Project Overview
Yz English is an independent, scenario-based learning product that helps newcomers preview and rehearse unfamiliar everyday interactions. As founder and product designer, I defined the positioning, built the content system and AI-assisted production workflow, launched and grew its social channels, and translated audience traction into a live web MVP.
Project
Scenario-based English learning product
Role
Founder & Product Designer
Duration
July 2026–Present
Platforms
Web, Douyin, Xiaohongshu, WeChat Channels
Scope
Results at a Glance
Moving abroad involves more than learning a new language. People also have to navigate unfamiliar processes, social expectations and service interactions - often without knowing what will happen next. Traditional lessons may teach the words, but they rarely show how a real interaction unfolds.
Design challenge
How might I help people preview and rehearse an unfamiliar situation before it happens, so they can navigate it with greater confidence?
Core insight
“When people can preview how an interaction may unfold, they can prepare both their language and their expectations.”
Design the experience around life moments, not language levels.
Yz English is a practical rehearsal tool, not a linear academic course.
01 Find a real-life situation
Browse moments such as a clinic visit, parent-teacher meeting, product return or phone appointment.
02 Preview what will happen
Understand the setting, participants, likely steps, common questions and realistic dialogue.
03 Practice at the right depth
Watch the full scenario, replay line by line, complete listening practice, review key phrases and download a study guide.
04 Save or download a guide
Keep key phrases and scenario notes close at hand by saving or downloading a guide for quick review before the real interaction.
I built a repeatable production system, not a collection of disconnected AI outputs.
Real-life trigger → Dialogue → Storyboard → AI footage → Edit & subtitles → Multi-platform release → Performance learning
For every scene I defined the communication problem, refined the dialogue, directed recurring characters, broke the story into shots, evaluated visual continuity, edited subtitles and pacing, then learned from audience response.
Turning a growing content library into a working MVP.
Social posts created discovery, but learners needed a structured place to browse, watch, study and return.
Explore by scenario and category
—
Review vocabulary and language notes
—
Watch short situational videos
—
Download learning guides
—
Study line-by-line dialogue
—
What the system made possible.
Cross-platform audience
10K followers on Douyin
4,668 followers on Xiaohongshuaohongshu
1.3K followers on WeChat Channels
As of August 2026. Audiences may overlap across platforms.
Product outcomes
40+ scenarios produced
Working web MVP
Repeatable multi-format workflow
Video, dialogue, dictation and downloadable guides connected in one experience
The project created a foundation for testing paid learning resources and future membership experiences.
Performance Insights
Douyin — Reach and Discovery
Signal: 10K followers and 21K likes.
Interpretation: Stronger opening hooks and faster context setup were associated with better retention and follower acquisition, making Douyin the primary discovery channel.
Design implication: Use a problem-first opening, immediate context, concise dialogue, and a clear follow prompt; test each version against retention and view-to-follow conversion.
Source: Douyin analytics · Aug 2026
Xiaohongshu — Save Intent and Reference Value
Signal: 4,668 followers and 10.7K likes and saves. Vancouver was the largest audience location.
Interpretation: Strong save behaviour and geographic concentration suggested demand for practical, locally relevant guidance for newcomers in Canada.
Design decision: Prioritize a Canada-focused collection while retaining transferable topics, then compare local scenarios by saves, engagement, and repeat use.
Source: Xiaohongshu analytics · Aug 2026
Top-performing Video — Topic Validation
Signal: 86K views, 2,157 likes, 1,314 saves, and 466 shares.
Interpretation: A familiar, high-stakes situation with an unexpected outcome created a curiosity gap, while strong saves and shares indicated preparation value.
Design implication: Develop more expectation VS reality scenarios with progressive resolution, then measure retention, completion, and view-to-follow conversion.
Source: Douyin analytics · Aug 2026
What I learned
The most important design decision was to treat content, product and production as one connected system. AI made the work faster, but the quality came from context, judgment and iteration. The next phase is to validate which scenarios create the strongest learning value and translate audience interest into sustained product use.
Source: Douyin and Xiaohongshu Creator Analytics, August 2026. Platform follower counts may include overlapping audiences.