Selene Dominguez-Diaz

 

​​

​

​

​

​

​

​

​

​

 

 

 

 

 

 

 

 

 

 

Project Brief 

 

Key Deliverables

 

 

Our mission is to design a user-centric experience for a Discord-based chatbot. At the end of the design phase, our team will present comprehensive findings and actionable recommendations by:

​

• Conducting interviews with 5-10 users to gather practical usability insights

• Compiling user personas to better understand our target audience

• Documenting the entire design process from start to finish

• Analyzing user feedback to identify significant usability challenges

• Offering strategic, actionable recommendations for improved usability

• Presenting the results and collecting feedback for further iterations

​

 

 

 

Research Plan​:

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​​​​​

 

​

​

​

​​​

User Interview Overview

 

 

The users interviewed can be categorized broadly, though there are distinctions based on age, experience, and socio-economic status. Those with higher risk tolerance typically had lower financial literacy and diverse backgrounds (e.g., car sales person and expats in Japan). At the same time, risk-averse individuals tended to be financially stable (e.g., a husband trading in six digits with a penchant for holding TSLA). However, it's important to note that the small sample size limits the ability to make broad generalizations.

 

When discussing AI and automation, two short-term traders expressed significant interest in the potential utility. One trader, who owns nine monitors, highlighted the benefit of an AI synthesizing data to save time. The other trader, felt the investment system was rigged against them and believed an AI could democratize the investment landscape.

​

​​

​

​​​

User Personas base on Interviews:

​​​

​

​

​

​

​

​

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

​

​​

 

The Bot

 

Here’s how the bot measures up against others

on a heuristic level:

​

​

​

​

​

​

​

​

 

 

 

​​​Red Routes & User Flows

​​

​

​​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​​​​

Low Fidelity Prototype

​

In developing the chatbot, we identified two primary considerations that needed to be addressed before moving forward:

1. Given the text-based nature of the project, it was essential to explore visualization tools beyond Figma to accurately represent interactions.

2. Discord's inherent limitations required a thorough investigation into the capabilities of a bot within the Figma framework.

To tackle the first concern, we opted to use Google Docs to create a basic chat layout, ensuring a clear and effective representation of the conversation flow. The second concern will be addressed in detail during the High Fidelity prototype phase, where we will assess the bot's functionality within the specified constraints.

​

​

 

​Stock Inquiry:

​

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Onboard - Establish Risk:

​

 

​

​

​

​

​

​

 

​

 

Updates to the community:

​

​

​

​

​

​

​

​​

​

​

​

​

 

​

​

High Fidelity Prototype

​

Addressing the second challenge—visualizing Adabot within the limitations of a Discord bot—required the team to dive into some research on Discord bot functionalities and what’s expected within this environment.

Discord bots rely on discord.js to interface with the Discord API, which means our design had to adhere to specific guidelines. We needed to use embeds to structure data effectively and leverage code blocks for creative text formatting, including the use of colors.

To meet these requirements, we identified the Embed Generator as the most suitable tool for crafting the prototype that follows:

​

​

​

​​​

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

How might we inform risk-averse users of their options?

​

​

Recommendations

 

Ultimately, the decision on which findings to implement lies with the relevant stakeholders and team members. Given the constraints of time and resources, the recommendations are divided into two categories:

 

​

Prescriptions

 

These are the essential, easily implementable actions that can be maintained throughout the project's lifespan without much modification.

​

1. Implement the Systems Usability Scale: Begin collecting data immediately upon going live and continue to do so.

2. Run A/B Tests: Conduct these tests as the system evolves to ensure future UX teams have solid metrics for ideation.

3. Document Bot Operations: Create comprehensive documentation on how the bot operates and make it accessible to future UX teams.

4. Communicate Bot Status: Inform users when Ada is processing information, even if the response time is minimal. Users should know what is happening, even during a short wait.

5. Consistency in Responses: Ensure AdaBot provides consistent answers to similar questions, regardless of wording. As a financial advisor bot, the responses should be reliable and uniform.

6. Direct and Actionable Advice: AdaBot should provide clear, actionable advice in her responses.

7. Independent Research: AdaBot should conduct her own research. For example, if asked, “What is a good price to buy X stock at?”, she should provide a researched response rather than suggesting the user do their own research.

 

​

​Subscriptions

 

These are important recommendations that require a greater investment of time and energy to implement.

1. Implement Advanced Features: Invest in developing advanced features that may enhance user experience but require significant resources to implement.

2. Continuous Improvement: Allocate time and resources for ongoing updates and improvements based on user feedback and data analysis.

3. User Education: Develop educational materials and resources to help users understand and make the most of AdaBot's capabilities.

 

By following these recommendations, we can ensure AdaBot provides a reliable, efficient, and user-friendly experience, positioning it as a valuable tool for financial advice and stock market analysis.

​

​

​

Final Product​

​

​

​

​

​

​

 

 

​

​

​

​

​

​

​