Sales Data Analysis Using Power Bi

sales dashboard

Introduction

During my virtual Data Analyst internship at MeriSKILL.I was tasked with analyzing a large sales dataset to identify trends, create visualizations, and share actionable recommendations. With over 185,000 records, this dataset provided a wealth of information that helped me uncover insights into customer behaviors, product performance, and sales trends. Here’s a look at how I approached the analysis, what I discovered, and my final recommendations.

Raw Sales Data

Data Collection and Structure

The dataset consisted of 11 columns capturing various attributes, such as:

• City: Location of the sale
• Product: Category of the product sold
• Order Date: Date of the transaction
• Revenue: Total revenue from each sale
• Quantity Ordered: Number of units sold
• Profit Margin: Percentage profit on each sale

Each row represented an individual transaction, allowing for granular analysis across metrics like revenue, sales quantity, and profit margins.

Data Cleaning

Before digging into the data, I needed to clean it up a bit. This is a crucial step to eliminate factors that could cause errors and create bias in my analysis. I checked for duplicate entries in the data, and there were none.

I transformed the first row to headers

Transformed the data types of each column to the appropriate type

I split the Date and Time into two columns by choosing the ‘Split Column’ option and selecting space as the delimiter.

I further extracted the Days of the week as I will need them during my analysis. I did that by using the formula:

Day Of Week = FORMAT('Sales data'[Order Date].[Day],"dddd")

Split Days, Time, and Date of Sales Data

Exploratory Data Analysis (EDA) and Visualizations

To highlight various aspects of the business, I created visualizations using Power BI, allowing for dynamic filtering and interactive exploration.

1. Sales Overview

A high-level dashboard provided a summary of key metrics:

• Revenue: $34.49M
• Sales Quantity: 209K units
• Profit Margin: 58.82%

Analysis

Now comes the fun part! I made different visualizations to highlight various aspects of the business growth.

I used a line graph to show the sales trend in each month.

Sales trend over time using the line chart

I used a Clustered Bar Chart to highlight the five most bought products.

Top best-selling products

I used a map to show the Top 5 cities by sales

Top 5 cities by sales

I also created slicers by Month, Days of the week, Product, and City.

Finally, I organized everything in a Dashboard for easier visualization.

After critically examining all the charts, I noticed some interesting things. For example, Macbook Pro Laptops were selling well, with a revenue sales of 8.0M, but some other products didn’t sell as much despite higher turn-off rates, such as AAA Batteries, which accounted for the lowest sales of 0.1M.

macbook pro sales analysis

iphone sales analysis

Highest sales by product

Weekly Sales Analysis

A bar chart was used to explore sales by day of the week. The data shows that Sundays, Mondays, and Tuesdays have the highest sales volume, with a noticeable drop-off from Wednesday onward.

Additionally, it highlights the importance of understanding customer behavior patterns and aligning sales strategies accordingly to maximize revenue opportunities.

The sum of Sales by Days of The Week

Key Insights and Recommendations

This analysis provided a comprehensive view of customer behavior, product performance, and geographic demand. Based on these insights, here are my top recommendations:

1. Focus on High-Value Products
   Prioritize high-revenue products like the MacBook Pro and ThinkPad Laptop in stock and marketing efforts. Consider bundling accessories with these high-demand products to increase average order value.
2. Leverage Weekly Shopping Trends
   Launch targeted “Weekend Kickoff Deals” on Sundays and Mondays to capitalize on high sales volume. Use these promotions to cross-sell complementary items, such as headphones with laptops.
3. Increase Market Presence in Top Cities
   Invest in digital ads and localized promotions in high-demand regions. For instance, targeted offers could be tailored to preferences in major urban areas to boost engagement and sales.
4. Optimize Inventory and Marketing for Low-Performing Products
   Consider reducing stock of low-margin items, like USB-C cables and batteries, and refocus budgets on high-demand products. Alternatively, bundle these items with popular products to increase visibility without heavy discounts.
5. Prepare for Seasonal Sales Peaks
   Ensure high-margin products are well-stocked before the holiday season to meet demand. Starting promotions in October could capture early shoppers and sustain momentum throughout November and December.
6. Data-Driven Marketing
   Implement a routine for tracking weekly and monthly trends to enable agile marketing strategies. Customer segmentation based on purchase frequency and preferences can help design personalized campaigns, enhancing engagement and boosting lifetime customer value.

Conclusion

Analyzing sales data can provide valuable insights into a business’s performance and highlight opportunities for growth. This project taught me the importance of data cleaning, visualization, and data-driven decision-making. Through this internship, I was able to refine my analytical skills and deliver actionable recommendations that align with business goals.

Thanks for reading!