Highlighting & Categorizing
I designed a feature for conversational interfaces that allows users to highlight specific answers and categorize them.
Client
Appcent
Services
Benchmarking
Brainstorming
Prototyping
Date
January 2024 - March 2024
Link
Download App
mail Extension
I designed an AI-powered mail extension for mail.
I designed an AI-powered mail extension for mail.
Mail Extension
Problem
Information overload in conversational chatbots leads to forgetfulness after multiple interactions.
Information overload in conversational chatbots leads to forgetfulness after multiple interactions.
Information overload in conversational chatbots leads to forgetfulness after multiple interactions.
Information overload in conversational chatbots leads to forgetfulness after multiple interactions.
After wandering around some GenAI projects and doing some research with users I detected that users engaging with conversational chatbots often encounter challenges in organizing and retaining relevant information provided during interactions.
Without a mechanism for capturing and categorizing key details, users may struggle to recall important information later, leading to frustration and inefficiency. Additionally, chatbots lack the ability to adapt responses based on user preferences and interests, resulting in a less personalized user experience.
Benchmarking
Analysing what already exists
Once I start a project, I usually like to observe and study what already exists on the market. This helps me get a deeper idea of what sort of functionalities do exist.
The goal is to understand the competition better and to asses how can we differentiate in terms of product and functionalities.
Overview
Designing an AI mail extension
In my tenure with Appcent's AI team, I've contributed to various AI-related projects, including designing Safari extension, designing a text-based interface or introducing new features. One project involved crafting an AI-powered mail tool extension for AppleMail.
My Role
Finding
functionality
As a UX/UI designer for this project, my responsibilities were twofold: understanding the technical constraints and meeting user needs to ensure the extension was both user-friendly and valuable.
I firmly believe that AI tools must serve a genuine purpose for users; otherwise, they risk being perceived as redundant.I’m a firm believer that AI tools can be very powerful, as long as they provide a real purpose to users, if not they are useless.
Timeline
January 2024 - April 2024
Role
UX Designer
Engineering
Questions appeared
-How would a user interact with the mail extension?
-In what situations users can find the extension useful?
-Do users use text-based interfaces platforms to reply mails?
Those and many more questions popped up, when joining the project, which means the project required a deep understanding of users in that matter.
Design Considerations
Key findings helped to articulate design decisions
Key findings from our research underscored the importance of context, simplicity, and integration within the mail app. Each email carries its unique context, demanding tailored responses.
Due to technical constraints, we focused on enabling feasible tasks within the extension, emphasizing simplicity to minimize friction. Integrating seamlessly into the mail app was crucial to maintain user familiarity and efficiency.
Additionally, providing users with the flexibility to adjust the tone of their responses ensured a personalized experience.
Solution
Generating responses with adaptive tone
With these insights in mind, I crafted a streamlined and intuitive mail extension that adds genuine value to users' email workflows. The extension offers simplicity and purpose, empowering users to respond effectively while retaining flexibility and control over their messages.
My role
Empowering User Engagement and Personalization
Empowering User Engagement and Personalization
I have developed a solution that enables users to highlight answers or pieces of answers within chatbot conversations and categorize them into themed folders or categories. When users come across information they find valuable or relevant, they simply select the text they wish to highlight and choose a category to store it in. These categories can be pre-defined or customized by the user based on their specific interests.
Furthermore, I leverage this highlighted and categorized information to enhance the conversational chatbot's ability to curate responses and tailor future interactions to the user's preferences. By analyzing the user's saved categories and highlighted content, the chatbot gains insights into the user's interests and preferences, allowing it to provide more personalized and relevant responses over time. This approach also enables user profiling, as the chatbot learns more about each user's preferences and behavior through their interactions with the highlighted content.
