Context

AMA Interview is an AI-powered interview prep platform offering mock interviews, practice questions, and coaching. As features expanded, engagement stayed strong—but paid conversion plateaued at 2.9%.

At first, this looked like a UX clarity problem. But deeper analysis showed a different issue: users weren’t disengaged—they were hesitant.

The core challenge wasn’t usability. It was decision paralysis at the moment of entry.

Outcome & Impact — Quick View

+58% Paid Conversion: from 2.9% to 4.59%.

Peak Daily Conversion: 11.11%

The Real Problem - Too Many Valid Ways to Start

New users landed in a feature-dense product and were immediately asked to decide:

What should I practice?
How should I prepare?
Which tool fits my situation?

Each option made sense on its own. Together, they created friction before users experienced value. When everything looks valuable, nothing feels urgent.

The Question We Had to Answer

The hardest part wasn’t designing flows—it was answering a product question: What should this product ask users to do first? We identified three viable entry paths:

1. Question Practice — low friction, low commitment
2. Study Plan — structured, but heavy upfront
3. Mock Interview — high commitment, but instant value

Choosing one meant explicitly deprioritizing the others. This was a product decision, not a layout tweak.

Strategic Decision — Clarity Over Comfort

We intentionally chose the highest-commitment action—starting a mock interview—as the entry point. This was a risky decision.

It asked users to act before fully exploring the product.
It could discourage cautious or undecided users.

But the upside was compelling:

Mock interviews already drove the most meaningful engagement.
They made the product’s purpose immediately clear.
They dramatically shortened time-to-value.

We decided to optimize for clarity over comfort.

Core Solution — A Guided, Task-First Entry

To support this decision, we introduced a dedicated Getting Started experience. Instead of exposing all capabilities upfront, the product now guides users toward a single, high-value action:start a mock interview.

Key changes included:

Removing feature-level choices from the entry moment
Framing the experience around action rather than exploration
Aligning the first interaction with real usage patterns

This reduced cognitive load and let users experience value before making further decisions.

Parallel Strategy — Why Study Plan Came Second

While the entry experience focused on immediate action, we also addressed long-term engagement through a parallel strategy: Study Plan. Mock interviews enabled fast activation. Study Plan supported control and continuity.

While mock interviews helped users start quickly, Study Plan existed to help users regulate their pace, feel grounded, and see progress. The goal wasn’t strict sequencing, but balance: speed without anxiety, momentum with reassurance.

Entry optimized for action. Study Plan optimized for confidence and trust.

Navigation After Activation

Once users understood where to start and how to continue, we reorganized the rest of the product around clear, intent-based destinations:

Question Practice
AI Mock Interviews
1:1 Human Coaching

This structure preserved flexibility for advanced users while keeping navigation predictable and easy to scan.

Reflection — What This Project Changed for Me

This project reshaped how I think about entry design.

Entry is not about showing possibilities.
It’s about deciding what kind of commitment a product asks for first.

In complex products—especially AI-driven ones—more intelligence doesn’t always mean better experiences. Restraint, clarity, and sequencing matter just as much as capability.This principle later informed how I approached agent-based systems and AI workflows, where trust is built not through breadth, but through clear, intentional first actions.