EWA Test Assignment

EWA is an edutainment app for learning languages. Lessons already run on a roadmap, and the books section was just a library: the user picks what to read, in what order and for which level, and tracks progress on their own. The brief was to replace it with a linear chain of books inside the personal curriculum.

The books are split into CEFR groups (A1, A2, B1 and so on), 4-5 books per level. The user has to see where they are in the course. The brief was one overview screen, with the visual style free as long as it doesn’t feel foreign to the rest of the app. Development complexity wasn’t part of the brief.

I turned the task into 3 hypotheses and drew a concept for each.

1. Familiar pattern. The same navigation as in Lessons (path, progress dots, mascot) cuts adaptation time. Books go in the familiar snake, levels move to a separate “Study plan” screen. Check: A/B on new users, Time to First Read.
2. Gamified journey. A journey map with a final goal and a balloon avatar lifts engagement and Retention 7/30D. Every level is its own island. Check: Retention against the current library, level Completion Rate.
3. Content first. A showcase of shelves where covers dominate shifts focus to the content and lifts Click-Through Rate. Check: CTR of covers, Premium Conversion on locked books.

Designed the UI and built 3 high-fidelity interactive prototypes in Figma. All 3D island assets I generated through AI plugins.

Concept 1, Classic. The conservative one: the whole course on one Study plan screen, a final test at the end of every level.

Concept 2, Modern journey. An interactive map of islands, one per level, with a balloon that moves as the user levels up. An island can be finished, current or locked.

Concept 3, Avant-garde. Interactive shelves with cover previews, so the difficulty grows in plain sight. Inside a group: Continue reading, Details and Reroll.

All the concepts share one set of book states, so the user reads progress from the cover itself. A finished book has a green outline and a “Read again” button. The current one pulses, shows a progress bar and offers “Continue reading”. A locked one is desaturated, with a lock and an inactive “Unavailable” button. The last card of the level is the final exam, available ahead of time via “Take it early”.

The test left several scenarios out of the UI, so I described them as ideas.

• Onboarding in 2 steps. Mark favorite genres with icon tags, then favorite books, either as a cover grid or as Tinder-like swipes. Swipe right is “want to read”, left is “not for me”, up is “already read”. In 30 seconds this gives the data to build the personal roadmap, and the first book more likely matches the taste.
• Reroll. Change a book within the current CEFR level: three reels spin and the user picks one of 3 books. The number of rerolls is limited, extra ones cost coins.
• After reading. A word test at the end of a level. The format has to vary, because closing 100+ words with one mechanic is hard.
• Common word library. A separate view with all the words the user marked in every level and book.
• Book club. Highlight that friends or other people read the same books. It gives a sense of belonging without pressure.

• 3 different versions covering the entire end-to-end reading flow and progression loops to run qualitative UX testing.
• AI-driven 3D asset pipeline used to rapidly self-produce spatial graphics and map out the core “User Journey.”
• 100% autonomous delivery across layout architecture, product hypothesis mapping, and 3 distinct UI iterations within a 2-day deadline.