Alt-Data Pulse Dashboard

Alt-Data Pulse is a portfolio-ready analytics dashboard that blends alternative data (spend, traffic, hiring, and app engagement) with consensus context to surface revenue momentum ahead of earnings. The interface is designed to feel like a modern, Bloomberg-style terminal: dense, fast to scan, and tuned for decision-making.

Screenshots

Why It Matters

Traditional earnings models lag reality. Alt-Data Pulse focuses on real-world behavior (transactions, visits, hiring demand, and engagement) so trends show up weeks earlier, with clear deltas versus consensus.

Key Capabilities

• Revenue Proxy vs Wall Street Consensus with beat/miss deltas.
• Brand selector and time-range filters to compare momentum quickly.
• Foot traffic and ticket-size trends with rolling averages.
• Hiring insights dashboard for labor demand and staffing velocity.
• App engagement signals to triangulate brand momentum.
• News panel for qualitative context alongside quantitative signals.
• Bloomberg-style dark theme optimized for long sessions.

Signal Coverage

• Spend and transactions (daily granularity).
• Foot traffic index with rolling averages.
• Hiring signals (job postings, growth, and velocity).
• App engagement metrics (DAU, installs, usage intensity).
• Optional consensus data (LSEG-compatible CSV).

Tech Stack

• Python, Pandas, NumPy, SciPy, Statsmodels
• Plotly Dash with TradingView-style interactivity
• Streamlit (legacy UI with Altair charts)
• Google Cloud Storage + BigQuery loaders

Architecture

alt-data-pulse/
├── dash_app/             # Plotly Dash UI (primary)
├── dashboard/            # Streamlit UI (legacy)
├── ingest/               # Data pipelines + predictor
├── data/                 # Cleaned CSVs + cached outputs
├── notebooks/            # Analysis + experimentation
└── docs/screenshots/     # Portfolio screenshots

Run Locally

python -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt

# Plotly Dash dashboard (recommended)
python dash_app/app.py    # runs on http://127.0.0.1:8050

# Or Streamlit dashboard (legacy)
streamlit run dashboard/app.py    # runs on http://localhost:8501

Data Pipeline

python ingest/pipeline.py --full --start-date 2023-01-01

Outputs land in data/ and are loaded by the dashboard.

Roadmap

• Incremental refresh pipeline for near-real-time updates.
• Expanded coverage beyond QSR to other consumer categories.
• Workspace tabs and multi-layout views.
• Deeper news and sentiment integrations.

---