Add new item
Capture garment details with imagery, descriptions, and colour selection.
Drobe
A cross-platform capsule wardrobe app that reduces decision fatigue and encourages sustainable fashion habits through colour-first outfit planning and offline-first architecture.
Overview
Drobe is a mobile application designed to help users build a sustainable capsule wardrobe by digitising their clothing, planning outfits via a colour-first workflow, and generating harmonious colour palettes.
Built with an offline-first architecture using Hive for local persistence, the app remains fully functional without internet connectivity — respecting user privacy and device constraints.
Developed as my final-year Computer Science project, Drobe demonstrates end-to-end product thinking — from research-informed design decisions to production Flutter implementation.
Source Code
Explore the full Flutter implementation and run the app locally via the public GitHub repository.
Problem
Users often own an excess of clothing but still feel they have "nothing to wear." The challenge: help users reduce decision fatigue, limit overconsumption, and plan outfits that align with their personal style — without adding friction or complexity.
Research & Insights
Competitive & literature-informed research
- Reviewed existing wardrobe apps (Stylebook, Cladwell, Pureple) to identify gaps in decision-fatigue support, sustainability nudges, and offline flexibility.
- Analysed app store feedback to understand common user frustrations: complex interfaces, feature overload, and reliance on cloud connectivity.
- Anchored the problem in decision fatigue theory and capsule wardrobe research showing reduced stress and more intentional consumption with constrained choices.
- Applied colour harmony principles (monochromatic, complementary, analogous) to support mix-and-match behaviour in capsule wardrobes.
What that changed in the product
- Filters and saved outfits became core "decision load reducers".
- Palette selection became the entry point to outfit creation.
- Offline-first became a must-have requirement, not a nice-to-have.
Design Rationale
Design principles
- Reduce decision fatigue: narrow choices via colour + category filtering and reusable saved outfits.
- Encourage sustainable habits by default: increase awareness of owned items and support rotation/reuse without "preaching".
- Offline-first & privacy-respecting: wardrobe data stays on-device; the app remains useful without connectivity.
Key UX decisions (and why)
- Colour-first outfit building: selecting a palette first encourages cohesion and makes capsule dressing easier.
- Structured outfit slots over drag-and-drop: better for small screens and reduces interaction complexity during planning.
- Background removal: cleaner tiles reduce visual noise and help users "see the outfit" faster.
Prioritisation
Using the MoSCoW method to define scope and prevent feature creep:
Must-have
Add/edit/delete items, save outfits, full offline use, colour filtering.
Should-have
Background removal, palette generation, stronger search/filtering.
Could-have
Laundry tracking, daily outfit suggestions.
Won't-have (yet)
Seamless cloud sync + social/community features (deferred to protect MVP scope).
This prevented feature creep and kept the build focused on reducing outfit-planning friction.
Key Flows
Wireframing & Prototyping
Designed key screens in Figma with a neutral, minimal aesthetic, starting from low-fidelity Whimsical wireframes.
Palette-first outfit creation
Users select or generate a colour palette before browsing items. This constrains choice and guides cohesive outfit building.
Structured outfit slots
Rather than freeform drag-and-drop, users place items into predefined slots (top, bottom, shoes, accessories) for faster mobile interaction.
Filtering & search
Filters by category, colour, and occasion reduce the number of items to browse at once, lowering cognitive load.
Build & Architecture
Constraints
Needed to run fully offline and remain fast with many image assets.
Technical choices (in service of UX)
- Flutter for consistent cross-platform UI and performance for image-heavy workflows.
- Hive for lightweight local persistence (fast reads/writes; privacy-first storage model).
- MVVM + modular services (e.g., image processing, persistence, outfit logic) to keep the codebase scalable and maintainable.
Testing & Validation
- Exploratory testing, scenario-based testing, checklist testing, regression testing to validate core flows.
- Heuristic evaluation against Nielsen's principles (clarity, feedback, consistency, error prevention).
- Cross-platform checks on iOS + Android; verified offline behaviour (e.g., airplane mode).
Challenges & Solutions
Reliable image persistence
Fixed iOS path issues by storing images in the documents directory and normalising orientation metadata.
Palette extraction accuracy on patterned items
Extract colours after background removal; cap stored colours; allow manual edits.
UI not refreshing after data changes
Integrated Hive updates with Provider state + notifyListeners() for reactive UI refresh.
Responsive layout across devices
Adaptive grids + theme/text scaling checks.
Key Screens
Wardrobe inventory
Browse item collections with search, filters, and quick actions.
Outfit planner
Compose head-to-toe looks using draggable garment cards.
Editable outfit layouts
Fine-tune pieces on the fly with inline edit and delete controls.
Laundry tracking
Move items through laundry states to keep the wardrobe status accurate.
Accessories catalogue
Maintain complementary pieces with rich imagery and metadata.
Fabric care hub
Curated content cards that educate users on garment care.
Inspiration lookbook
Taggable inspiration boards to spark new outfit combinations.
Long-form editorials
Immersive reading experience to reinforce sustainable habits.
Outcome
- Delivered a functional cross-platform prototype that reduces outfit planning friction through structured flows, filtering, and reusable saved outfits.
- Reinforces sustainable behaviour by increasing awareness of owned items and encouraging rotation/reuse.
- Built an extensible architecture for future features like recommendations, analytics, and multi-device sync.
- Awarded a First-Class for innovation, technical depth, and usability.
Project Details
Role
UX Designer, Flutter Developer
Duration
Oct 2023 – Apr 2024
Tools
Flutter, Hive, Figma, Whimsical