The best AI products should make themselves less needed over time
There's something counterintuitive about building an AI product in the mental health and self-awareness space: if you're doing it right, your users should eventually need you less.
Most product teams optimize for stickiness. More sessions, more time in app, more daily returns. But at Murror, we've been wrestling with a different question — what if the goal of our product is to help someone build enough self-understanding that they don't need to open the app as often?
When someone uses Murror to process a difficult emotion or reflect on a pattern in their relationships, the ideal outcome isn't that they come back tomorrow to do the same thing. It's that they start recognizing those patterns on their own, in real time, without us.
This creates a genuinely hard product challenge. How do you build a sustainable business around a product that's designed to reduce its own usage?
Here's what we've been learning:
Users who "graduate" from daily use become your best advocates. They don't churn — they shift to occasional, intentional use and tell everyone about the product that actually changed something for them.
Depth of impact beats frequency of use. One reflection that leads to a real conversation with a partner is worth more than 30 days of streak-based check-ins.
You can measure success differently. Instead of DAU, we look at whether people report applying insights from Murror in their real lives. That metric has been far more predictive of long-term retention and referrals than any engagement number.
I think this tension is going to define the next wave of AI products, especially those in health, education, and personal development. The companies that figure out how to align their business model with genuine user progress — even when that progress means less usage — will build something really lasting.
Would love to hear from others building in this space. How do you think about the tension between engagement and genuine user outcomes?
Replies
I am curious how you measure real world application 🤔 It sounds powerful, but also harder to track compared to simple engagement metrics
Murror
@rakesh_gupta20 Great question! It is harder to track, honestly. We use two main signals: (1) a short follow-up prompt 3 days after a session asking whether the user applied any insight from their reflection, and (2) behavioral patterns — users who shift from daily use to weekly but keep returning for months tend to be the ones reporting the most life impact. We also track referral rates, since people who genuinely benefit tend to tell others. It's messier than DAU dashboards, but the signal quality is much higher.
This idea feels right but also uncomfortable. I have always though of retention as the main goal, so building something that reduces usage sounds almost against instinct.
Murror
@new_user___0932026a86e905cf4b2b7f7 Totally get that tension. It felt counterintuitive to us too at first. The reframe that helped: reducing usage frequency isn't the same as losing users. Our "graduated" users still come back during big life moments — a conflict with a partner, a career decision, a tough week. They just don't need us every single day, and that's actually the point. Retention still matters, it just looks different — it's measured in months and years, not daily streaks.
This is a really refreshing take.
Designing for “graduation” instead of addiction feels much more aligned with actual user outcomes.
Murror
@judit10 Love the way you framed that — "graduation" instead of addiction is exactly how we think about it internally. It also changes how we design features. Every new capability we build gets filtered through the question: "Does this help the user eventually need us less?" If it doesn't pass that test, we rethink the approach. It's a harder design constraint but it keeps the product honest.