Role: UX/UI Designer + Brand Designer

Project Type: End-to-end niche marketplace

Timeline: 8 weeks

Industry: Specialty Food & Beverage / Travel-Tech

Tools: Figma · Photoshop · FigJam

The Challenge

The Oportunity

Mainstream navigation and food discovery platforms prioritize data volume over quality curation. For the specialty coffee enthusiast, searching for a good cup of coffee has become a process of sifting through "fast-food chains, and generic breakfast joints which leads to wasted time and a degraded morning ritual.

Travelers and niche coffee enthusiasts crave a zero-friction discovery tool that guarantees quality. By filtering the market to only include specialty roasters, DRIP. connects the high-frequency daily ritual with artisanal businesses, reclaiming the user's time and morning experience.

Problem Statement

“How might we eliminate the search-fatigue of the specialty coffee drinker by creating a curated ecosystem that rewards their high-frequency habit with instant, guaranteed access to specialty coffee?”

User Personas

Pain Points

  • 90% of map results are trivial to specialty coffee drinkers.

  • Users waste 15–20 minutes vetting shops in new cities via reviews.

  • No trusted system exists to filter "specialty" from "commodity" coffee.

  • Users bounce between apps to find, order, and track their daily ritual.

Methodology

Methods:

  • 15 user interviews

  • Competitive analysis of Starbucks, Dunkin, Google reviews, Yelp.

Findings:

  • 86% of respondents stated they would rather walk 15 minutes further for a "Good cup” than get a lower quality cup across the street.

Research

User Insight

Competitive Landscape

User Journey

Persona: Elena (The Displaced Purist)

Scenario: Arriving in a new city for a 9:00 AM meeting and needing to find a specialty coffee shop.

Design Analysis

Screen 1: The "Pure" Entry

I chose a high-contrast, immersive hero image. The goal is emotional resonance.

  • Design Decision: The "Get Started" button is high-contrast terracotta, specifically placed at the bottom "thumb-zone" for immediate interaction.

Screen 2: Curation over Quantity

The home screen doesn't just show "everything." It shows the Best Nearby.

  • Analysis: Categories like "Macchiato" and "Cappuccino" are prioritized over generic "Hot Drinks" or "Coffee" signaling to the user that this app speaks their language.

Screen 3: Technical Transparency

The Detail page highlights what matters to the niche: Bean Type, Roast level, and Size.

  • Analysis: By using icons for temperature and milk, we reduce cognitive load while maintaining the a luxury feel.

Screens 4 & 5: The "Anxiety-Free" Wait

For a traveler, knowing exactly when their coffee arrives is vital for scheduling.

  • Analysis: The tracking map is clean, removing building labels to focus strictly on the delivery path, providing a easy waiting experience.

Learnings

  • Design Finding: I discovered that specialty drinkers value bean origin and roast type as much as convenience.

  • The "Niche" Advantage: By ignoring the mass market, I created a brand language that feels like a "private club" for coffee lovers.

  • Success Metric: I would measure success by the "Time to Order" (aiming for <60 seconds) and "Recurring Usage" (aiming for 3.5x orders per week).

Previous
Previous

ReShop | Buy Better. Waste Less.

Next
Next

Dr. Ana Sofia Cardenas | Medico Esteticista