The Engagement Trap

Here’s the uncomfortable truth: a recent review of 93 mental health apps found a median 30-day retention rate of just 3.3%—meaning nearly 97% of users stop engaging within the first month. Companies point to daily active users (DAUs) and call it success. But opening an app isn’t the same as improving.

Despite millions of downloads across therapy, meditation, and wellness apps, the same cycle repeats. People come in with hope, try for a while, then vanish. Yet the industry still celebrates those fleeting moments of activity as a win. Here’s what makes this even more problematic: effective mental health support doesn’t always look like daily engagement. Someone could check in occasionally—exactly when they need support—and experience genuine improvement. Conversely, someone could log in religiously every single day and see zero change in their anxiety or depression. The frequency of engagement tells you nothing about whether your product actually works.

Health and wellness aren’t defined by clicks or session count. They’re measured by lasting change.

What Actually Matters

In research with our partner Rula, an analysis of more than 5 million therapy visits showed that symptoms of anxiety and depression improved rapidly within the first five sessions, with progress continuing through session twelve. The findings highlight what truly matters: meaningful improvement over time, not just engagement metrics.

We also helped Bend Health (now part of Lyra) measure outcomes across youth participants. Seventy-three percent showed reductions in mental health symptoms, including anxiety and depression. They aren’t touting engagement rates or session counts—this is proof of real progress.

This is the standard the industry should be holding itself to.

The Retention Reality Check

The companies we partner with aren’t chasing clicks—they’re driving change. They recognize that evidence of improvement matters far more than numbers that only reflect activity.

Engagement isn’t meaningless. It shows action, momentum, even hope. It can signal that something’s working. But on its own, engagement isn’t enough. If people aren’t feeling better, DAU graphs don’t matter.

In our published work with Moment for Parents, retention exceeded industry norms even at six months. In their pilot, 63.9% of users who had stopped interacting for at least a week returned at least once—a clear sign of genuine value, not just habit.

The pattern is clear: retention follows outcomes. When people feel better, they keep coming back. When they don’t, they leave. Companies that focus on improving lives—not inflating metrics—naturally see engagement as a side effect of real impact.

The Drop-Off Nobody Talks About

The industry rarely talks about what happens in weeks two to four after someone starts using a mental health app. The initial excitement fades. The novelty wears off. And if nothing meaningful happens by then, people drift away.

Research shows a sharp drop in active users between weeks two and three, proving the early experience matters most.

Building habits is hard. It requires consistency, motivation, and a sense that the effort is paying off. When using an app doesn’t lead to real improvement, that habit never forms. Companies that succeed during this critical window don’t rely on streaks or badges; they deliver outcomes users can feel.

Because in the end, retention follows impact. Engagement that doesn’t improve lives isn’t progress—it’s distraction.

What Science-Backed Actually Means

Unfortunately, “evidence-based” has become a marketing buzzword often detached from real data.

What it should mean is measuring real outcomes—tracking whether symptoms improve, quality of life increases, and benefits last. Comparing results to alternatives. Being transparent about who it works for, and who it doesn’t.

It doesn’t mean citing someone else’s loosely related study and calling it validation. Collect your own evidence, and start early.

That’s what real science looks like: measured outcomes that prove impact, not vague claims like “people who meditate feel calmer.”

The Cost of Getting This Wrong

When companies optimize for engagement instead of outcomes, people spend months—or years—using tools that don’t help them. They believe they’re improving their mental health, but they’re just clicking, swiping, and logging in.

Meanwhile, companies celebrate DAU charts in board meetings as if attention equals progress. It doesn’t.

Every month someone spends using a tool that isn’t working is a month lost—time not spent on something that could help. It’s continued suffering and eroded trust in digital mental health itself. When engagement becomes the goal, people lose faith in the idea that these tools can work at all. That’s the real cost.

What Should Be Measured Instead

If engagement isn’t the answer, measure what you actually aim to improve. The goal isn’t to collect vanity data—it’s to understand whether your product truly helps people in ways that matter. To capture this evidence measure: 

• Change in the outcomes you target. If your product supports sleep, track improvements in sleep quality and consistency. If it reduces stress, measure the real-world impact on stress levels.
• Time to reach results. How quickly do users start to feel or function better?
• Sustained benefit. Do improvements last after people stop using your product?
• Who it helps best. No solution works for everyone. Be transparent about who benefits most.

Most importantly, partner with experts who know how to measure outcomes properly, like applied scientists, clinicians, or research strategists who can identify the right tools and interpret results with integrity. That’s how credibility is built.