What Really Makes People Want to Use Fitness Apps? A sweeping review finds usefulness beats gadgets

Fitness apps have become one of the most visible promises of digital health: carry a coach, a logbook, and a motivational trainer in your pocket, and the logic goes, you will move more and manage your weight better. Yet the app store is crowded, downloads are easy, and sustained use is anything but guaranteed. A research team led by Xuecheng Li and Jiafu Huang of South China Normal University, working with colleagues at Zhaoqing University, Hainan Normal University, and the University of Victoria, has now tackled a deceptively simple question with unusual statistical rigor: what actually drives a person’s intention to use a fitness app? Their answer, published in BMC Psychology, comes from a systematic review and meta-analysis of twenty-nine studies covering 22,971 participants, and it cuts through years of conflicting single-study findings with pooled effect estimates that are hard to argue with.

The method behind the study is the kind of evidence synthesis that health psychologists consider the gold standard when individual surveys point in different directions. The team systematically searched six electronic databases, including PubMed, Web of Science, and Scopus, from their inception through August 1, 2026. To be included, a study had to empirically investigate fitness app usage intention and report quantitative path coefficients alongside sample sizes, the raw numbers needed for a statistically defensible pooling of results. Each article was then coded to extract the information required for the meta-analysis, which the authors conducted using a random-effects model. That modeling choice matters: a random-effects framework assumes that the true association between a psychological factor and usage intention may genuinely differ across studies, populations, and app types, rather than assuming a single universal value. Methodological quality of the included studies was assessed with the AXIS tool, a widely used checklist for cross-sectional research, and the team screened for publication bias with funnel plots, Egger’s tests, fail-safe N calculations, and trim-and-fill analyses, a battery of statistical probes designed to detect whether small or flattering studies have skewed the literature.

The headline result is strikingly clear. Perceived Usefulness, the belief that a fitness app actually helps a user achieve their health and fitness goals, showed the strongest pooled association with usage intention of any factor examined, with a correlation coefficient of r = 0.423 and a 95 percent confidence interval of 0.332 to 0.505. In the language of behavioral science, this is a moderate-to-strong effect, and its confidence interval sits comfortably away from zero, meaning the association is unlikely to be a statistical fluke. The finding aligns with the core logic of the Technology Acceptance Model, one of the most influential frameworks in information systems research, which holds that people adopt tools they judge to be useful before they worry about anything else.

Hot on its heels came a cluster of factors that the authors describe as social and experiential. Subjective Norm, the perception that important people in one’s life approve of or expect the behavior, produced a pooled correlation of r = 0.394, though its confidence interval of 0.037 to 0.662 is notably wide, reflecting less consistency across the underlying studies. Habit, the automated pull of repeated use, came in at r = 0.391 (95 percent CI: 0.283 to 0.489). Perceived Enjoyment, the intrinsic pleasure of using the app itself, reached r = 0.360 (95 percent CI: 0.193 to 0.507), and Satisfaction, the accumulated evaluation of past experience with the product, scored r = 0.356 (95 percent CI: 0.285 to 0.422). Together, these results sketch a user who starts with a rational assessment of utility, then stays because the app has become habitual, socially endorsed, and genuinely pleasant to use.

A second tier of moderate associations includes Performance Expectancy, Hedonic Motivation, and Perceived Ease-of-Use, concepts drawn from frameworks such as the Unified Theory of Acceptance and Use of Technology. These factors matter, but their pooled effects were smaller than the leaders. Price Value, Effort Expectancy, and Social Influence showed only small associations, a finding that may surprise marketers who emphasize discounts and social buzz. Perhaps the most provocative result, however, is the one that failed to appear. Facilitating Conditions, the belief that technical infrastructure and support are available to make app use possible, showed no significant association with usage intention at all, with a pooled correlation of r = −0.070 and a confidence interval spanning −0.375 to 0.249 and a p-value of 0.672. In other words, having the devices, the connectivity, and the support in place does not, by itself, make anyone want to open the app.

That null finding deserves careful interpretation. Facilitating conditions are typically conceived as enablers of actual use rather than of intention, so their absence from the intention equation is not necessarily contradictory. Still, the result carries a practical message for an industry that often invests heavily in compatibility and onboarding infrastructure while assuming that utility and social pull will take care of themselves. The meta-analysis suggests the opposite emphasis may be wiser: developers who want users to form an intention to engage should first ask whether the app demonstrably helps people reach their goals, and second whether it is woven into social expectations, enjoyable in its own right, and habit-forming through satisfying use.

The authors are candid about the limits of their synthesis, and those limits matter for how the findings should be used. Substantial heterogeneity was observed for most predictors, with I-squared statistics exceeding 75 percent for ten of the twelve variables examined. In plain terms, the variation among studies was far greater than chance alone could explain, pointing to genuine contextual variability across populations, countries, app types, and measurement approaches. A correlation measured among university students in one country using a running tracker may simply not generalize to middle-aged adults in another country evaluating a calorie-counting platform. The team also emphasizes that the included studies are observational, typically cross-sectional surveys, so the results should be read as preliminary associations rather than causal effects. An intention survey cannot prove that usefulness causes adoption; it can only show that the two travel together.

Despite these caveats, the scale and consistency of the evidence lend the conclusions real weight. The authors argue that an integrated framework combining utilitarian, social, and experiential factors is the appropriate lens for understanding user engagement with fitness apps, rather than any single technology-acceptance model applied in isolation. That integrative stance reflects a broader shift in behavioral science toward recognizing that health technologies are not merely tools but social and emotional environments. The global stakes are considerable: physical inactivity and obesity remain among the most pressing public health challenges worldwide, and fitness apps represent a scalable, low-cost channel for promoting physical activity and self-health management at population level, provided the science of engagement keeps pace with the engineering of features.

For the millions of people who have downloaded a fitness app and quietly abandoned it, the study offers a quiet vindication. The failure, the data suggest, is rarely about missing hardware or insufficient social media campaigns. It is about whether the app proves genuinely useful, whether it becomes a satisfying habit, and whether it enjoys the blessing of the people around the user. For developers, the to-do list that emerges from nearly twenty-three thousand participants is refreshingly concrete: prioritize functional utility and social features, adapt design strategies to specific user contexts, and treat ease of use and pricing as necessary but decidedly secondary concerns. The palm-sized gym, it turns out, succeeds not when it is most convenient or most connected, but when it first convinces its user that it actually works.

Subject of Research: Psychological determinants of the intention to use fitness apps

Article Title: Factors associated with the intention to use fitness apps: A systematic review and meta-analysis

Article References: Li, X., Huang, J., Deng, Y., Li, D., & Feng, X. (2026). Factors associated with the intention to use fitness apps: A systematic review and meta-analysis. BMC Psychology. https://doi.org/10.1186/s40359-026-05616-w

Image Credits: AI Generated

DOI: 10.1186/s40359-026-05616-w

Keywords: fitness apps, meta-analysis, systematic review, technology acceptance, perceived usefulness, subjective norm, habit, health psychology, physical activity, digital health, user intention, mobile health