# Effectiveness of Step Goal Personalization Strategies on Physical Activity in a Mobile Health App: A Field Study

## TL;DR

This study compares two ways of personalizing step goals in a mobile health app and finds that effectiveness varies depending on users' initial activity levels.

## Contribution

The study provides empirical evidence on the effectiveness of user-set versus algorithm-set step goals and identifies how these strategies impact different user segments.

## Key findings

- Personalized-by-you and personalized-by-the-algorithm strategies both increased weekly steps compared to default goals.
- Effectiveness of personalization strategies varied by baseline activity level, with different impacts on low, medium, and high-active users.
- The not-changed group also showed a modest increase in weekly steps, suggesting a placebo or Hawthorne effect.

## Abstract

Goal personalization features integrated into mobile health apps have the potential to enhance physical activity, as some evidence shows that the personalized goals generated by algorithms are more effective than default or fixed goals. However, it remains unclear whether goals set by users are more effective than fixed goals and which personalization strategy is more effective for different user segments.

This field study aimed to evaluate (1) the efficacy of 2 step goal personalization strategies—personalized-by-you and personalized-by-the-algorithm—and (2) which strategy is more effective among users with different activity levels.

All users of SamenGezond, a Dutch mobile health app, have a default goal of 2000 steps per day, 5 days a week. For this study, 2 random groups were selected, totaling 5800 users. Subsequently, an email was sent to 3800 users in group 1, asking whether they were satisfied with their current goal. Those who were not satisfied were offered 2 personalization options: to set a goal themselves or to have the algorithm integrated in the app set goals for them. In total, 1399 users responded: 230 chose to set their own goals (personalized-by-you group), 236 opted for setting the goal by the algorithm (personalized-by-the-algorithm group), and 933 chose to keep the default goal (not-changed group). The algorithm used a moving-window percentile rank method based on step data from the previous 4 weeks. Users who did not personalize retained the default goal. The remaining 2000 users in group 2 did not receive the email and also retained the default goal. To evaluate the effectiveness of step goal personalization strategies, we used propensity score matching and difference-in-difference analysis.

Users in the personalized-by-you group increased weekly step count by 3793 a week, while those in the personalized-by-the-algorithm group increased by 4315 steps a week, compared with the not-changed group (users with default goals). The 2 strategies appear to have a similar effect. Interestingly, users in the not-changed group also increased their weekly steps by 1759. Furthermore, the effectiveness of each strategy varied by baseline activity level. The personalized-by-you strategy was effective for medium- (increase of 5842 steps) and high-active users (increase of 4266 steps) but not for low-active users (increase of 384 steps; P=.82). Conversely, the personalized-by-the-algorithm strategy was effective for low- (increase of 5095 steps) and medium-active users (increase of 5278 steps) but not for high-active users (increase of 1446 steps; P=.51).

Step goal personalization demonstrates short-term effectiveness. However, their impact varies by users’ baseline activity levels, indicating the need for a tailored approach for different user segments. Future studies should examine the long-term effects of such interventions to design sustainable health behavior change strategies.