ChatGPT
An easy to understand user metrics dashboard for ChatGPT

Timeline
1 Day
Role
Interaction Designer
Visual Designer
Team
Individual
Methodology
User Research, Wireframes, High Fidelity Mockups & Prototypes, Iterations
The Challenge
I designed the dashboard that allows users to view their user metrics for ChatGPT.
In today's time, the pervasive integration of AI across diverse professions, particularly the widespread adoption of chat-based models like ChatGPT and generative AI, underscores the necessity for comprehensive user activity monitoring.
The Goal
To design an easy to understand visually appealing and comprehensive dashboard seamlessly integrated in the ChatGPT app.

User Metrics Dashboard
The Dashboard comes as a pop-up overlay similar to the existing settings tab
Process
Considering limited time, the process began with secondary research on user-friendly data presentation and usability testing of relevant apps. Essential data points to be presented were selected, and low-fidelity wireframes were created for layout clarity. Using the ChatGPT design system, a high-fidelity mockup was developed, ensuring effective data presentation and positive user engagement.
.png)
Research
I started by understanding who are the primary users for ChatGPT from the existing data. I conducted secondary research to understand who are the prime users for ChatGPT.
What is ChatGPT?
An interactive web application that interacts in a conversational way. "The dialogue format makes it possible for ChatGPT to answer follow up questions, admit its mistakes, challenge incorrect premises, and reject inappropriate requests."
Quantitative Analysis
Another finding was that the users came from all sorts of professional backgrounds, thus, I wanted to create a dashboard that was easy to understand, minimal and something that resonates with the targeted user group between 18- 35 years of age.

.png)
What should the dashboard display?
After studying the user group and how ChatGPT works, I listed down some key data points to be included in the user metrics dashboard.
Total conversations
Popular Topics or Keywords
Active Conversations
Positive vs. Negative Responses Ratio
Conversations per Hour/Day/Month/Year
Overall Satisfaction Score
Average Session Duration
Messages per User
Define
To define the key information and criteria for designing the dashboard, I employed the "How might we" method to articulate the problem statement effectively. I identified 3 criteria to guide the dashboard design process. These criteria were carefully chosen to ensure the dashboard's effectiveness in meeting user needs and facilitating user engagement.
Simple & Minimal
Clear and easy to understand, with intuitive navigation and minimal clutter
Relevant & Actionable
Provide insights that users can use to make informed decisions
Visually Appealing
Provide insights that users can use to make informed decisions
How might we?
How might we create a user friendly and easy to understand dashboard of user metrics?
Ideation
After gathering all the necessary data points to be included and narrowing down the criteria to be included in the user metrics dashboard, I started to brainstorm about the way I would present this information. I started to sketch taking inspiration from the existing dashboards on other similar applications like Facebook, Google Analytics, and Apple Health, to name a few.





High Fidelity Prototype Cards
Using the design guide followed by ChatGPT, I designed the Hi-Fi Prototypes in form of cards.
User Flow
I leveraged the existing user flow of ChatGPT to insert the dashboard seamlessly so it doesn't come as a surprise. I used Jacob's law (user's prefer to see the site the same way they've used other sites) to define the user flow and the design of the dashboard.

Design System
I used the existing design system for the dashboard design adhering to established visual elements, such as color schemes, typography, icons, and layout grids, to maintain consistency with the ChatGPT brand identity.




HIGH FIDELITY PROTOTYPE
What I learned
-
Understanding what users require from a metrics dashboard, including the type of information they find valuable and how they prefer it to be presented in a clear and concise manner.
-
Analysing data and identifying key information to be presented for an effective user engagement.
-
Presenting quantitative data in a graphical manner with minimal elements so as to have the focus on information than the design itself.
What I'd do in the future
-
I would experiment with interactive visualizations or customizable dashboards to provide users with greater flexibility in exploring and understanding the data.
-
I'd like to conduct usability testing to get user feedback and iterate on the design.
-
I'd also like pay closer attention to accessibility considerations, ensuring that the dashboard is usable by individuals with diverse abilities.