1.Brainstorming process

1.1 Brainstorming Methodology

After analyzing the interviews and journey maps for the UniMarket solution, the team conducted a brainstorming session focused specifically on identifying smart features that could reduce user effort and friction when participating in sustainable clothing practices.

The goal of this brainstorming was to explore how intelligent systems could support users in decision-making, reduce perceived effort, increase trust, and improve overall usability.

Problem Framing

From the research phase, we identified that users are willing to participate in sustainable clothing exchange, but often avoid it due to:

• Perceived effort (taking photos, writing descriptions, searching)

• Lack of trust in second-hand items

• Difficulty finding items that match their style

• Low motivation to actively engage

These insights guided the brainstorming toward automation, personalization, and intelligent assistance.

Individual Idea Generation

Each team member independently proposed ideas using sticky notes, focusing on how AI or smart systems could simplify the user experience. The ideas emphasized reducing manual tasks, improving trust, and supporting user decisions.

Key ideas generated during this phase included:

• AI-based photo recognition to analyze clothing items

• Automatic tagging (size, color, style, category)

• See how the clothes look on you (outfit inspiration)

• Smart recommendations based on user preferences and behavior using tags

• Image-based search (“shop the look”)

• Quality validation through image analysis

• Rating and badge systems supported by AI insights (app mascot that tells you how good of a user you are)

Evidence Idea Clustering

After idea generation, the smart features were grouped according to their purpose:

• Automation: photo analysis, automatic tags

• Personalization: recommendations, preference-based filtering

• Trust & Validation: quality recognition, ratings, badges

• Discovery: image-based search, “shop the look” functionality

This organization helped clarify how each smart feature addressed a specific user pain point.

2. Decision Process – Divergence and Convergence of Smart Features

2.1 Divergence Strategy

During the divergent phase, all proposed smart features were considered without restriction. The team explored both simple and advanced AI solutions, including high-effort ideas (e.g., full outfit generation) and low-effort ones (e.g., auto-tagging).

After completing the brainstorming process, we moved into a structured decision-making phase based on the divergence and convergence strategy taught in class.

During the divergent phase, all generated ideas from both problem spaces were considered equally, without filtering or prioritization. This allowed the team to explore a wide range of possible solutions and avoid prematurely discarding innovative or unconventional ideas.

2.2 Convergence Strategy

To converge toward feasible and valuable smart features, the team applied the four-category convergence technique taught in class:

The Rational Choice:

• Automatic clothing tagging (size, type, color)

• Preference-based filters

The Most Likely to Delight:

• Smart recommendations

• Image-based search (“shop the look”)

• Interactive Mascot that gives small rewards to users

The Darling:

• AI-based quality recognition and approval

• Outfit combination suggestions

The Long Shot:

• Fully automated outfit generation

• Advanced similarity search using deep image models

Each team member participated in voting across these categories. This process allowed the team to balance technical feasibility, user value, and innovation.

2.3 Selection of Final Smart Features

Based on the convergence results, the team selected a set of core smart features to integrate into UniMarket:

• AI-based photo recognition for clothing analysis

• Automatic tagging and categorization

• Smart recommendations based on user preferences

• Interactive app mascot that provides feedback with small rewards

These features were selected because they:

• Reduce effort for “lazy but willing” users

• Increase trust in second-hand clothing

• Improve discovery and personalization

• Align with the technological scope of the project

By focusing on smart features, UniMarket transforms sustainability into a low-effort, intelligent, and attractive experience, encouraging participation without demanding active commitment from users.

Empathy Maps

The following table has the links for each empathy map that correspond to a different prospective user:

Member	Empathy Map
Felipe Mesa	Empathy Map
Joseph Linares	Empathy Map
Isabella Caputi	Empathy Map
Mariana Pineda	Empathy Map
Giuliana Volpi	Empathy Map
Sofia Vasquez	Empathy Map

Personas

The following table has the links for each persona:

Member	Persona Made	Diagram
Felipe Mesa and Mariana Pineda	Seller	Diagram
Isabella Caputi and Giuliana Volpi	Buyer	Diagram
Joseph Linares and Sofia Vasquez	Volunteer	Diagram

Description of Solution

The proposed solution is UniMarket, a university-based second-hand clothing platform that uses smart, AI-driven features to reduce the effort and friction associated with sustainable clothing practices. The solution addresses key barriers identified during research, such as perceived effort, lack of trust, low motivation, and difficulty finding items that match users’ style. UniMarket focuses on making sustainability convenient and accessible for university students.

The platform integrates AI-based photo recognition to automatically analyze clothing images and generate tags such as size, color, category, and style, reducing the manual work required from users. In addition, smart recommendations and image-based similarity search help buyers easily discover relevant items based on their preferences and visual cues, improving the overall browsing and decision-making experience.

To increase motivation and trust, UniMarket introduces an interactive app mascot that provides feedback on users’ sustainable actions. The mascot evaluates behaviors such as selling, buying second-hand, or donating clothes and responds with messages, levels, and small rewards. This gamified approach encourages consistent participation, helps users feel recognized for their efforts, and makes sustainable behavior more engaging and rewarding.