Retail Sales Analysis: An Interactive Power BI Dashboard
Welcome to my Power BI portfolio! This project showcases my end-to-end skills in data analytics, from cleaning and transforming data to building a feature-rich, interactive dashboard that delivers actionable business insights.
Author: Ernando Taufiq Nur Hidayat
๐ Interactive Power BI Report
You can interact with the dashboard preview below. For a full-screen experience, open the PDF.
Download dashboard preview (PDF)
๐ Project Executive Summary
This project analyzes sales data from a fictional retail company to identify key trends, profit drivers, and areas for growth. The dashboard is designed to provide stakeholders with quick, data-driven answers to critical business questions.
- Data Source: Maven Analytics Guided Project Dataset.
- Tools Used: Microsoft Power BI (Power Query, DAX), Excel.
๐ฏ Project Objectives:
- Identify the best and worst-performing product categories and individual products.
- Analyze sales performance across different geographical locations (cities and store types).
- Uncover sales patterns and trends over time.
๐ Key Findings & Insights:
- Healthy Profitability Driven by High Volume: The business operates with a healthy 28.6% profit margin, but relies heavily on high-volume sales of low-cost items, resulting in a low Average Order Value (AOV) of ~$16.89.
- “Downtown” Store Dominance: The “Downtown” store location is a clear outlier, contributing 56% of the total profit, indicating successful strategies that could be replicated elsewhere.
- Strong Seasonal Sales Peak: A significant sales spike occurs around the year-end holiday season (Q4) and the beginning of the year (Q1), which is critical for inventory and marketing planning.
๐ ๏ธ Technical Process & Methodology
1. Data Extraction, Transformation, and Loading (ETL)
- Used Power Query to connect to and import data from multiple CSV files.
- Performed extensive data cleaning and transformation, including:
- Correcting data types (e.g., text to date).
- Handling null or missing values.
- Creating new custom columns to support deeper analysis (e.g., extracting Year and Month from a date column).
2. Data Modeling
- Designed an efficient Star Schema to optimize report performance and ensure accurate calculations.
- Established correct relationships between fact and dimension tables.
- Created a dedicated Date Table using DAX to enable complex time intelligence analysis.
3. DAX Calculations
I wrote several dynamic Measures using DAX to create key business metrics. Some notable examples include:
- Total Revenue & Profit:
Total Revenue = SUM(Sales[Revenue]) - Profit Margin:
Profit Margin = DIVIDE([Total Profit], [Total Revenue]) - Time Intelligence: Used functions like
SAMEPERIODLASTYEARandDATESYTDto compare performance over time.
4. Visualization & Dashboard Design
- Focused on data storytelling, where each visual was chosen to answer a specific business question.
- Maintained a consistent color scheme and layout to ensure a clean and intuitive User Experience (UX).
- Added interactive features like slicers, informative tooltips, and drill-through capabilities to allow for self-service analysis by the end-user.
๐ก Data-Driven Business Recommendations
Based on the insights uncovered, here are several actionable recommendations:
- Implement an Average Order Value (AOV) Growth Strategy: Given the low AOV, the company should introduce product bundling or frequently-bought-together recommendations at checkout.
- Replicate the “Downtown” Success Model: Conduct a deep-dive case study on the Downtown store’s operations, staffing, and local marketing strategies to create a blueprint for underperforming locations like “Airport”.
- Diversify the “Electronics” Category: Re-evaluate the product portfolio and create targeted promotions to boost sales volume.
๐ฌ Hubungi Saya
Jika Anda memiliki pertanyaan atau ingin berdiskusi mengenai proyek ini, jangan ragu untuk menghubungi saya.
- LinkedIn: Ernando Taufiq Nur Hidayat
- Email: ernando.taufiq29@gmail.com