UofT

2026

QoL changes with PokemonTCGP's collection experience

Role

UX Researcher and UI Designer

Timeline

September 2026

Team

Individual

Platform

Mobile (ver 1.7.2)

Role

UX Researcher and UI Designer

Timeline

September 2026

Team

Individual

Platform

Mobile (ver 1.7.2)

Overview

Pokémon TCG Pocket Loses Nearly Half its Playerbase in August 2026

Pokémon TCG Pocket launched in October 2024 as a free-to-play mobile card collector and quickly became a hit. Within a year, monthly active users fell from about 39 million to 23 million, and DeNA pointed out in improving the collection experience as one way to win players back (Source: Gamebiz.jp).

I chose this subject for my UX in Video Games course, where I benchmarked three collectible card games (Pokémon TCG Pocket, Hearthstone, and Yu-Gi-Oh! Master Duel) to see how their collection experiences help players find the cards they need. This case study picks up where that project left off.

Research

Heavy metadata in need of a clean UI

All three games share the same core components (search, filters, duplicate management, and a similar information architecture). Where they differ is the trade-off: every tool means clutter, while a clean UI means less flexibility.

To dive in deeper into game specifc components for PokemonTCGP, I conducted:

IA and heuristic audit

Evaluated the "My cards" screen's information architecture and conducted a heuristic evaluation

Analyzed 70+ Reddit posts

Ran a thematic analysis to surface recurring player pain points.

Across 100+ observations, affinity mapping showed the same problems resurfacing:

Players bounce between pages to find unowned / related cards

Players search by how a card looks, not by number, so chasing cards they don’t own sends them to workarounds outside the Pokédex view.

Players bounce between pages to find unowned / related cards

Players search by how a card looks, not by number, so chasing cards they don’t own sends them to workarounds outside the Pokédex view.

Basic search methods did not allow for human error

Search assumes players already know the exact name — one typo or a missing space, and they hit a dead end.

Basic search methods did not allow for human error

Search assumes players already know the exact name — one typo or a missing space, and they hit a dead end.

Filter and sort did not match the player's mental model

Sorting and filtering do not fit how players think about their collection.

Filter and sort did not match the player's mental model

Sorting and filtering do not fit how players think about their collection.

Wishlist is not surfaced properly

The wishlist feels disconnected from the collection. Players must navigate to their profile to update it, can’t easily see other players’ tradables, and have limited options for organizing their wishlist.

Wishlist is not surfaced properly

The wishlist feels disconnected from the collection. Players must navigate to their profile to update it, can’t easily see other players’ tradables, and have limited options for organizing their wishlist.

Problem

The search system is not broken. They just don't offer enough flexibility for the unknown.

I visually highlighted pain points from my heuristic evaluation and the community feedback, then looked for where they overlapped. Those overlaps showed me which UI components to prioritize.

Snapshot of the main UI issues found in "My Cards" page and unowned cards view

Ideation

The UI Components

Crazy 8s Sketches (actually…maybe more than 8)

I sketched several versions of the components users struggled with most (search bar, filter drawer, unowned card screen, card dex view), each paired with notes on requirements and ideas.

Notes and sketches for the filter drawer

Designing wireframes that made sense

After the July 2026 update, I re-evaluated my collection toggles against Pokémon TCG Pocket’s new search and filter system. I updated my wireframes to fit its IA direction and added a screen for discovering unowned cards, which my research showed players needed.

Carousel image

[HUD] Increased visibility of search and filter + filter reset

Resetting a search took three taps (open search, clear, OK), and nothing made it obvious that a search or filter was active.

Carousel image

[Search Bar - Drawer] Added search history and search recommendations

Typos and ability-effect searches often hit “no results” dead ends. With search history removed, users are left with date-based options they can’t easily recall.

Carousel image

[Collection Filter - Drawer] Increased visibility of search and filter + filter reset

The original toggles allowed impossible combinations, like favorites and non-favorites at once, which had no effect. I merged each pair into one control toggle and section to save space.

Carousel image

[Collection Filter - Drawer] Increased visibility of search and filter + filter reset

The original toggles allowed impossible combinations, like favorites and non-favorites at once, which had no effect. I merged each pair into one control toggle and section to save space.

Carousel image

[Pokemon Type Filter - Drawer] Added Retreat cost and Stage

My research showed players wanted stage and retreat cost filters for competitive deck building, since other online deck builders offer them and the game doesn’t.

Carousel image

[Pokemon Type Filter - Drawer] Added Retreat cost and Stage

My research showed players wanted stage and retreat cost filters for competitive deck building, since other online deck builders offer them and the game doesn’t.

Carousel image

[Search Results - Drawer] Shows visibility of active filters / available cards

Users had to scroll the full filter list to see what was active. Resetting filters one by one or all at once makes searching faster and more flexible.

Carousel image

[Unowned card] Related card sections in evolution order / same pokemon

Cards are ordered only by expansion number, which does not reflect how players think about their collection. Grouping by Pokémon and evolution line would better match their mental model and improve browsing.

Designs

Current Design

I recreated all of the elements and designed for feasibility and actionable changes within the current neumorphic design system. Catering to both players and DeNA, I made small quality-of-life UI updates rather than huge changes that would disrupt the current design system.

I recreated all of the elements and designed for feasibility and actionable changes within the current neumorphic design system. Catering to both players and DeNA, I made small quality-of-life UI updates rather than huge changes that would disrupt the current design system.

I recreated all of the elements and designed for feasibility and actionable changes within the current neumorphic design system. Catering to both players and DeNA, I made small quality-of-life UI updates rather than huge changes that would disrupt the current design system.

Lessons

UI changes catered for the game vs the players

One important thing I learned is that games are designed around their own goals, like mystery, collecting, and progression, which don't always match what players need to get things done. In a typical app, the fix for a hard-to-find item is to surface them directly. Here, showing all unowned art would have broken the mystery that makes collecting fun.

So instead of adding more tools, I looked for changes that served both sides. The main feature that demonstrates this is the brief art preview that flips back to the card back, so the Pokédex stays in context without spoiling the surprise, and removing toggles that conflicted with the game's existing system. The lesson was that good UI in a game isn't just about removing friction. It's about knowing which friction is part of the experience and which is just in the way.