Youdao Dictionary

Personalized Vocabulary for Professional Contexts

EdTech

User Research

Product Thinking

April 2024

Overview

Youdao Dictionary is a widely-used mobile tool for bilingual translation and vocabulary learning. Despite its strong reputation, user feedback revealed a gap in personalized learning paths, especially for professionals needing industry-specific vocabulary. This project aimed to improve engagement and learning outcomes by introducing a Thematic Learning Mode that allows users to customize their word learning journey based on personal interests or professional fields.

Keywords

Finance, Retirement, Figma

My Role

UX Designer & User Researcher

Timeline

3 months

Team

Collaboration with 1 PM, 2 developers, and the content strategy team

Tools

Figma | User Surveys (WJX.cn) | Excel/Python (basic data analysis)

Main Feature 1: Topic-Based Entry Point

A new module on the homepage allowing users to select from themes like Finance, Medicine, Law, Technology, and Travel.

Main Feature 2: Curated Learning Interface

Users browse vocabulary lists with contextual definitions, pronunciation guides, and real-world usage examples.

From Researcher to Strategist

At the beginning, I approached the project as a UX researcher—simply uncovering user pain points. But as insights solidified, I gradually transitioned into a product-thinking designer, identifying opportunities, prioritizing features, and shaping the user journey. This taught me the importance of connecting user voice with product vision, not just reporting data but transforming it into actionable design direction.

Reflection 2: On Balancing Depth & Simplicity

One key challenge was translating fuzzy user desires into implementable features, especially when collaborating with engineers and content strategists. I took initiative to host mini co-design workshops, using low-fidelity sketches to align everyone's understanding. These informal sessions minimized misunderstandings and allowed us to reach consensus on feature scope and logic early on.

Context

How I transformed scattered survey feedback into a feature that reshaped the product’s direction

From Insight to Initiative – Turning User Cries into Product Strategy

When I first joined the Youdao team, user feedback was abundant—but directionless. I led the design and analysis of a user survey with 1000+ responses, surfacing a recurring demand: users wanted specialized vocabulary learning. Instead of merely reporting the data, I proposed a Thematic Learning Mode—a scalable feature that could bridge user needs and business growth.

I didn’t stop at the insight—I translated it into a clear product opportunity. By mapping professional users’ needs into themed content clusters (e.g., Finance, Medicine), I helped shift the product team’s mindset from “one-size-fits-all” to targeted, role-based learning. My early wireframes became the blueprint for development discussions.

How I aligned design, content, and engineering to ship a personalized learning flow

Designing the Bridge – Between What Users Want and What Teams Can Build

Designing a great idea is one thing. Getting a team to believe in it, align on it, and build it is another. While working on Thematic Learning Mode, I noticed early tension between content strategists (who feared complexity), engineers (concerned about scalability), and our design goals.

To break the silo, I hosted a series of mini co-design sessions. I shared low-fi concepts and invited honest feedback—not just on visuals, but on logic, feasibility, and content constraints. This helped the team align quickly, and even empowered content team members to own their part of the interface with confidence.

By facilitating dialogue and framing technical trade-offs in user value terms, I positioned myself not just as a designer, but as a design facilitator who could move ideas forward.

Outcome

Impact & Result

What Happened When We Let Users Take the Lead

While the full rollout was scheduled post-internship, internal testing and stakeholder feedback showed:

Increased engagement in prototype testing vs. standard wordlists (based on task completion rates and feedback).
Positive feedback from domain users during A/B testing mockups.