https://github.com/dcostachar/pet-store-transactions-analysis
An analysis of a pet store's transactions dataset using Python to clean, explore, and visualize key trends in product sales.
https://github.com/dcostachar/pet-store-transactions-analysis
data-cleaning data-visualization exploratory-data-analysis matplotlib pandas python seaborn
Last synced: 2 months ago
JSON representation
An analysis of a pet store's transactions dataset using Python to clean, explore, and visualize key trends in product sales.
- Host: GitHub
- URL: https://github.com/dcostachar/pet-store-transactions-analysis
- Owner: dcostachar
- Created: 2025-02-12T18:35:56.000Z (over 1 year ago)
- Default Branch: main
- Last Pushed: 2025-02-12T18:46:56.000Z (over 1 year ago)
- Last Synced: 2025-06-29T07:41:52.461Z (12 months ago)
- Topics: data-cleaning, data-visualization, exploratory-data-analysis, matplotlib, pandas, python, seaborn
- Language: Jupyter Notebook
- Homepage:
- Size: 105 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Pet Store Transactions Analysis: Uncovering Sales Insights with Python
Author: Charlene D'Costa
Date: February 10, 2025
Coursework for the Meta Marketing Analytics Professional Certificate.
[Pet Store Transactions Dataset](https://github.com/dcostachar/pet-store-transactions-analysis/blob/main/data/transactions.csv)
# Project Overview
Defining the business problem.
For this project, I used Python to perform data cleaning and exploratory data analysis (EDA) on a fictitious pet store transactions dataset. I began by cleaning the data—handling missing values, removing data anomalies, and dropping columns with excessive nulls. Next, I conducted EDA to uncover product sales trends, analyze the quantity sold by category, and identify both the most popular and the highest-priced categories within different product lines. Finally, I created visualizations, including bar and box plots, to provide stakeholders with actionable insights on top-selling categories and price distributions, informing business decisions about which product lines drive the most revenue and where pricing adjustments might be considered.
# Data Cleaning
Cleaning the data for analysis.
In this section, I will import and clean the dataset to prepare it for analysis. I will use PyCharm as my integrated development environment (IDE) along with its Jupyter Notebook integration to perform the analysis. Additionally, I import the following libraries, which will be used throughout the project: pandas, matplotlib, and seaborn.
```python
# Importing libraries.
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
# Importing dataset.
df = pd.read_csv("./data/transactions.csv")
```
.dataframe tbody tr th:only-of-type {
vertical-align: middle;
}
.dataframe tbody tr th {
vertical-align: top;
}
.dataframe thead th {
text-align: right;
}
Date
Order_Number
Customer_ID
Product_Name
SKU
Price
Size
Quantity
Product_Category
Product_Line
0
5/22/2021
SXF-7309-1727-1334
476582ea-1bba-4289-8775-3fcd8074821c
Feline Fix Mix
RKAPY3I1TP
39.55
NaN
1
treat
cat
1
5/22/2021
SXF-7309-1727-1334
476582ea-1bba-4289-8775-3fcd8074821c
Scratchy Post
MPH6SCD7UT
26.95
NaN
3
toy
cat
2
5/22/2021
SXF-7309-1727-1334
476582ea-1bba-4289-8775-3fcd8074821c
Reddy Beddy
DJWE1V9LZK
23.07
large
3
bedding
dog
3
3/23/2020
DG7-5410-5845-1340
5929a0e9-95a7-4dbf-896e-c11d1988615e
Snoozer Essentails
GABWVMEL2R
28.04
NaN
3
bedding
dog
4
3/23/2020
DG7-5410-5845-1340
5929a0e9-95a7-4dbf-896e-c11d1988615e
Reddy Beddy
KDTMPSBZKZ
13.84
small
1
bedding
dog
...
...
...
...
...
...
...
...
...
...
...
2898
10/16/2020
P8K-8079-0264-6569
468f40b3-59ca-47fd-9739-c7f19cf48f32
Cat Cave
058G0P7V60
29.66
NaN
1
bedding
cat
2899
10/16/2020
P8K-8079-0264-6569
468f40b3-59ca-47fd-9739-c7f19cf48f32
Kitty Climber
W86BRJ9SSG
39.32
NaN
1
toy
cat
2900
10/16/2020
P8K-8079-0264-6569
468f40b3-59ca-47fd-9739-c7f19cf48f32
Fetch Blaster
M291KHJ4LW
29.47
NaN
1
toy
dog
2901
10/16/2020
P8K-8079-0264-6569
468f40b3-59ca-47fd-9739-c7f19cf48f32
Snoozer Essentails
GABWVMEL2R
28.04
NaN
1
bedding
dog
2902
12/10/2019
6ZD-7972-0320-6653
f2a090b3-ec77-4018-939e-1a18d2b4f4ef
Snoozer Essentails
GABWVMEL2R
28.04
NaN
1
bedding
dog
2903 rows × 10 columns
```python
# Using df.info() to display a concise summary of the DataFrame, including the index, column names, data types, non-null counts, and memory usage.
df.info()
```
Index: 2758 entries, 0 to 2902
Data columns (total 10 columns):
# Column Non-Null Count Dtype
--- ------ -------------- -----
0 Date 2758 non-null object
1 Order_Number 2758 non-null object
2 Customer_ID 2716 non-null object
3 Product_Name 2758 non-null object
4 SKU 2758 non-null object
5 Price 2758 non-null float64
6 Size 626 non-null object
7 Quantity 2758 non-null int64
8 Product_Category 2758 non-null object
9 Product_Line 2758 non-null object
dtypes: float64(1), int64(1), object(8)
memory usage: 237.0+ KB
### Question 1: Remove all rows that are missing either the `Product_Name` or the `Product_Category`.
```python
# Dropping rows where either the 'Product_Name' or the 'Product_Category' column has a missing value.
df = df.dropna(subset = ['Product_Name', 'Product_Category'])
```
.dataframe tbody tr th:only-of-type {
vertical-align: middle;
}
.dataframe tbody tr th {
vertical-align: top;
}
.dataframe thead th {
text-align: right;
}
Date
Order_Number
Customer_ID
Product_Name
SKU
Price
Size
Quantity
Product_Category
Product_Line
0
5/22/2021
SXF-7309-1727-1334
476582ea-1bba-4289-8775-3fcd8074821c
Feline Fix Mix
RKAPY3I1TP
39.55
NaN
1
treat
cat
1
5/22/2021
SXF-7309-1727-1334
476582ea-1bba-4289-8775-3fcd8074821c
Scratchy Post
MPH6SCD7UT
26.95
NaN
3
toy
cat
2
5/22/2021
SXF-7309-1727-1334
476582ea-1bba-4289-8775-3fcd8074821c
Reddy Beddy
DJWE1V9LZK
23.07
large
3
bedding
dog
3
3/23/2020
DG7-5410-5845-1340
5929a0e9-95a7-4dbf-896e-c11d1988615e
Snoozer Essentails
GABWVMEL2R
28.04
NaN
3
bedding
dog
4
3/23/2020
DG7-5410-5845-1340
5929a0e9-95a7-4dbf-896e-c11d1988615e
Reddy Beddy
KDTMPSBZKZ
13.84
small
1
bedding
dog
...
...
...
...
...
...
...
...
...
...
...
2898
10/16/2020
P8K-8079-0264-6569
468f40b3-59ca-47fd-9739-c7f19cf48f32
Cat Cave
058G0P7V60
29.66
NaN
1
bedding
cat
2899
10/16/2020
P8K-8079-0264-6569
468f40b3-59ca-47fd-9739-c7f19cf48f32
Kitty Climber
W86BRJ9SSG
39.32
NaN
1
toy
cat
2900
10/16/2020
P8K-8079-0264-6569
468f40b3-59ca-47fd-9739-c7f19cf48f32
Fetch Blaster
M291KHJ4LW
29.47
NaN
1
toy
dog
2901
10/16/2020
P8K-8079-0264-6569
468f40b3-59ca-47fd-9739-c7f19cf48f32
Snoozer Essentails
GABWVMEL2R
28.04
NaN
1
bedding
dog
2902
12/10/2019
6ZD-7972-0320-6653
f2a090b3-ec77-4018-939e-1a18d2b4f4ef
Snoozer Essentails
GABWVMEL2R
28.04
NaN
1
bedding
dog
2758 rows × 10 columns
### Question 2: Find any clearly "incorrect" values in the `Price` column and clean the DataFrame to address those values.
```python
# Based on the earlier output from df.info(), we know that the 'Price' column is numeric (float64).
# Therefore, instead of checking for non-numeric data, we need to verify that the price values are valid.
# For example, negative prices are not acceptable.
# Check the minimum and maximum values of the 'Price' column.
# (A negative minimum indicates an invalid value.)
df.Price.min(), df.Price.max()
# Create a mask to retain only rows where 'Price' is greater than 0 and less than 15,000.
# This filters out any negative prices and any values that are unreasonably high.
valid_mask = (df.Price > 0) & (df.Price < 15000)
# Apply the mask to filter the DataFrame.
df = df[valid_mask]
```
(np.float64(10.8), np.float64(39.55))
### Question 3: After you've done the cleaning above, remove any column that has more than 500 missing values.
```python
# From counting the number of missing values in each column, we see that the 'Size' column has over 500 missing values (2222 missing values).
df.isna().sum()
# Dropping the 'Size' column from the DataFrame.
df = df.drop(columns='Size')
```
### Question 4: Address the other missing values. You can replace the values or remove them, but whatever method you decide to clean the DataFrame, you should no longer have any missing values.
```python
# Using df.dropna() to remove any remaining rows that contain missing values.
df = df.dropna()
# Verifying that all missing values have been removed by counting the missing values in each column. Since there are none, we can move on.
df.isna().sum()
```
Date 0
Order_Number 0
Customer_ID 0
Product_Name 0
SKU 0
Price 0
Quantity 0
Product_Category 0
Product_Line 0
dtype: int64
# Exploratory Data Analysis
Uncovering patterns and analyzing relationships between variables.
### Question 5: Create a `Subtotal` column by multiplying the `Price` and `Quantity` values.
```python
# Creating 'Subtotal' column
df['Subtotal'] = df.Price * df.Quantity
```
.dataframe tbody tr th:only-of-type {
vertical-align: middle;
}
.dataframe tbody tr th {
vertical-align: top;
}
.dataframe thead th {
text-align: right;
}
Date
Order_Number
Customer_ID
Product_Name
SKU
Price
Quantity
Product_Category
Product_Line
Subtotal
0
5/22/2021
SXF-7309-1727-1334
476582ea-1bba-4289-8775-3fcd8074821c
Feline Fix Mix
RKAPY3I1TP
39.55
1
treat
cat
39.55
1
5/22/2021
SXF-7309-1727-1334
476582ea-1bba-4289-8775-3fcd8074821c
Scratchy Post
MPH6SCD7UT
26.95
3
toy
cat
80.85
2
5/22/2021
SXF-7309-1727-1334
476582ea-1bba-4289-8775-3fcd8074821c
Reddy Beddy
DJWE1V9LZK
23.07
3
bedding
dog
69.21
3
3/23/2020
DG7-5410-5845-1340
5929a0e9-95a7-4dbf-896e-c11d1988615e
Snoozer Essentails
GABWVMEL2R
28.04
3
bedding
dog
84.12
4
3/23/2020
DG7-5410-5845-1340
5929a0e9-95a7-4dbf-896e-c11d1988615e
Reddy Beddy
KDTMPSBZKZ
13.84
1
bedding
dog
13.84
...
...
...
...
...
...
...
...
...
...
...
2898
10/16/2020
P8K-8079-0264-6569
468f40b3-59ca-47fd-9739-c7f19cf48f32
Cat Cave
058G0P7V60
29.66
1
bedding
cat
29.66
2899
10/16/2020
P8K-8079-0264-6569
468f40b3-59ca-47fd-9739-c7f19cf48f32
Kitty Climber
W86BRJ9SSG
39.32
1
toy
cat
39.32
2900
10/16/2020
P8K-8079-0264-6569
468f40b3-59ca-47fd-9739-c7f19cf48f32
Fetch Blaster
M291KHJ4LW
29.47
1
toy
dog
29.47
2901
10/16/2020
P8K-8079-0264-6569
468f40b3-59ca-47fd-9739-c7f19cf48f32
Snoozer Essentails
GABWVMEL2R
28.04
1
bedding
dog
28.04
2902
12/10/2019
6ZD-7972-0320-6653
f2a090b3-ec77-4018-939e-1a18d2b4f4ef
Snoozer Essentails
GABWVMEL2R
28.04
1
bedding
dog
28.04
2714 rows × 10 columns
### Question 6: Identify the most common `Product_Category` purchased for each `Product_Line`. Then, assign the name of that category (as a string) to the variables common_category_cat and common_category_dog.
```python
# Grouping the DataFrame by 'Product_Line' and 'Product_Category' and summing the numeric columns (e.g., Quantity) to identify the most purchased item for each Product Line (cat and dog).
df.groupby(['Product_Line', 'Product_Category']).sum()
# Based on the aggregated results, assigning the most common product category (i.e., the one with the highest total items sold) for each product line to the respective variables.
common_category_cat = 'treat'
common_category_dog = 'bedding'
```
.dataframe tbody tr th:only-of-type {
vertical-align: middle;
}
.dataframe tbody tr th {
vertical-align: top;
}
.dataframe thead th {
text-align: right;
}
Date
Order_Number
Customer_ID
Product_Name
SKU
Price
Quantity
Subtotal
Product_Line
Product_Category
cat
bedding
9/21/20199/12/20209/14/20198/24/202111/6/20197...
ENT-5271-0660-48509K8-5765-4461-7159G7Y-9331-8...
d13c8860-67ae-44fb-8827-7afa9b4eec7a5c91c913-6...
Cat CaveSnoozer HammockSnoozer HammockCat Cave...
058G0P7V60V4B4RNS3ZPV4B4RNS3ZP058G0P7V60058G0P...
8371.70
526
16733.96
food
6/30/20219/14/20191/27/202010/9/20198/24/20217...
TCS-6223-7628-2720G7Y-9331-8313-29499FH-9267-0...
aacbf226-43d2-4b06-987b-5c4224c11b9d5bc15bf9-0...
Yum Fish-DishPurr MixYum Fish-DishPurr MixYum ...
GZCJZ3ET04O5FYJLBE0HGZCJZ3ET04O5FYJLBE0HGZCJZ3...
5957.41
422
11638.76
toy
5/22/20215/14/20209/12/20205/30/20219/14/20191...
SXF-7309-1727-1334VQE-2656-9729-61949K8-5765-4...
476582ea-1bba-4289-8775-3fcd8074821c041686b3-e...
Scratchy PostFoozy MouseFoozy MouseFoozy Mouse...
MPH6SCD7UT8PYSMLYINS8PYSMLYINS8PYSMLYINS8PYSML...
10116.60
851
19651.79
treat
5/22/20211/5/20205/14/202011/2/202012/2/20191/...
SXF-7309-1727-133414C-6286-0019-4676VQE-2656-9...
476582ea-1bba-4289-8775-3fcd8074821ce7b79f56-2...
Feline Fix MixSnack-em FishFeline Fix MixPurrf...
RKAPY3I1TPORGRBTIKZRRKAPY3I1TP28LQOI0LSKA8SU9C...
11453.96
1026
22817.43
dog
bedding
5/22/20213/23/20203/23/20205/16/20215/14/20209...
SXF-7309-1727-1334DG7-5410-5845-1340DG7-5410-5...
476582ea-1bba-4289-8775-3fcd8074821c5929a0e9-9...
Reddy BeddySnoozer EssentailsReddy BeddyReddy ...
DJWE1V9LZKGABWVMEL2RKDTMPSBZKZDJWE1V9LZKI2GQUN...
11906.83
953
24253.04
food
3/23/20205/14/20209/12/20205/30/20219/26/20203...
DG7-5410-5845-1340VQE-2656-9729-61949K8-5765-4...
5929a0e9-95a7-4dbf-896e-c11d1988615e041686b3-e...
Whole Chemistry RecipeWhole Chemistry RecipeWh...
6K4AUUS7306K4AUUS7306K4AUUS730NYW2F6CPBY6K4AUU...
4593.70
505
9022.00
toy
1/5/20201/5/20206/30/20216/22/20217/25/20219/2...
14C-6286-0019-467614C-6286-0019-4676TCS-6223-7...
e7b79f56-2196-49a0-852a-3d329ad7cb57e7b79f56-2...
Tug-a-BackChomp-a PlushFetch BlasterChomp-a Pl...
IZBHF5KR793HDX5H4WTMM291KHJ4LW3HDX5H4WTMIZBHF5...
9746.13
662
19605.11
treat
9/21/20199/14/20195/27/20217/7/20204/3/20195/2...
ENT-5271-0660-4850G7Y-9331-8313-2949R3O-6541-4...
d13c8860-67ae-44fb-8827-7afa9b4eec7a5bc15bf9-0...
Chewie DentalChewie DentalAll Veggie YummiesAl...
CG3531YP08CG3531YP08OWFPW3WZHGOWFPW3WZHGOWFPW3...
7409.18
453
14207.34
### Question 7: Determine which categories (`Product_Category`) by `Product_Line` have the median highest `Price`. Assign the (string) name of these categories to their respective variables priciest_category_cat and priciest_category_dog.
```python
# Grouping the DataFrame by 'Product_Line' and 'Product_Category' and calculating the median 'Price' for each group to identify which product category has the highest median price within each Product Line (cat and dog).
df.groupby(['Product_Line','Product_Category'])['Price'].median()
# Based on the median prices calculated above, assigning the name of the product category with the highest median price for each product line to the respective variables.
priciest_category_cat = 'bedding'
priciest_category_dog = 'toy'
```
Product_Line Product_Category
cat bedding 29.66
food 24.53
toy 16.71
treat 19.96
dog bedding 28.04
food 18.53
toy 29.47
treat 25.48
Name: Price, dtype: float64
# Data Visualization
Using the matplotlib and seaborn libraries to create visualizations.
### Question 8: You want to emphasize to your stakeholders that the total number of product categories sold differ between the two `Product_Line` categories ('cat' & 'dog'). Create a horizontal bar plot that has `Product_Category` on the y-axis and the total number of that category sold (using the `Quantity`) by each `Product_Line` category. Also change the axis labels to something meaningful and add a title.
```python
# Creating a horizontal bar plot that meets the question requirements.
ax = sns.barplot(data=df, y='Product_Category', x='Quantity', estimator=sum, ci=None, hue='Product_Line')
# Setting a meaningful title and labels.
ax.set_ylabel('Product Category')
ax.set_xlabel ('Total Products Sold')
ax.set_title = ('Total Number of Products Sold')
```
/var/folders/73/442cck753ndc4t4bclgcnbr80000gn/T/ipykernel_41509/1645217559.py:2: FutureWarning:
The `ci` parameter is deprecated. Use `errorbar=None` for the same effect.
ax = sns.barplot(data=df, y='Product_Category', x='Quantity', estimator=sum, ci=None, hue='Product_Line')
### Question 9: Based on the plot from Question 8, what would you conclude for your stakeholders about what products they should sell? What would be the considerations and/or caveats you'd communicate to your stakeholders?
```python
answer_to_9 = '''
Based on the visualization, I would advise stakeholders to focus on treats for cats and bedding for dogs, as these appear to be the most popular products within their respective product lines, with toys also performing well. However, it's important to note that this analysis is based on a single dataset, which may not be the most up-to-date or account for seasonal trends.
'''
```
### Question 10: Create an explanatory visualization that gives business stakeholders deeper insights into product sales trends.
```python
# Creating a box plot
ax = df.Price.plot.box()
answer_to_10 = '''
I created a box plot to illustrate the distribution of product prices. The plot reveals that most product prices fall between approximately $17 and $35, with some notable outliers—prices as high as around $40 and as low as $11 or $12. This visualization not only highlights the central tendency and variability in pricing but also helps stakeholders understand the overall price range of the products. Such insights can inform pricing strategies, inventory decisions, and identify opportunities for adjustments to maximize revenue.
'''
```
