RCC Dashboard and Engagement Tool

Project Overview 📊

The RCC Engagement Analytics Dashboard is a full-stack analytics platform designed to help the Responsible Computing Club (RCC) understand community growth, engagement, diversity, and retention over time. The system ingests member registration data and event attendance data, normalizes and reconciles them across multiple sources, and exposes chart-ready analytics APIs used by a dashboard interface. The goal is to provide clear, decision-ready insights for RCC leadership without manual spreadsheet analysis.

Key Questions Answered 🤔

• How many members does RCC have? How many members are active?
• Who makes up the RCC community (academics, year levels)?
• Are events attracting a diverse audience?
• How well does RCC retain attendees across multiple events?

Analytics 📊

• KPI cards (total members, active %, growth rate)
• Line charts (cumulative membership growth over time)
• Pie charts (major category distribution, class year distribution)
• Event-level stacked charts (attendance diversity by major category)
• Bar charts ( explaining how many people attended 0, 1, 2, 3, or 4+ events)
• Grouped / stacked views by major category

User Flow (High-Level) 🔀

1. Members register through a standardized form → stored in members
2. Admin uploads event attendance CSVs → parsed and normalized
3. Attendance is linked to members when possible
4. Active membership is recalculated automatically
5. Dashboard fetches analytics via read-only endpoints
6. Charts render using pre-aggregated backend data

Tech Stack 🛠️

• Next.js 16 (App Router) and React 19
• Tailwind CSS v4, shadcn/ui, Base UI
• next-themes, sonner, motion
• FastAPI + Uvicorn backend (Python 3.10+)
• TypeScript and Zod-based env validation
• Bun and Turborepo
• Supabase auth & Supabase (PostgreSQL)

Repository Layout 📁

RCC Dashboard and Engagement Tool is a Turborepo monorepo for a dashboard web app and shared packages.

Turborepo/
├── apps/
│   ├── web/           # Next.js web app
│   └── server/        # FastAPI backend
└── packages/
    ├── config/        # Shared TS config
    └── env/           # Shared env schemas (server/web/native)

Architecture (High-Level) 🏛️

• Frontend fetches analytics via GET endpoints
• Admin actions (event imports) use POST multipart/form-data
• Backend returns fully computed JSON

1️⃣ Member Registration

Google Form Submission
        ↓
Google Apps Script (onFormSubmit / onEdit)
        ↓
Supabase (members table)

2️⃣ Event Attendance

CSV & Event Form Upload
          ↓
FastAPI Import Layer
          ↓
Supabase (PostgreSQL)
          ↓
Analytics SQL / RPC Functions
          ↓
REST Analytics Endpoints
          ↓
Next.js Dashboard UI

API Endpoints 🔗

Overview Analytics

GET /analytics/overview Returns:

• Total members
• Active members + percentage
• 30-day growth rate
• Monthly cumulative growth time series

GET /analytics/mission Returns:

• Member distribution by major category
• Member distribution by class year
• Top events with attendance diversity breakdown

GET /analytics/retention Returns:

• Attendance frequency buckets (0, 1, 2, 3, 4+)
• Overall distribution
• Distribution by major category

POST /api/import/event-attendance

• Content-Type: multipart/form-data
• Required fields:
  • import_type = "event_attendance"
  • title
  • starts_at
  • event_kind
  • file (CSV)
• Optional:
  • event_type
  • location
  • committee

Returns:

• Event ID
• Validation summary
• Success summary
• Skipped rows with reasons
• Warnings

Data Assumptions & Logic 📝

Members vs Non-Members

• Members are identified by email match with members.email
• Non-members are still included in:
  • Event diversity analytics
  • Retention analytics

Active Member Logic

• Only members can become active
• A member becomes active if they attend qualifying events
• is_active_member and active_member_start_date are updated automatically

Missing Data Handling

Field	Behavior
Major	Falls back to member record or Other/Unknown
Program	Inferred from class year, major tokens, or member record
Class Year	Normalized or set to Unknown
Check-in time	Stored as NULL if malformed
Duplicate emails	Skipped during import

Normalization Rules

Email

Lowercased, whitespace removed, used as global unique identifier

Major

Free-text majors are normalized into major_normalized & major_category (Technical, Business, Humanities & Arts, Health Sciences, Other/Unknown)

Degree Program

Derived using:

• Explicit class year (Freshman → Undergraduate)
• Major tokens (e.g., “M.S.” → Graduate)
• Member record fallback

Data Schema 📑

• members: Stores registered RCC members and normalized demographic data.
• events: Stores event metadata (no attendance).
• event_attendance: Stores attendee engagement, supports members + non-members.
• profiles: Controls platform access (admin vs member).

Getting Started (monorepo) ⚙️

Prerequisites: Bun 1.2+, Python 3.10+, Node.js (for Turbo)

Install dependencies:

bun install
bun run server:install               # creates .venv in apps/server (no pip yet)
cd apps/server
source .venv/bin/activate            # Windows: .venv\Scripts\Activate
pip install -r requirements.txt

Run all apps (web + server):

bun run dev

Run only one app:

bun run dev:web
bun run dev:server

Open:

• Web: http://localhost:3000
• Server: http://localhost:8000 (docs at /docs)

Backend (apps/server)

Requirements

• Python 3.10+
• bun (or npm/pnpm) for repo scripts
• Node.js (for Turbo)

Quick start

From repo root:

bun install                  # JS deps
bun run server:install       # makes .venv in apps/server

Install Python deps (once the venv exists):

cd apps/server
source .venv/bin/activate         # Windows: .venv\Scripts\Activate
pip install -r requirements.txt

Run the API

# via Turbo (recommended, repo root)
bun run dev

# or directly (backend only)
cd apps/server
source .venv/bin/activate
python -m uvicorn app.main:app --reload --host 0.0.0.0 --port 8000

Endpoints (default)

• API base: http://localhost:8000
• Swagger: http://localhost:8000/docs
• OpenAPI JSON: http://localhost:8000/openapi.json

Notes

• server:install only creates the venv; it does not pip install.
• Turbo assumes python resolves inside the venv; activate it before bun run dev.
• For fresh shells, re-run the venv activation step.

Troubleshooting

• python: command not found: use python3 -m venv .venv then activate.
• No module named uvicorn: pip install -r requirements.txt inside the venv.
• Turbo exits 127: ensure the venv is active so python is on PATH.

Fallback

OS	Command
macOS / Linux	source .venv/bin/activate
Windows	.venv\\Scripts\\activate

Environment Variables

Edit apps/web/.env as needed:

SERVER_URL=http://localhost:8000

Shared env schemas live in packages/env/src/*.ts.

Scripts

• bun run dev: Start all apps in dev mode
• bun run dev:web: Start the web app only
• bun run build: Build all apps
• bun run check-types: Typecheck across the repo
• bun run server:install: Create the server venv in apps/server (no pip)

Future Improvements 🚀

• Improved major data collection and analytics
• Event cohort analysis (first-time vs. returning)
• Attendance trend forecasting

Contributors & Roles 🤝

Name	Role / Title	Key Contributions
Julia Husainzada
Asmita Dulla
Izabella	Backend / Data Developer	Did a little bit of everything related to analytics and data
Samriddhi Matharu
Emily Thach	Frontend developer	Dashboard design & ui/ux, data schema, customized ui charts/components
Pouya Anvari	Full-Stack developer	Figma design, UI wiring, API & Endpoint Aggregation
Om Shah	Builder.	Can write Code and owns a computer