Fix user retention leaks for AI SaaS
Challenge
AlgoIRL launched their first MVP which received a lot of interest among users. However, they experienced a high bounce rate, unable to retain customers. My role was to detect the leak in the funnel and redesign the experience to improve retention and increase the number of paying customers.
Process
The first step was to define a preliminary research protocol to map research goals and questions, and methods to get answers.
Key Insights after conducting interviews and speaking with the users:
Our main challenge is communicating why it's worth paying for over the free, DIY alternatives.
Our marketing and UX must project authority and trust, show social proof.
Users are tech savvy and demand transparency and control.
With this, I explored the user's golden path from their perspective to identify where users were dropping off.
Critical insights:
Our 10-step demo wizard creates friction and contradicts our “effortless coach” promise, leading to user drop-off and validating their “do it myself” objection. To reduce time-to-value and prove our unique benefit, we should guide users to their personalized “Aha!” moment (the plan) within seconds.
Currently, users see us as a generic “ChatGPT wrapper” because we fail to differentiate. We need to surface our core value, such as visual “Before/After” comparisons and our in-built code editor.
Redirecting to LeetCode encourages users to skip our platform, hurts habit-building, and increases churn. Introducing in-page hints and updating completion criteria can drive engagement with the editor, building investment and improving retention.
Solution
This is where the research and strategy turn into a real solution!
We are currently in the design phase, building the wireframes and user flows needed to tackle our 3 core opportunities. The "effortless coach" experience is being built right now.
Updates and the first look at the prototypes will be posted soon! 🚀