Skip to content

Repository files navigation

💧 FlowSpring

The Reporting Infrastructure for SDG 6

Turning every smartphone into a water-safety sensor — and every citizen into a verified reporter.

SDG 6 Status Stack License

Live Demo · The Problem · How It Works · Architecture · Traction · Business Model · Roadmap · Team


⚡ The 10-Second Pitch

Only 10% of India's monitored water bodies meet good ambient water quality (UNEP, 2023). The barrier isn't technology — it's a broken information relay between the people who see contamination and the authorities who can fix it. FlowSpring removes every link in that chain. One report. Ninety seconds. Zero intermediaries.


🌍 The Problem

Nearly 2 billion people worldwide lack access to safely managed drinking water. In rural India alone:

Statistic Source
163 million Indians lack clean water access near their homes WHO/UNICEF JMP
70% of India's surface water is contaminated NITI Aayog, 2023
Only 10% of monitored water bodies meet good ambient water quality UNEP GEMS/Water, SDG Indicator 6.3.2
29% of India's domestic wastewater is safely treated SDG Indicator 6.3.1

These numbers persist not because the problem is invisible — communities see it, smell it, and live with its consequences daily — but because there is no direct channel between a citizen who spots contamination and the water authority responsible for fixing it.

The Relay Problem

Today, a contamination report travels like this:

Citizen  →  Neighbour  →  Local Councillor  →  NGO Field Officer  →  Authority
   |              |                |                    |                |
 sees it      tells someone    passes it on        logs it (maybe)   acts (maybe)

Every arrow in that chain is a point of failure, delay, or silence. The original reporter has no record, no visibility, and no way to hold anyone accountable. Authorities, in turn, have no documented proof they were ever notified — so ignoring a report costs them nothing.

This is not a water scarcity problem. It is an accountability infrastructure problem.


💡 The Solution

FlowSpring inverts the model. Citizens become the sensors. Authorities become the responders.

A community member who spots contaminated water opens the app, taps four times, and within 90 seconds their GPS-tagged, photo-evidenced, severity-rated report lands simultaneously in:

  • 📍 The district water board's live dashboard
  • 🤝 The WaSH NGO partner's daily digest
  • 🗺️ The public community map, visible to every neighbour

No intermediary. No silence. No way to pretend it wasn't seen.


📱 Live Demo

🔗 flowspring.lovable.app

Scan to try it on your phone:

Home Report Map Status Tracker
Live incident map, district safety score, one-tap reporting Icon-grid form, GPS auto-fill, voice input, severity selector Colour-coded pins, real-time updates, filterable 5-stage pipeline with timestamps & accountability trail

🔧 How It Works

The 90-Second Report Flow

graph LR
    A[Open App] --> B[Tap Report]
    B --> C[Select Type & Severity]
    C --> D[GPS Auto-Detected]
    D --> E[Optional Photo / Voice]
    E --> F[Submit]
    F --> G((Live on Map<br/>+ NGO Dashboard<br/>+ Water Board))
Loading

The 5-Stage Accountability Pipeline

Every report moves through a transparent, timestamped pipeline — the core innovation that no SMS hotline or complaint form provides:

📋 Submitted  →  🔍 Under Review  →  ✅ Verified  →  🔨 Action in Progress  →  ✔️ Resolved
  • Each transition is timestamped and tagged with the responsible party
  • Community peer-review: 3 independent confirmations auto-upgrades a report to Verified
  • Resolution accountability: 2+ "still exists" votes auto-reopens a falsely-closed report
  • Escalation: reports pending 7+ days are automatically flagged — publicly, visibly, permanently

This isn't a suggestion box. It's a public accountability ledger.


🏗️ Technical Architecture

FlowSpring is a production-grade, full-stack application — not a prototype.

┌─────────────────────────────────────────────────────────┐
│  FRONTEND — React 18 + TypeScript + Tailwind CSS         │
│  Framer Motion · Leaflet/OpenStreetMap · i18next         │
└───────────────────────┬─────────────────────────────────┘
                         │
┌────────────────────────▼────────────────────────────────┐
│  BACKEND — Supabase                                       │
│  • PostgreSQL (with Row Level Security on every table)   │
│  • Auth (email/password, role-based access control)      │
│  • Realtime (WebSocket subscriptions for live map)       │
│  • Storage (incident & resolution photo uploads)          │
│  • Edge Functions (status-triggered notifications)        │
└─────────────────────────────────────────────────────────┘

Key Engineering Decisions

Decision Why It Matters
2G-network compatible Built for rural connectivity, not urban broadband — <1MB per report submission
Offline-first report queue Reports cache to localStorage and auto-sync on reconnect — zero data loss
Secure user_roles table (not a column on profiles) Prevents privilege escalation; field agents, NGO partners, and admins are cryptographically separated from regular users via a has_role() security-definer function
Severity ≠ Resolution colour system Red/orange/yellow = active severity. Green = resolved only. Prevents a critical UX failure mode where colour blindness or quick scanning could mistake a low-severity active incident for a fixed one
Web Speech API voice input No paid API — works in Hindi, Telugu, and English for low-literacy users
SDG 6.3.2-formatted CSV export NGO dashboard exports data in the exact schema India needs for its UNEP submission — turning citizen reports into national policy evidence

Database Schema (simplified)

incidents              -- core report: location, severity, type, status, photos
incident_status_history -- full audit trail: who changed what, when, why
verifications           -- community peer-review votes
resolution_feedback     -- "confirmed fixed" / "still exists" votes
notifications           -- persistent, DB-backed (not in-memory) feedback loop
user_roles               -- secure RBAC: user / field_agent / ngo_partner / admin

Every table enforces Row Level Security — no client can read or write data it isn't authorized for, verified at the database layer, not just the UI.


🎯 Why This Wins (Decision Matrix)

We didn't assume FlowSpring was the right approach — we proved it against two alternatives using a weighted evaluation framework:

Criteria Weight FlowSpring ClearPipe (SMS-only) WaterLens (IoT sensors)
SDG Impact Potential 30% 5 3 4
Technical Feasibility 25% 5 4 2
Scalability & Partnerships 25% 5 2 3
Cost Effectiveness 20% 4 4 2
Total Score 🏆 4.80 / 5.00 3.20 2.85

FlowSpring outperformed an IoT sensor network on cost and feasibility — because hardware doesn't scale in rural India, but smartphones already do. GSMA reports rural Android penetration in India now exceeds 42% and climbing.


📊 Traction & Validation

  • Decision-matrix validated against 2 competing approaches (weighted scoring, not guesswork)
  • 80 ASHA community health worker beta partners identified for 2-district pilot
  • Full production app deployed — not a wireframe, a working platform
  • SDG 6.3.2-compliant data export built and tested
  • ✅ Design validated through low-fidelity → high-fidelity → production pipeline, with a real UX bug caught and fixed pre-launch (the severity/resolution colour conflict)

Target Metrics (14-Month Goal)

Metric Target
Active users 6,500
Verified field reports 3,800
District water board partnerships 3
WaSH NGO partnerships 2
Pilot districts 2 (Rajasthan, Telangana)

💰 Business Model

FlowSpring is built as a sustainable social enterprise, not a grant-dependent project.

┌──────────────────┬─────────────────────────────────────────────┐
│  REVENUE STREAM   │  WHY IT WORKS                                │
├──────────────────┼─────────────────────────────────────────────┤
│  Government SDG    │  India's Jal Jeevan Mission actively funds  │
│  grants             │  tools advancing SDG 6 — measurable impact  │
│                     │  data makes FlowSpring grant-eligible       │
├──────────────────┼─────────────────────────────────────────────┤
│  NGO dashboard fees │  WaSH NGOs need SDG-compliant impact data   │
│                     │  for their own funding reports — we are     │
│                     │  the data infrastructure they don't have    │
├──────────────────┼─────────────────────────────────────────────┤
│  Research dataset   │  Verified, geo-tagged water quality data    │
│  licensing          │  is a genuine public health research asset  │
└──────────────────┴─────────────────────────────────────────────┘

Cost structure is deliberately lean: free-tier cloud infrastructure (Supabase, OpenStreetMap), a small core engineering team, and field onboarding costs scoped to ASHA worker training rather than paid acquisition.

Total addressable market: India alone has 600,000+ villages. FlowSpring's architecture — 2G-compatible, multilingual, zero-hardware-dependency — is designed to scale to all of them.


🗺️ Roadmap

  • Phase 1 — Core platform: reporting, mapping, status pipeline, real-time sync
  • Phase 2 — Community verification, escalation system, NGO dashboard
  • Phase 3 — Voice input, role-based field agent tools, offline-first architecture
  • Phase 4 — WhatsApp bot for users without smartphones (WABA API)
  • Phase 5 — Water Source Registry — persistent health tracking for named wells, tanks, and standpipes
  • Phase 6 — District Water Safety Score — live 0–100 metric feeding directly into state SDG 6.3.2 reporting
  • Phase 7 — Low-cost IoT turbidity/pH sensor integration into the same incident pipeline
  • Phase 8 — State-level scale-up: Rajasthan → Telangana → national

🎨 Design Philosophy

FlowSpring is engineered for its actual user: a rural, potentially low-literacy, low-bandwidth, first-time smartphone user — not a Silicon Valley power user.

  • Icon-first, not text-first — every action is recognizable without reading
  • Colour with redundancy — colour is never the only signal; every status also has an icon and label (WCAG AA compliant)
  • 3-tap reporting — type, severity, submit. Everything else is optional or automatic
  • 2G-first performance budget — every screen, every asset, built for the network conditions of rural India, not a campus WiFi demo

🛠️ Tech Stack

Layer Technology
Frontend React 18, TypeScript, Tailwind CSS, Framer Motion
Backend Supabase (PostgreSQL, Auth, Realtime, Storage, Edge Functions)
Maps Leaflet.js + OpenStreetMap (zero API key cost)
Voice Web Speech API (English, Hindi, Telugu)
i18n i18next
Deployment Lovable

🙏 Acknowledgements

Built as part of Innovation and Design Thinking, Spring 2026, BITS Pilani Dubai Campus, under the guidance of Dr. Chakradhar Iyyuni and Dr. Padmanabhan Seshaiyer. Special thanks to our ASHA health worker field partners for the insight that shaped every design decision in this product.


The relay is broken. FlowSpring fixes it.

SDG 6 · Clean Water and SanitationCitizens as Sensors. Authorities as Responders.

About

Resources

Stars

2 stars

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages