This project focuses on building a data-driven sales forecasting and inventory intelligence solution for a jewellery brand using Power BI.
This project presents an end-to-end Business Intelligence solution built using Power BI for a jewellery brand, focused on sales forecasting, product demand analysis, and reorder intelligence.
The dashboard converts raw transactional data into actionable insights to support:
- Short-term sales forecasting
- SKU-level demand tracking
- Inventory planning and reorder decisions
- City and marketplace performance analysis
Jewellery retail faces unique challenges:
- High-value inventory
- Fluctuating product demand
- Regional and marketplace-level variability
The business needed clear answers to:
- What will sales look like in the next 30 days?
- Which SKUs are driving demand?
- Which products show declining or negative trends?
- Where should inventory be replenished or optimized?
- Time-series forecasting of net sales
- Confidence intervals to represent forecast uncertainty
- Helps in revenue planning and inventory budgeting
- SKU-wise demand trend analysis
- Identification of:
- Fast-moving products
- Declining or low-demand SKUs
- Enables data-driven reorder planning
Key KPIs
- Total Net Sales
- Total Quantity Sold
- Total Orders
- Total Active Products
Analytical Views
- City-wise sales and quantity contribution
- Top 5 products by revenue
- Marketplace comparison (Amazon vs Myntra)
- Interactive slicers for SKU-level deep dives
- A small percentage of SKUs contribute disproportionately to total revenue
- Certain cities consistently outperform others in sales and quantity
- Some products show declining demand and require cautious reordering
- Forecast confidence bands help mitigate inventory risk
- Power BI
- DAX (Measures, KPIs, trend calculations)
- Time-Series Forecasting
- Interactive Dashboards & Slicers
- Retail & E-commerce Analytics
The dataset contains anonymized transactional sales data, including:
- Order date
- SKU code
- Quantity sold
- Net sales value
- City
- Marketplace (Amazon / Myntra)
⚠️ Note: Data has been anonymized to maintain confidentiality.
This dashboard enables:
- Improved demand forecasting accuracy
- Reduced stockouts and excess inventory
- Smarter SKU-level inventory decisions
- Better regional and marketplace performance tracking
- Automated reorder quantity recommendations
- Seasonality-aware forecasting models
- Python-based forecasting (ARIMA / Prophet)
- Low-stock alerts and exception reporting
- Download the
.pbixfile - Open it in Power BI Desktop
- Refresh the data
- Use slicers to explore SKUs, cities, and marketplaces
Prachi Bhuwad
Data Analyst | Business Intelligence | Forecasting & Retail Analytics
📫 Open to Data Analyst / BI / Forecasting roles
Recommended repo name:
jewellery-sales-forecasting-powerbi
Include:
.pbixfile/imagesfolder with dashboard screenshotsREADME.md
GitHub Topics:
power-bi, data-analytics, forecasting, retail-analytics, dax, inventory-management