. Photos, collected by all apps except for JustFit, are also optional.

Some noteworthy data points are optional for some apps but not others. For example, health information is collected by the Calorie Counter by Lose It!, WeightWatchers, BetterMe, and Fitbit apps, but is optional in the Fitbit app.

Fitness information is collected by all 9 apps, but is optional in 4 of them: Strava, Peloton, Calorie Counter by Lose It!, and Yazio. Users’ purchase histories are collected by Strava, where they’re not optional, and also by Fitbit and BetterMe, where this data point is optional. Email addresses are collected by all apps except for JustFit, and are optional only in the Yazio and BetterMe apps.

Some data points are mandatory for every app that collects them. Yazio collects sexual orientation information from its users without offering the option to skip this detail. Information concerning installed apps (collected by the Strava, Calorie Counter by Lose It!, and Fitbit apps) is not optional in any of the apps that collect it.

Conclusions

There’s no doubt that fitness apps—whether used for planning and logging training sessions or formulating and sticking to a diet—can be a huge help in achieving health and fitness goals. It’s not all benefits and gains with these apps, though. Whether the user pays money for a given app or not, they certainly pay with their privacy.

But privacy can mean many things to many people. Users of apps like these exchange personal information such as their contact details, demographic data (including sexual orientation), and precise location for access. Many of these apps will also track and report on the user’s interactions with the app.

The problem is that most users are likely unaware that this is the deal they’re making with the app developers. Even fewer users are likely to realize what consequences can come from having their personal data harvested and sold to data brokers.

Health and fitness information is both particularly sensitive and particularly valuable to data brokers. Why? There’s a range of interested parties willing to pay for such information, including insurance companies, marketers, and all manner of spammers and scammers.

Allowing personal data to reach these third parties can lead to more robocalls, spam emails, and junkmail. More than that, it can result in users receiving more and better-targeted scam attempts, having loans mysteriously rejected, and seeing insurance premiums go up for no apparent reason.

Thankfully, people looking for a new fitness app have tools like the Google Play data collection disclosures at their disposal. Armed with these and the knowledge that at least some collected data points are optional, users are able to make informed decisions about which apps to allow onto their devices and into their lives, and how to limit the data collection potential of those they do end up choosing.

Methodology

Having identified the top 9 fitness and sports apps by revenue², Incogni’s researchers collected information about each of them from the Google Play Store. In cases where several apps were published by the same company, apps aimed primarily at health monitoring were prioritized.

Incogni’s researchers then noted what user information the selected apps’ developers declared collecting and sharing and what purposes they gave for doing so. This data was collected from the apps’ data collection disclosures on the Google Play Store on March 27, 2024.

The data used in this study is available here: public dataset.

Note on data:

The list of most popular fitness apps by revenue included two apps made by Fitbit. Only one of those was available on the Google Play Store, so only it made it into the analysis above.

Sources

1. Electronic Frontier Foundation (EFF). “Location Data Brokers.” Accessed April 8, 2024. https://www.eff.org/issues/location-data-brokers.
2. Statista. “Leading fitness and sport apps worldwide in January 2024, by revenue.” Last modified March 4, 2024. https://www.statista.com/statistics/1239716/top-fitness-and-sport-apps-by-revenue/.
3. Google. “Understand app privacy & security practices with Google Play’s Data safety section.” Google Play Help. Accessed April 8, 2024. https://support.google.com/googleplay/answer/11416267.

Visuals

We welcome the reuse of our images if proper attribution is given to Incogni. The charts, graphs, and tables used in this research can seamlessly embed into your website. Use the menu that appears at the top right of the visual when you hover over it with your mouse. When embedded, these visuals maintain their interactivity and preserve their original quality.