Fable Forge
Overview
This project is a concept for a fictional game called Fable Forge. It's a mobile text-based RPG that uses AI-generated content. This case study focuses on the main user flow of creating a new default game. I completed this flow's research, ideation, and UI development.
Business Goal
The primary goal is to build a mobile text-based role-playing game with endless replayability. Success is to be measured by initial Adoption and retention.
Design Problem
The user wants a way to play an open-ended, immersive, emergent, and cooperative storytelling experience where their choices impact the storyline.
Design Solution
A multiplayer, mobile text-based RPG that leverages AI-generated content to keep gameplay fresh and exciting. This case study focuses on the main user flow of creating a new default game.
Discovery
I started the project by imagining what kind of application I'd like to create and what would be ideal for a portfolio piece. I decided on a mobile text-based role-playing game that uses AI for three reasons, one being that a text-based game closely resembles all kinds of applications because it stays in a 2D environment. The second reason is that I'm a big fan of Role-Playing Games, so creating a product I'd use makes it easier to make on my own and keeps it interesting. Fun fact: I had no idea text-based RPGs were even a thing - I discovered them while scouring the app store for ideas.
Lately, I've been fascinated with discovering the potential and limits of generative AI models. I use ChatGPT and Midjourney daily for both professional and personal projects. I used AI even in this project to analyze interviews and cross-reference them with each other to find insights and themes I may have missed on my own. Surprisingly, not surprisingly, it not only uncovered what I found, it returned even more - with references to when each interview participant mentioned it.
This fascination with AI is the final reason for choosing a piece showcasing a great potential application of generative AI. With AI, a game like this comes alive.
Foundational Research
High-Level Touchpoint Map
My first task was getting context and deciding on direction, so I mapped out the potential customer touchpoints. This process would have been a cross-team collaborative exercise if I brought this game to market and directed UX across the entire product.
Research Goals
- Gain insight into operations and how the customer interacts with the product.
- Map out problems/pain points, and select a target moment to focus on.
Competitive Audit
One initial step was to evaluate Fable Forge's feasibility by conducting a competitive audit of existing text-based RPGs on the app store. This audit included the type and quality of Products, Market Position, Competitor's Value Proposition, Potential Opportunities, and Threats from competitors.
Research Goals
- Validate the market demand for the core features of the game.
- Discover untapped opportunities in the market.
Key Learnings
Market Demand Validation
The audit confirmed that there is indeed a market for the game's primary features. This validation reinforces the value proposition of Fable Forge.
Limited AI Competition
The audit found that only one direct competitor uses AI. This presents a unique opportunity for Fable Forge to differentiate itself by leveraging AI-generated content.
Broad Target Audience
The target audience for text-based RPGs is quite diverse, ranging from teens to adults. Some players even explicitly seek text-based RPGs, suggesting a niche but enthusiastic market segment.
User Interviews
Research Goals
- Understand the expectations and desires of potential users for a text-based RPG that utilizes AI-generated content.
- To identify pain points that potential users may experience when playing a mobile text-based RPG.
- Gather additional feedback and suggestions that could guide the development of the game.
Key Learnings
Some helpful information emerged. The gains confirmed the market demand and helped prioritize features. Their pains and concerns brought up new issues regarding monetization models and AI-generated content.
Positive Aspects
AI-Driven Dynamic and Personalized Experiences
Participants appreciated the promise of AI in creating dynamic, personalized narratives that can adapt to player choices. This confirms that the core concept has market appeal.
Collaborative Storytelling
They highlighted the value of shared narrative experiences and player-driven events in the game's story.
Balance Between User Input and AI-generated Content
They emphasized the importance of striking the right balance between user input and AI-generated content
Pain Points & Concerns
UI/UX Concerns
All participants stressed the necessity for an intuitive and polished user interface, including accessibility features. This highlights the importance of investing in UI/UX design.
Monetization Concerns
All four emphasized the need for fair in-game monetization that doesn't hinder gameplay or create a "pay-to-win" environment. Subscription models are seen as viable if they provide clear value.
AI-Generated Content Concerns
Two of the four interviewees expressed concerns about the potential for AI-generated content to become repetitive or generic, pointing to the need for quality control in AI-driven narratives.
Personas
Using the research, I synthesized two personas from some of the insights gained. Additional personas would be beneficial. However, I only made two because of the minor project scope and timing.
Problem Statements
I extracted some problem statements from each persona. However, I didn't need them just yet because of the user flow I decided on in the next step.
Jess
Jess needs new, unique storytelling because she games consistently, and the lack of new content prevents her from playing the same game.
Adam
Adam needs a way to play a remote D&D type of game with his friends because they can't play in person as often as they would like.
Flow & Architecture
Define Target & Direction
Next, I prioritized potential features and chose what to focus on for the rest of this design project.
I decided on the primary user flow of a new player. This flow includes creating a New Game using the starter content, except for a custom profile image and name. It ends with the player starting the game and being shown a loading screen.
Why I Chose This Direction
- It's likely most players' first experience of the game
- Using a default game type and character is the easiest way for a new player to get started
- The custom profile image and name showcase the unique content that can be generated using generative AI techniques.
- It sets the groundwork for creating a custom game type and character during the process.
User Flow
Now that I had decided which flow to design, I made a flow map. I also included some potential gains and pains with opportunity prompts in the form of How Might We's. However, They are assumptions and would need to be evaluated later in usability tests.
In later usability tests, this flow needed to be updated. As you'll read later, some key steps were added to the flow.
Information Architecture
While working on the user flow, I worked on an architecture map of the game. The map includes the primary game flow and the other probable main menu links. It had multiple iterations and served as a way to help visualize the context of the flow.
Wireframes & Prototyping
Prototyping
I designed a prototype in Figma for each wireframe and conducted unmoderated usability tests. The usability tests were conducted at their homes using the Maze application. Following the usability test, they completed a SUS.
Research Goals
- The research goals for each prototype largely remained the same.
- Determine whether or not users can create a new game
- Determine whether or not users can create a character
- Gauge the user's overall opinion of the flow
Wireframes v1
These initial designs for the main flow were created directly in Figma for simplicity and ease of creating multiple variations. When I began my career over a decade ago, paper wireframes did help reduce my need to add detail. However, nowadays, starting from digital saves me time and increases my number of iterations.
Research Insights
Need Character Review
Before starting the game, a final review of the player's character should be included.
New Game Progress Bar
Having some sense of how many more steps it would take to start a new game alleviated some issues of feeling like there are too many steps to create a new game.
Wireframes v2
These wireframes are of higher fidelity than usual because I used ChatGPT and Midjourney to generate text and images. Using AI-generated content made it possible to create concepts quickly with actual content instead of placeholder text and boxes. I also added loading pages.
Research Insights
Changing Default Character Profile Image
Even though players select a default character for simplicity and to get started quickly, they wanted a way to change their character's image.
Distracted By Unrelated & Incomplete Flow
When choosing a realm, the options of "My Realms" and "Community Realms" became a distraction during testing.
Wireframes v3
This next set of wireframes incorporated the changes from the previous usability study and improved the layout. A significant difference was how the screens focused on the primary flow by removing unrelated elements such as main menu links and realm categories. This change made more space for content by reducing the page header height. Another improvement was changing the aspect ratio of character selection. I also adjusted the overlays to make it feel more natural and easier to read.
Research Insights
Ready for UI Design
Players responded positively to the updated design. The findings indicated the design was ready for designing the UI - feedback was primarily graphical based.
UI / Visual Design
Final UI Design
Finally, I designed a UI that aligned with the initial artwork vision. I also created a logo and finalized some placeholder content. An interactive prototype was completed in Figma as well.
This design is the first UI that would be tested with players and iterated on. However, because of time constraints, this is the final UI design.
UI & Artwork Vision
- A stylized UI
- Should "feel" like a storybook
- Elements of Adventure
- Does not reflect a specific game theme or time period.
- The style must work with various game themes, such as high fantasy and sci-fi.
Design System & Style Guide
A basic design system and style guide were created in Figma.
Post-Mortem
The game seems to fit the market. Although, true success needs to be gauged post-launch. The biggest challenge was working solo. It slowed the design process, and the lack of internal review most likely resulted in missed opportunities and solutions.
Next Steps
This concludes the case study of this project. Although only a fraction of the entire game, it provides a solid base for continued research and development.
Taking the project to launch requires continued design iterations, development, and implementation, as well as ongoing evaluation and iteration post-launch to succeed.
Continued Design Improvements
- Finish designing other Users
- JourneysContinued User Testing & Visual Design
- Finalize Design System & Prep for Development
Development & Implementation
- Content Creation
- Develop Game
- AI Content Generation via OpenAI API &/or internal generative AI system
- Finalize Business Model
- Marketing
- Launch
Evaluation & Iteration
- Measure performance against KPIs
- Improve the game based on user feedback