How I Designed and Built Washcast — A Weather-Aware Laundry Assistant That Makes Everyday Life Easier
By Anuoluwapo Sebiomo — Designer, Developer, and the Person Who Kept Forgetting Laundry Outside When It Rained
"Should I hang laundry now?"
That simple question was the spark.
Like many others living in the UK with limited indoor drying space, I found myself constantly battling unpredictable weather forecasts, wet clothes, and repeat wash cycles. Despite access to multiple weather apps, none were telling me what I actually needed to know: Can I do laundry now and expect it to dry?
That insight led me to design and build Washcast, a mobile app that gives actionable, location-aware laundry advice powered by real-time weather data and AI-generated guidance.
The Problem
Weather apps tell you if it's going to rain.
They don't tell you if your laundry will dry.
The Gaps I Observed:
Generic forecasts lacked laundry relevance: "Partly cloudy" isn't helpful.
No consideration of fabric drying times.
Laundry planning across multiple days was guesswork.
Weather data ≠ Advice. People don't want to interpret data—they want a confident decision.
My Role
I played the roles of:
Product Strategist – Defining the value proposition and positioning.
UX Designer – Designing for simplicity, trust, and clarity.
iOS Developer – Building the app using SwiftUI and multiple APIs.
Systems Thinker – Architecting data flows, caching, and failover strategies.
This was a solo project completed in ~8 hours of focused design and build time.
Research & Insights
Talking to Users
I conducted short interviews and observations with friends and online communities. Some recurring themes:
"I don't care about UV index—I just want to know if my jeans will dry."
"I always forget to check the weather before doing laundry."
"It rained even though the app said it wouldn't. Can't trust it."
"Would be great if an app just told me when to wash."
These frustrations pointed me to the core need: people want a decision, not data.
Design Approach
3-Second Rule:
Users should know what to do within 3 seconds of opening the app.
I embraced a "zero-friction" UX philosophy:
No onboarding.
Auto-detect location.
Immediate advice at launch.
Every design decision hinged on contextual clarity, visual simplicity, and trust.
The MVP
Washcast's MVP had three core principles:
1. Intelligence
• AI-generated advice that accounts for time, forecast, and fabrics.
• Forecast-based planning with laundry-specific language.
2. Precision
• GPS-accurate location advice (not just "Manchester").
• Real-time adjustments as weather or location changed.
3. Clarity
• One main call-to-action: "Can I hang laundry now?"
• Supporting sections: forecast scroll, fabric tips, and location info.
Key UX Decisions
📱 On Launch: Instant Clarity
- • Show actionable recommendation immediately.
- • Show current location to build trust.
📆 Planning View: Scrollable Forecast
- • Horizontal scroll mimicking a mental calendar.
- • Tap into future days for advice previews.
🧺 Advice Layout: Three-Paragraph Structure
- • What to do now.
- • When and how to dry.
- • Tips for specific fabrics or backup plans.
Building Washcast
I built Washcast using:
Layer | Tech/Tool |
---|---|
Frontend | SwiftUI (iOS 15+) |
Weather API | OpenWeather (current & forecast) |
Location | CoreLocation + Google Places Search |
AI Advice | OpenAI GPT-4 for fabric-aware guidance |
Data Storage | UserDefaults + File-based Caching |
Notifications | UserNotifications framework |
Dev Challenge Highlights
1. Forecast Failures at Launch
Problem: The app didn't load forecasts correctly on cold start.
Fix: Refactored init flow to unify weather-fetch logic across launches and search.
Lesson: Initialization inconsistencies cause silent failures—unify data pipelines early.
2. Ambiguous Location Bugs
Problem: Users in "Denton, UK" got data for "Denton, TX" 🤦♂️
Fix: Constructed location strings using City,CountryCode for API accuracy.
Lesson: Don't trust third-party location matching. Be explicit.
3. Cache Confusion
Problem: Advice persisted across locations due to stale cache.
Fix: Cache invalidation on location change with unique keys per coordinate.
Lesson: Smart cache is only as smart as your invalidation logic.
4. Broken Forecast Data
Problem: JSON decoding failed because the visibility field was sometimes missing.
Fix: Made it optional and added defensive parsing.
Lesson: External APIs are messy. Code like you expect it to break.
Key Metrics
Metric | Outcome |
---|---|
Crash Rate | 99.9% crash-free |
Weather Load Time | Under 1 second (after optimizations) |
API Cost Reduction | 70% via intelligent caching |
User Feedback Loops | 3 testing rounds with direct changes |
WCAG Accessibility | AA compliant |
Product Learnings
1. Solve One Pain Exceptionally Well
The temptation to add timers, washing instructions, and cycle tracking was strong. But solving "When should I hang laundry?" really well made the app useful from day one.
2. Build Trust Through Precision
Users will never forgive bad advice about rain. So I invested in:
- • GPS-based precision
- • Forecast checks on location change
- • Conservative recommendations when unsure
3. Don't Just Show Data. Interpret It.
Raw weather info wasn't useful. I layered on AI that interpreted:
- • Fabric type
- • Timing urgency
- • Drying probability
And framed it in plain, helpful language.
Monetization Strategy
💵 Why £3.99 One-Time Purchase?
I evaluated several models:
Model | Pros | Cons |
---|---|---|
Free + Ads | Low barrier to entry | Distracting, low trust for utility app |
Subscription (£0.99/month) | Recurring revenue | Overkill for occasional-use app |
Annual (£5/year) | Reasonable long-term value | Adds cognitive load vs. one-time buy |
One-Time (£3.99) | Simple, honest pricing — pay once, use forever | Lower revenue ceiling |
Decision Rationale:
- • Users perform laundry weekly — not daily — so a monthly sub felt excessive.
- • One-time pricing aligned with the app's utility nature.
- • At £3.99, it's priced low enough to feel like a "no-brainer" utility, while covering API costs.
What I'd Improve Next
- ✅ Widgets: Quick glances from Home Screen
- ✅ Apple Watch support: For when you're already loading the washer
- ✅ Custom preferences: E.g., "I always wash heavy cotton"
Final Thoughts
Washcast isn't a flashy product.
It's not solving a billion-dollar problem.
But it solves a real, daily pain with clarity, trust, and thoughtfulness.
In a world of overwhelming information, Washcast gives people a confident answer to a simple question.
And sometimes, that's all you need.
TL;DR
- • Designed and built Washcast, a weather-aware laundry assistant for iOS
- • Combined OpenWeather, OpenAI, and GPS data for AI-powered drying advice
- • Solved edge-case bugs with location disambiguation and forecast caching
- • Prioritized immediate value delivery and frictionless UX
- • Resulted in a crash-free, fast, and accessible utility people actually use