Shopping Data Analysis And Visualization

The project is based on the Customer Shopping Dataset retail Sales Data provided by Kaggle. We will do data visualization studies with PowerBI using this data set.

The dataset includes shopping information from 10 different shopping centers between the years 2021–2023. Data is available from various age groups and genders to provide a comprehensive view of shopping habits in Istanbul. The dataset includes basic information such as invoice numbers, customer ID, age, gender, payment methods, product categories, quantity, price, order dates, and mall locations.

Data Overview with Pandas

We examined the first five and last five rows of our data.

It consists of 3 numeric and 7 categorical variables.

There are no missing values in our dataset.

Column headings are ok, no spaces. Our dataset consists of 99457 observations and 10 attributes.

We have seen the unique elements of the columns.

Most shopping is done in Mall of Istanbul, least shopping is done in Emaar Square Mall. The average age of customers shopping in shopping malls is generally the same (43).

Based on the gender of the shoppers, the age statistics of male and female shoppers are almost the same, women have done more shopping, and if we look at the average of expenditures, men have spent more.

44447 customers completed their purchases with cash, 34931 customers with customer credit cards and 20079 customers with debit cards, if we take the average expenditures as a basis, cash expenditures were relatively higher.

Most of the shopping expenses were made for the clothes and the least for the books.

Data Visualization with PowerBI

First, let’s decide which colors to use. For this, I got help from the image of Kanyon AVM in Istanbul. I used this tool https://coolors.co/

1- Total Customers, Total Sales Quantity, Total Price, Average Age

Total customers, total sales quantity, total price and avegrage age are key measures that are important in understanding the overall customer performance. Therefore, in order to clearly visualize these numbers, a chart called a Card that displays these figures is often used. The Card chart is a useful tool quickly assess the performance of a company or department.

2- Total Spend Based Shopping Centers

The best way to view and compare spending in shopping malls is to use clustered bar charts. Here, we can clearly see that the most spending is made at the Mall of Istanbul, and the least is spent at Forum Istanbul.

3- Average Quantity Over Time and Total Price Based on Age

Line charts allow users to easily identify trends and patterns in data changes over time. Here, we can clearly see the average of sales quantities and the change over time. Moreover, we can adjust the time bar at any scale we want.
In the second chart, we see the total expenditures based on age. These values are very helpful in determining the age ranges we will address.

4- Categories

Bar charts are a common and effective way to present sales data as they allow for easy comparison of sales figures between different categories. By drilling down into the data, analysts can gain a more detailed understanding of the sales performance of each category and make informed decisions about product strategy and inventory management.

5- Customers Paymend Method

Showing the payment methods used in the donut chart is the easiest way to tell the difference. By using a donut chart to display this information, businesses can quickly and easily identify which payment method customers use the most and adjust their sales strategy accordingly.

6- Slicer

Sales Performance Dashboard has a feature called “Slicer” which filters specific data to be displayed on the dashboard, making data analysis easier and more convenient. This dashboard has three slicers in total:

Invoice Date: Filters data to display a specific time period.
Age range: Filters data to display only the selected age range. You can swipe the bar or type the age range.
Gender: Presents data in two ways, male or female.

Finally, the Customer Shopping Dashboard can be created by arranging and combining charts and slicers as shown below.

The project link: my GitHub.

I hope i was helpful. Thank you!