Abstract
Physical activity helps people maintain a healthy weight and reduces the risk for several chronic diseases. Although this knowledge is widely recognized, adults and children in many countries around the world do not get recommended amounts of physical activity. While many interventions are found to be ineffective at increasing physical activity or reaching inactive populations, there have been anecdotal reports of increased physical activity due to novel mobile games that embed game play in the physical world. The most recent and salient example of such a game is Pokémon Go, which has reportedly reached tens of millions of users in the US and worldwide. We study the effect of Pokémon Go on physical activity through a combination of signals from large-scale corpora of wearable sensor data and search engine logs for 32 thousand users over a period of three months. Pokémon Go players are identified through search engine queries and activity is measured through accelerometry. We find that Pokémon Go leads to significant increases in physical activity over a period of 30 days, with particularly engaged users (i.e., those making multiple search queries for details about game usage) increasing their activity by 1473 steps a day on average, a more than 25% increase compared to their prior activity level ($p<10^{-15}$). In the short time span of the study, we estimate that Pokémon Go has added a total of 144 billion steps to US physical activity. Furthermore, Pokémon Go has been able to increase physical activity across men and women of all ages, weight status, and prior activity levels showing this form of game leads to increases in physical activity with significant implications for public health. We find that Pokémon Go is able to reach low activity populations while all four leading mobile health apps studied in this work largely draw from an already very active population.
Introduction
Physical inactivity is a global crisis, contributing to millions of deaths annually and billions in economic costs. While mobile health (mHealth) apps aim to solve this, they face a "preaching to the choir" problem: they primarily attract individuals who are already active. Researchers from Stanford and Microsoft Research utilized a massive dataset of wearable sensor data and search logs to quantify the impact of Pokémon Go on physical activity. They investigated whether "exergames"—games that require physical movement—can serve as a more effective intervention than traditional health apps. Using the 2016 Pokémon Go phenomenon as a natural experiment, the study sought to determine if a game-first approach actually increases steps, if the effect is sustained, and if it reaches the sedentary populations that need it most.
Methodology
The study employs a multi-modal approach to bypass the limitations of small-scale clinical trials, utilizing objective data from approximately 32,000 users over three months. To identify Pokémon Go players without direct access to game servers, researchers analyzed Bing search logs, categorizing 454 unique queries into "experiential" (e.g., "how to play") versus "non-experiential" (e.g., news about accidents). This allowed them to pinpoint exactly when a user started playing ($t_0$).
Activity was measured using objective accelerometer data from Microsoft Band wearables rather than self-reported surveys. The team compared 1,420 identified players against a random control group of 50,000 wearable users, matching start dates to account for seasonal or weather-related changes. Finally, Pokémon Go’s performance was benchmarked against four leading, anonymized mobile health applications.
Key Results
The most striking finding was that Pokémon Go players were less active than the average wearable user before they started playing. In contrast, users of traditional health apps were significantly more active than average before even downloading the apps. This suggests games can penetrate sedentary demographics that clinical interventions cannot.
On average, players increased their activity by 192 steps per day (+13% over control). However, there was a clear dose-response relationship: particularly engaged users (those searching for game guides/mechanics) increased their activity by 1,473 steps per day—a massive 25% increase over their baseline. For these highly engaged users, the proportion of people meeting official physical activity guidelines (approx. 8,000 steps/day) jumped from 12.2% to 31.7% after they started playing.
In the first 30 days of release, the researchers estimate Pokémon Go added 144 billion steps to US physical activity. By applying life-table analysis, the authors estimated that if the engagement seen in the first 30 days were sustained, the game could add 2.8 million years of life to its US user base.
Discussion and Limitations
Unlike many fitness trends that skew toward specific niches, the "Pokémon Go effect" was observed across all ages, genders, and BMI categories. Even obese individuals (BMI 30–40) saw significant activity gains. However, the data also shows a "novelty decay": activity levels peaked and then began to decline after 3 to 4 weeks. While the highly engaged group stayed above their baseline, the "honeymoon phase" of the intervention is limited.
Future work is needed to determine how to design game mechanics that turn a short-term spike in activity into a lifelong habit. Additionally, while the health benefits are massive, the societal costs—such as anecdotal reports of accidents and trespassing—require further study. The primary challenge for future "games-for-health" remains sustaining the high levels of engagement seen in the initial weeks of a global phenomenon like Pokémon Go.
Conclusion
Pokémon Go has demonstrated a unique ability to reach "activity-poor" populations that traditional health apps often miss. By prioritizing gameplay over explicit health goals, the app accidentally became a public health powerhouse, adding billions of steps to the population total and showing that exergames can be a viable tool for large-scale health interventions.
Summary updated