From Options to Action: Evaluating Adoption of Privacy Features in Fitness - Tracking Platforms

Paper Title

From Options to Action: Evaluating Adoption of Privacy Features in Fitness-Tracking Platforms

Publication Info

• Topic area: Privacy feature adoption in fitness-tracking platforms.
• Keywords: Privacy features, fitness tracking, user behavior, Strava, Garmin Connect, Endpoint Privacy Zones, usability, privacy adoption, social platforms, data sharing.

Background and Problem

• Problem / challenge: Despite the availability of privacy features in fitness-tracking platforms, actual user adoption remains low, and the reasons for this underutilization are not well understood.
• Significance: Fitness platforms collect sensitive data (e.g., location, heart rate), which can expose users to privacy risks such as location inference and personal data misuse. Understanding adoption gaps is critical for designing effective privacy mechanisms.
• Motivation and related work: Prior studies have documented privacy risks and user perceptions but lack large-scale empirical analysis of privacy feature adoption. This paper addresses this gap by combining empirical data and user surveys to evaluate adoption patterns and barriers.

Solution

• Proposed approach: A mixed-methods study combining large-scale empirical analysis of 197,873 public activity records from Strava and Garmin Connect with a survey of 182 users to quantify adoption rates and explore reasons for non-use.
• Novelty:
  1. Systematic categorization of privacy features across major fitness-tracking platforms.
  2. Large-scale empirical measurement of privacy feature adoption on Strava and Garmin Connect.
  3. User study identifying barriers to adoption and motivations for privacy behavior.
  4. Analysis of hybrid privacy configurations and their implications for user behavior.
• Procedure and key techniques:
  • Categorization of privacy features into five categories: User Profile Controls, Activity Privacy, Location Protection, Data Management, and Social and Community Settings.
  • Empirical analysis of public data to measure adoption rates of privacy features.
  • User survey to gather self-reported adoption rates and reasons for non-use.
  • Statistical modeling, clustering, and thematic analysis to uncover adoption patterns and user motivations.

Results

• Concrete findings:
  • Only 36.42% of users set their profiles to private, and 2.12% of activities utilized additional privacy controls.
  • Endpoint Privacy Zones (EPZs) were used in 14.52% of activities.
  • Hybrid privacy mode was adopted by 25.4% of users.
  • Awareness of privacy features exceeded adoption, with lack of awareness (754 mentions) and perceived low necessity (367 mentions) being the most common barriers.
• Advantage over baselines:
  • First large-scale empirical measurement of privacy feature adoption in fitness platforms.
  • Integration of quantitative and qualitative insights to provide a comprehensive understanding of adoption barriers.
• Experiments / evaluation:
  • Analysis of 197,873 public activity records from Strava and Garmin Connect.
  • Survey of 182 participants to explore adoption rates and reasons for non-use.
  • Statistical techniques (logistic regression, chi-square tests) and clustering (PCA, k-means) to analyze adoption patterns.
• Limitations and future work:
  • Limited to publicly available data and self-reported survey responses.
  • Future work could explore real-time interventions, platform-specific usability improvements, and longitudinal studies on privacy behavior.

Summary

This study provides the first large-scale evaluation of privacy feature adoption in fitness-tracking platforms, revealing low usage rates despite the availability of robust privacy controls. Empirical analysis of 197,873 public records and a survey of 182 users identified key barriers, including lack of awareness, perceived low necessity, and usability challenges. The findings highlight the need for simplified privacy settings, stronger default protections, and privacy-preserving competition modes. By combining empirical data with user insights, the study offers actionable recommendations for improving privacy engagement and advancing research on privacy behavior in social platforms.