E-commerce-Performance-Dashboard

Power BI dashboard that tracks daily revenue, purchase behavior and user efficiency for an online store.

---

1. Business Question

“How is revenue trending, what is driving spikes, and are we getting more value from each user?”

2. Data

File	Rows	Refresh	Notes
ecommerce_sales.csv	90	Manual (demo)	Date, users, purchases, revenue

Data transformation

• Removed $ and , from total_revenue, cast to decimal.
• Added Conversion Rate, AOV, ARPU, rolling 30-day average.
• Created Date dimension for proper YTD / MTD logic.

3. Key Metrics

Metric	Definition
Revenue	Sum of total_revenue
Purchases	Sum of purchase_events
Conversion Rate	Purchases ÷ Unique Users
Average Order Value	Revenue ÷ Purchases
Revenue Δ vs Prior 7 d	Week-over-week change

4. Report Pages

1. Overview – KPI strip, period buttons, revenue trend (raw & 30-day MA), purchases bars.
2. Details – Ranked revenue days, weekday/month heat-map, scatter (CR × AOV, bubble = users).

5. Insights

• Revenue down $812 vs prior 7 days, driven by lower traffic (-33 users) more than basket size.
• Highest single-day revenue: 24 Nov 2020 – Conversion Rate spike to 2.1 % (+0.9 pp).
• Fridays drive 35 % of monthly revenue; Sunday is consistently lowest.

6. How to Use

• Click Last 7 D / MTD / YTD buttons to shift time horizon.
• Hover the revenue line to read raw vs 30-day-avg values.
• Drill-through any bar in “Total Revenue by Date” to view raw order table (hidden page).

7. Future Work

• Merge marketing-spend to calculate ROAS.
• Prophet forecast for next 30 days revenue (Python notebook included).
• Anomaly alert in Power BI Service using data-driven alert + Teams webhook.

8. Setup

1. Clone repo, open powerbi/ecommerce_dashboard.pbix.
2. Change data source to your local path or set parameter DATA_DIR.
3. Refresh → Publish to your workspace.

---

Author — Trevor Sancho(https://www.linkedin.com/in/) • MIT-License