State of Behavior Change in Tech

Published on May 27, 2024

Most apps don’t support behavior change

We have seemingly perfected the art of engagement in the tech world, successfully dragging people down neverending rabbit holes of content. But yet, even the most popular digital products still have some way to go when it comes to real-world behavior change.

This is especially true for digital health and self-improvement apps. This category of behavior change apps has been on the rise for quite some time, driven by a shared mission of promoting healthier, more productive lives. Often designed to encourage healthy behaviors and discourage unhealthy ones, they have the potential to alter our habits and how we behave on a daily basis. Yet, the question remains: How effective are those apps really?

When looking at the effectiveness of many behavior-change apps, we see some major challenges. This article will highlight these and hopefully spark ideas for improving them.

“Behavioral features sprinkled on top”

A primary concern lies in the foundational aspects of how these apps are developed. While many apps aim to influence behavior, most lack a solid grounding in behavioral science. For instance, research covering 65 apps designed to improve physical activity shows that many apps tend to favor similar behavior change techniques while neglecting others. And out of these 65 apps, only 12 were backed by a peer-reviewed study.[1] To make matters worse, it’s more common for apps to have some ‘behavioral features sprinkled on top’ rather than to be built on a behavioral foundation at their core. This lack of scientific basis undermines their potential effectiveness in fostering meaningful engagement needed to achieve lasting behavioral changes.

Meaningful engagement: Going beyond mere entertainment

To engage and retain users, apps often aim to master the art of habit formation: a skill that separates the successful from the ephemeral in the digital world. Look at how companies like Meta have woven oversharing into the very social fabric of online culture, or how Twitter has nurtured a culture of rage-scrolling, and TikTok has perfected the pull of impulsive video browsing. Yet, the behavior change apps we’re talking about need to go beyond these examples – in the behavior change domain, apps need to activate, motivate, and support users rather than merely ‘engage’ them. These products aim to enrich lives and encourage healthy behaviors in fitness, mindfulness, or even therapy. To do so, they need to go beyond entertainment, empowering users on their journeys toward self-improvement and personal aspirations. Success requires people to take action in their lives outside of their screens.

And here comes the well-known challenge: our high aspirations don't always manifest as actual changes in behavior. On average, the 1-month retention rate for all health-related apps is only 4%.[2] Trying to copy social media engagement strategies will do little to solve this problem. The engagement that drives social media success – typically passive and superficial  – doesn't readily align with the active, meaningful engagement necessary for behavior change. There is a vast difference between the shallow, momentary interactions in mindless ‘doom-scrolling’ apps and the more profound transformative process of behavior change, such as establishing a new exercise routine.

Different levels of engagement

So how can we drive more meaningful engagement with behavior-change products? First of all, it’s helpful to distinguish between two distinct levels of engagement.[3] ‘Little e’ engagement is the interaction with a digital solution itself, including the features aimed at changing behavior, which is typically measured with traditional product metrics. For example, for an app like Noom, this could be program sign-up rates, weekly task completion rates, or coach chat engagement. However, solely optimizing for Little e falls short if you want to have a meaningful impact on behavioral outcomes.

This is where the second, crucial level of engagement comes in: ‘Big E’. This represents the actual behavior – the target behavior in users’ lives that the product aims to achieve. Coming back to Noom, that could be altering food habits, and potentially exercise frequency to support people in losing weight over time. It’s one thing to use an app every day; it’s another to have it support you in changing your behavior for the better. Little e is important, but Big E is that moment where engagement has been translated into real-world behavior change.

In essence, Little e is the building block that facilitates and enables the achievement of Big E. The successful design of a behavior change product therefore depends on effectively marrying these two forms of engagement – ensuring that the product is not only appealing and user-friendly (Little e) but also effective in meaningfully improving health behaviors in the real world (Big E). The product is merely the vehicle of change, a delivery system for what users truly value. And it’s important to not mistake the vehicle for the destination.

A mismatch between theory and application

Not many behavior change apps leverage the most effective techniques to help people change their real-world behavior. Several studies have underscored this challenge. [4,5] For example, the way most habit apps are designed often doesn’t align with the science of habit formation. An analysis of 115 popular habit apps revealed that features supporting habit formation are often lacking.[5] Still, many of these apps remain popular and widely promoted in app stores. This indicates a disconnect between what we know about how habits are formed and the practical application of features to support habit formation in app development.

The same can be said for behavior change techniques. The same analysis as mentioned above showed that self-monitoring is the most common behavior change technique in apps. Yes, tracking your behavior – and ideally getting feedback on it – can support the behavior change process, but tracking alone doesn’t help people to adopt different behaviors in their daily routines – or get rid of undesired ones. Furthermore, the ongoing task of logging behaviors can become a burden, potentially discouraging users and reducing the likelihood of having a lasting impact on behavior.

What’s more, most apps don’t help users adopt behavior beyond the product but instead create a dependency on the product to keep up the behavior. You could think: “Well, great! They need my product!”. But ultimately, this reliance on products means that once users stop using the app for whatever reason, their newly 'learned' behaviors are at risk of being abandoned with it.

A new method to assess the effectiveness of behavior change products

Measuring the effectiveness of digital behavior change solutions, such as health and well-being apps, is complex. While users’ digital footprints provide insights into users' digital behavior, they don’t fully capture whether interactions with the product lead to lasting behavioral impact in users’ lives. Behavior change happens in the real world, not just in a digital product. However, while changes in behavior and habit formation are more meaningful measures of success, they are also more difficult to quantify. Surveys and external measures can help, but those come at a price, as tracking burden can reduce engagement – not to mention the likelihood of socially desirable responses.

In light of this, there's a pressing need for a more nuanced framework to evaluate the behavior change potential of digital solutions in practice. This framework should go beyond superficial metrics and delve into how effectively an app can instill and sustain healthy behavior among its users. Because of the difficulty of measuring off-product behavior, such a framework will need to focus on process, and assess how well a product engages with known behavior change techniques.

In the coming weeks, we will share just that.


  1. Bondaronek, P., Alkhaldi, G., Slee, A., Hamilton, F. L., & Murray, E. (2018). Quality of Publicly Available Physical Activity Apps: Review and Content Analysis. JMIR mHealth and uHealth, 6(3), e53.

  2. Argyle, M. (2024, March 12). Mobile app user retention benchmarks broken down by industry. Sendbird.

  3. Cole-Lewis, H., Ezeanochie, N., & Turgiss, J. (2019). Understanding health behavior technology engagement: Pathway to measuring digital behavior change interventions. JMIR formative research, 3(4), e14052.

  4. Renfree, I., Harrison, D., Marshall, P., Stawarz, K., & Cox, A. (2016, May). Don't kick the habit: The role of dependency in habit formation apps. In Proceedings of the 2016 CHI conference extended abstracts on human factors in computing systems (pp. 2932-2939).

  5. Stawarz, K., Cox, A. L., & Blandford, A. (2015, April). Beyond self-tracking and reminders: designing smartphone apps that support habit formation. In Proceedings of the 33rd annual ACM conference on human factors in computing systems

    (pp. 2653-2662).