1. Background

Inspired by Data

In reviewing open-source dataset, I noticed that personalized language learning is not universally accessible for everyone. In multi-grade classes with higher peer pressure and less individualized context, the positive correlation between SES and verbal IQ is even stronger than in single-grade class (p<.01).

Using a Linear Discriminant Analysis classifier, I predicted the probability of an intelligent student to have an above-median language scores with different SES.

Industry Analysis

Mobile devices present a unique opportunity to enhance the accessibility of language learning. E-learning market’s mobile segment was projected to experience a 148.7% growth from 2019 to 2026, suggesting greater potential market share to be captured.

Mobile E-learning is cheap and accessible, but its course plans are often inflexible presets. Despite the opportunities of cross-industry collaboration, the technological barrier of this segment is low, so fierce competition should be expected.

2. Explore

Contextual Inquiry

I conducted a contextual inquiry with frequent users of Duolingo, observing how they complete practice sessions over an entire week and asked their overall impression towards the course.

User Interview

Three Duolingo users were interviewed about the desirable features of online learning experience from their perspectives. Referring to the e-learning products they liked, I centered the discussion around functions they frequently used. Their quotes are categorized into meaningful codes.

Sample Data

Persona

Journey Map

3. Define

Benchmarking

Based on key dimensions identified through thematic analysis, I constructed perceptual map of various benchmarks in e-learning industry. Strategically, I decided to position my product at “high context” and “mid proficiency” combination.

Product Requirement

By constructing a Kano model on the features (1) consistent with the theme of “contextual + proficient” and (2) consistently spotted in competitors, I evaluated their priorities and streamlined a subset of them into a coherent flow chart that depicts how a language learner might engage in their daily learning. With shortlisted features at hand, I conceived the user flow.

4. Solution

Wireframe

Low-fidelity wireframes are constructed to capture essential content and pivotal transitions. In Cosmospeaker, users can choose design their own AI-boosted learning agents with unique appearances, personality, and experiences, specializing at the topics of user’s interest. Each learning session would start with these topics supported by corresponding CEFR word bank.

Design System

Cosmospeaker’s pleasantly warm design system is established upon core concepts of “embrace, support, immerse”. All design elements are systematically and consistently organized into reusable variables and components variants that can directly implemented by software developers.