Consumer Trends and Online Shopping: Data Analysis

People now frequently make online purchases because it is much more convenient and faster than going to a physical store. Placing an order with a single click—that’s all it takes. On the Kaggle website, I found a dataset regarding consumer trends and online shopping. For me, it was important that the dataset included both numerical and categorical data so that it would be more interesting to compare and find correlations, and also to create a «portrait» of the buyer—their preferences and behavior.

The dataset contains the following information: Age — customer age Gender — customer gender Item Purchased — item name Category — item category Purchase Amount (USD) — total expenditure Location — purchase location Color — item color Season — season Review Rating — customer review rating Subscription Status — membership plan Shipping Type — delivery method Discount Applied — whether a discount was applied Promo Code Used — whether a promo code was used Previous Purchases — number of past purchases Payment Method — payment method used Frequency of Purchases — purchase frequency

The project will utilize a bar chart (to compare customer ages), a pie chart (to show proportional shares), a line graph (to display purchasing categories and their costs), a heatmap, and a Word cloud.

To create the color palette, I decided to generate a portrait of that very online shopper using Recraft. These colors served as the core palette for the project and its visualizations.

Prompt:

Shopping Trends And Customer Behaviour, customer portrait, close up.

Who is the online shopper? Before diving into consumer trends in online purchasing, I decided to first create a profile of the clients of online services. To do this, I made a pie chart reflecting the proportion of women and men who make purchases.

To my surprise, the majority of our customers are men (68%). Women account for only 32%, which is roughly half as many. This indicates that most online orders are placed by male customers.

Next, I decided to refine the «portrait» of our online buyers even further. To do this, I created a bar chart to determine the age of our clients. For clarity, I also highlighted several purchasing age categories. The largest age group is people aged 45 to 59; they make up the majority of buyers. Interestingly, people over 60 also make purchases, and their numbers are close to those in the 18–29 age bracket.

I also created a tag cloud indicating where online purchases were made. Thus, the majority of orders originated from California, Montana, Idaho, and Illinois.

I then became interested in the methods people use most frequently to pay for their online orders. To investigate this, I created a pie chart illustrating the different payment methods used in stores. The resulting image shows that the distribution of payment methods is quite uniform—the difference between the largest and smallest indicator is only 1,1%. This suggests that there are almost no particular preferences for paying for online orders; people use all the available methods equally.

And which categories do online store customers spend the most money on? To find this out, I decided to create a line chart to see which category is the most expensive. The image shows that online shoppers spend the most money on the clothing category, totaling over $100,000.

Next, I decided to find out which products are rated the best by online buyers. To do this, I created a heat map that groups products by name and shows the average customer review score for each item. The highest ratings were for gloves, while shirts received more negative feedback. Furthermore, the average rating wasn’t as high as expected, which might indicate buyer dissatisfaction.