C11BD Coursework

Submitted By

Oluwayomi Fadekemi Ojo

Student ID:H00447358

Date: March 18, 2024

Report title: Profitability Analysis and Insights for Superstore Business Growth

Executive Summary

The analysis of Superstore's dataset reveals insightful trends in profitability, customer segmentation, and regional performance and some patterns in its clusters. The business demonstrates growth and resilience, with certain segments and regions outperforming others. While product categories and customer segments varied in profitability, opportunities for improvement exist across the board. Recommendations include segment-specific strategies, proper segmentation, prioritizing customer satisfaction and clear data quality standards and continuous monitoring of profitability metrics. Regional strategies should focus on capitalizing on growth opportunities and addressing region-specific challenges.

Introduction

In this era of data proliferation and rapidly evolving environment, where businesses and industries are inundated with enormous amounts of information (Strydom & Buckley, 2020), data analytics has become an important tool for companies seeking to gain a competitive advantage and drive sustainable growth (Kaufmann, 2021). By harnessing and understanding this data, organisations can unlock valuable insights that guide strategic decision making and propel them towards achieving their strategic goals (Gyamfi & Williams, 2019). The profound potential of data analytics lies in its ability to sift through immense volumes of information to reveal hidden patterns (R. Sujatha et al., 2021), show correlations, and trends that may be hidden from traditional analysis. By leveraging advanced analytics techniques such as machine learning, predictive modelling, and data mining, organizations can distil complex datasets into actionable intelligence, guiding strategic initiatives and driving bottom-line results (Baesens, 2014).

Methodology

Considering the above, this report takes an in-depth approach to analyse Superstore’s dataset with the aim to uncovering hidden patterns, trends, and opportunities within the dataset that can drive profitability and foster business success. From identifying data entry errors and This report will take a systematic approach encompassing data collection, Data Cleaning and Preprocessing that involves a rigorous cleaning and preprocessing to remove inconsistencies, missing values, or outliers that may skew the analysis of our result. Exploratory data analysis is performed to gain insights into the distribution and relationships among different variables in the dataset. The data is segmented based on product categories, regions, and customer segments to analyse profitability trends within each segment. Comparisons are made across segments to identify areas of strength and opportunities for improvement. Continuous Monitoring and Optimization of profitability metrics will be done and tracked to identify emerging trends and areas needing corrective action and Insights from the analysis are used to optimize business strategies, pricing decisions, product offerings, and resource allocation.

Objective

The primary objective of this report is to conduct profitability analysis. This entails examining the profitability trends across different product categories, regions, and customer segments. By analysing sales revenue, profit margins, and associated costs, the report aims to identify areas of strength and opportunities for improvement within the retail operations of the Superstore.

#To begin, we have to import pandas # I'm assigning this to an object called Superstore #I am importing NumPy and matploblib and seaborn because of my large dataset and # analysis that include some mathematical function and graphs informative statistical graphics. import pandas as pd import numpy as np import matplotlib.pyplot as plt import seaborn as sns Superstore = pd.read_csv("Superstore_dataset.csv", parse_dates=['Order Date']) #I have used parse_dates in my code because I need for python to recognise the Order Date as dates. Superstore

Run to view results

#In order to quickly understand the structure of this dataset, #including the number of columns, their data types, and whether they contain missing values, #it is imperative that the .info() function is used. Superstore.info()

Run to view results

# To perform some exploratory data analysis in order to # understand the composition of the dataset and identify any patterns or trends. ship_mode_counts = Superstore['Ship Mode'].value_counts() segment_counts = Superstore['Segment'].value_counts() category_counts =Superstore['Category'].value_counts() sub_category_counts = Superstore['Sub-Category'].value_counts() region_counts = Superstore['Region'].value_counts() state_counts = Superstore['State'].value_counts() city_counts = Superstore ['City'].value_counts() productname_counts = Superstore['Product Name']. value_counts() print(ship_mode_counts) print(segment_counts) print(category_counts) print(sub_category_counts) print(region_counts)

Run to view results

Analysis

This statistical description summarises the information for Superstore. This shows that the company's profit is highly competitive despite the seemingly huge amount of losses generated from some sales. Statistical investigation shows that Superstore generated an approximate profit mean of 28.66 on sales mean of 229.88, generating a profit margin of 12.46% over the years. that is:

Profit Margin = (Profit/Sales) X 100 Therefore: (28.66/229.86) X 100 = 12.46%

This margin of 12.47% means that for every dollar of sales revenue generated, Superstore retains approximately $0.1247 as profit. This suggests that the profit margin is not below par given that the retail grocery business has weak profit margins typically between an average of one and two percent (Dorich, 2021). The description below also shows an average of 15.6% for every sale made. Given the huge difference between the minimum and the maximum value in the variables shown in the data below, it is imperative that we identify and take out outliers to avoid any bias estimates.

Outliers

The analysis identifies outliers in the dataset across different metrics. Firstly, outliers in the Discount variable are observed, with two extreme values exceeding 100% discount, attributed to rare sales occurrences in the corporate segment during specific years. Similarly, in Profit, three exceptionally high profits and one significant loss are considered outliers, impacting the calculated mean. Additionally, Quantity outliers, represented by values exceeding 14, are removed, notably impacting the mean quantity. Lastly, Sales outliers, identified as values surpassing 12,000, are linked to infrequent purchases by different customer segments across specific years, leading to a reduction in the mean sales value. Overall, addressing these outliers ensures a more accurate representation of central tendency and enhances the reliability of statistical analysis.

#in order to generate descriptive statistics of the numerical columns #in the given dataset, there is need to use the .describe() function Superstore.describe()

Run to view results

#From the description above, there is a huge difference between the min and the max of my discount. #Besides, 1.3 is considered too much of a discount to give a customer. #This means that customers would be paid to take the product receiving an amount equivalent to 130% #of the original price. From a business perspective, offering such a discount is not financially viable and would lead to significant losses. #Therefore, I will be considering any value greater than 0.8 as outliers for discount. fig, ax = plt.subplots() ax.scatter(Superstore['Discount'],Superstore['Profit']) plt.xlabel('Discount') plt.ylabel('Profit') plt.show()

Run to view results

#Quantity that are 10,000 and above will be considered an outliers as this will greatly distort the mean(average) #of Superstore's regular quantity. Looking at the dataset, quantity jumped from 14 to 10,000 and only #purchased by three customers just in 2014. No such quantity was ordered in 2015,2016 and 2017 that this dataset covers #Again, profits above 5000 will also be considered as an outlier. It only occured once every year entire 4years under consideration fig, ax = plt.subplots() ax.scatter(Superstore['Quantity'],Superstore['Profit']) plt.xlabel('Quantity') plt.ylabel('Profit') plt.show()

Run to view results

#Dropping sales greater than 12,000 because they seem to only occur once a year. #Including this in our analysis may distort our average thereby giving us inaccurate conlusions. fig, ax = plt.subplots() ax.scatter(Superstore['Sales'],Superstore['Quantity']) plt.xlabel('Sales') plt.ylabel('Quantity') plt.show()

Run to view results

#Dropping the outliers in Discount Superstore2= Superstore.drop(Superstore.index[(Superstore.Discount>1.0)]) Superstore2

Run to view results

#Dropping the outliers in Quantity Superstore3 = Superstore2.drop(Superstore2.index[(Superstore2.Quantity>14)]) Superstore3

Run to view results

#Dropping the outliers in Sales Superstore4 = Superstore3.drop(Superstore3.index[(Superstore3.Sales>12000)]) Superstore4

Run to view results

#Dropping the outliers in Profit and calling up the new dataframe Superstore_New = Superstore4.drop(Superstore4[Superstore4['Profit'] > 5000].index) Superstore_New

Run to view results

#There is a need to describe the new dataframe # This new description shows a significant drop in the mean and standard deviation of the original dataset. Superstore_New.describe()

Run to view results

Visualisation of Quantity purchased vs Sales, Profit and Discount

The table below shows relationship between the quantity of items purchased and their corresponding sales, profit, and discount values. There seem to be a common pattern among most of the data points regarding the relationship between quantity, sales, profit and discount. This implies that for most transactions, changes in quantity are associated with relatively consistent changes in sales, profit, and discount.

# Plotting scatter plot for Quantity, Sales, Profit, and Discount to see the relationship. plt.figure(figsize=(10, 6)) plt.scatter(Superstore_New['Quantity'], Superstore_New['Sales'], label='Sales', alpha=0.7) plt.scatter(Superstore_New['Quantity'], Superstore_New['Profit'], label='Profit', alpha=0.7) plt.scatter(Superstore_New['Quantity'], Superstore_New['Discount'], label='Discount', alpha=0.7) plt.xlabel('Quantity') plt.ylabel('Values') plt.title('Scatter plot of Quantity vs. Sales, Profit, Discount') plt.legend() plt.grid(True) plt.show()

Run to view results

#This is to show the profit that are greater than zero by region, shipmode of products, segment, category Profitby_Category_Region = Superstore_New[Superstore_New['Profit'] >0][['Profit', 'Segment','Order Date', 'Region', 'Category', 'Ship Mode']] #I want to convert this to a dataframe in order to visualise it. Profitby_Category_Region = pd.DataFrame(Profitby_Category_Region) Profitby_Category_Region

Run to view results

Product Analysis: Profitability by Product categories

The table and visualisation below help to understand the profit trends across Superstore's different product categories over time. Furniture has consistently performed lower in terms of profitability compared to the other categories. It only experiences some spikes around the same period every year which might be due factors such as seasonality. Office supplies and technology seem to be performing at par. While Office supplies experience a sharp growth in November 2016, it also had a sudden fall in January 2017 when all the product categories fell. This fall in all categories might be a result of intense competitive pressure.

Profitby_Category_Region['MonthYear'] = Profitby_Category_Region['Order Date'].dt.strftime('%Y%m') Profit_Pivot = Profitby_Category_Region.pivot_table(values='Profit', index='MonthYear', columns = 'Category', aggfunc='sum') Profit_Pivot.reset_index (inplace=True) Profit_Pivot

Run to view results

#Creating a line chart to see how the categories of product have behaved over the last four years. i.e, 2014 to 2017 plt.figure(figsize=(10, 6)) for category in Profit_Pivot.columns[1:]: plt.plot(Profit_Pivot['MonthYear'], Profit_Pivot[category], label=category) plt.xlabel('MonthYear') plt.ylabel('Profit') plt.title('Profit by Category over Time') plt.legend() plt.xticks(rotation=45) plt.tight_layout() plt.show()

Run to view results

Segment Analysis: Profitability by Segment in relation to discount

To understand the profitability of this segments, an analysis of the discount received by each segment is also done to enable us to understand the relationship between discount and profitability. The profit percentage for the Consumer segment has decreased from 54% in 2014 to 41% in 2017 even when the discount given to customers increased tremendously. Despite this increase in discounts over the years, the profit percentage has declined for the Consumer segment. This suggests that the increased discounts may not have translated into higher sales or revenue growth. Other factors such as changes in consumer preferences, competition, or operational inefficiencies could have contributed to the decline in profitability.

For Home Office Segment, the profit percentage for the Home Office segment increased from 18% in 2014 to 24.9% in 2017 when discount also increased. The Home Office segment experienced an improvement in profitability despite increased discount allowed. This could indicate that the segment has effectively managed its costs or targeting higher-margin products and customers.

Corporate segments experience a 25% growth between 2014 and 2017 despite the constant increment in discount given to its customers. Like the Home Office segment. This could be due to strategic pricing or focusing on high-value customers within the segment.

The loss situation is degenerating YoY between Corporate and Home office while consumer segment seems to be making effort in taming its loss. Still Consumer segment still has the highest amount of loss of these three segments.

#Filter of data with sales and discounts that made profit Profit_discounts_sales = Superstore_New[Superstore_New['Profit'] >0][['Discount', 'Sales', 'Quantity','Profit', 'Segment','Order Date', 'Region', 'Category', 'Ship Mode']] #I want to convert this to a dataframe in order to visualise it. Profit_discounts_sales = pd.DataFrame(Profit_discounts_sales) Profit_discounts_sales

Run to view results

#Pivot table to show the Segment that received the most discount Profit_discounts_sales['MonthYear'] = Profit_discounts_sales['Order Date'].dt.strftime('%Y') Discounted_Segment= Profit_discounts_sales.pivot_table(values= 'Discount', index='MonthYear', columns = 'Segment', aggfunc='sum') Discounted_Segment.reset_index (inplace=True) Discounted_Segment

Run to view results

#Visualisation of the segment that received the most discount in the last 4 years. i.e 2014 - 2017 Discounted_Segment.plot(kind='bar', x='MonthYear', stacked=False, figsize=(10, 6)) plt.xlabel('Month-Year') plt.ylabel('Total Discount') plt.title('Total Discount by Segment for Each Month-Year') plt.legend(title='Segment', bbox_to_anchor=(1.05, 1), loc='upper left') plt.xticks(rotation=45) plt.tight_layout() plt.show()

Run to view results

#Pivot table to show the Segment that received the most profit in relation to the discounts above Profit_discounts_sales['MonthYear'] = Profit_discounts_sales['Order Date'].dt.strftime('%Y') Segment_Profit= Profit_discounts_sales.pivot_table(values= 'Profit', index='MonthYear', columns = 'Segment', aggfunc='sum') Segment_Profit.reset_index (inplace=True) Segment_Profit

Run to view results

# Visualisation with pie chart of the profit of each segment in the last four years. for i, row in Segment_Profit.iterrows(): labels = row.index[1:] values = row[1:] plt.figure(figsize=(6, 6)) plt.pie(values, labels=labels, autopct='%1.1f%%', startangle=140) plt.title(f'Profit Distribution by Segment ({row["MonthYear"]})') plt.axis('equal') # Equal aspect ratio ensures that pie is drawn as a circle plt.show()

Run to view results

#Filter of discounts that led to loss of profit loss_discounts_sales = Superstore_New[Superstore_New['Profit'] < 0][['Discount', 'Sales', 'Quantity','Profit', 'Segment','Order Date', 'Region', 'Category', 'Ship Mode']] #I want to convert this to a dataframe in order to visualise it. loss_discounts_sales = pd.DataFrame(loss_discounts_sales) # There is a need to rename the Profit column to loss loss_discounts_sales.rename(columns={'Profit': 'Loss'}, inplace=True) loss_discounts_sales

Run to view results

#There is a need to get the highest loss making segment too loss_discounts_sales['MonthYear'] = loss_discounts_sales['Order Date'].dt.strftime('%Y') Segment_Loss= loss_discounts_sales.pivot_table(values= 'Loss', index='MonthYear', columns = 'Segment', aggfunc='sum') Segment_Loss.reset_index (inplace=True) Segment_Loss

Run to view results

Region Analysis: Profit by Region

The profitability of the Central region shows steady growth from 2014 to 2017. Profit increases from approximately $15,657 in 2014 to around $28,060 in 2017, indicating consistent growth over the period. This region demonstrates a positive trend in profitability. Like the central region, the east shows upward profitability trends with a steady rise from about $25,458 in 2014 to $40,087 in 2017. The profitability of the South region fluctuates over the years, with some variability in performance. Profit peaks in 2016 at approximately $20,766 but declines slightly in 2017 to around $21,214. While the South region experiences fluctuations, overall profitability remains relatively stable, suggesting a balanced performance. The West region demonstrates the most significant growth in profitability among all regions. Profit nearly doubles from approximately $23,346 in 2014 to over $46,155 in 2017. This region shows the highest profitability and the most robust growth trajectory, making it a key driver of overall profitability for the organization.

Despite the YoY profit growth experienced across the regions, they all had their losses increased in 2017. This suggests systemic challenges or external factors affecting the retail sector rather than region-specific issues.

##This aspect shows the region with the most Profit Profitby_Category_Region['MonthYear'] = Profitby_Category_Region['Order Date'].dt.strftime('%Y%') Region_Profit_Pivot = Profitby_Category_Region.pivot_table(values='Profit', index='MonthYear', columns = 'Region', aggfunc='sum') Region_Profit_Pivot.reset_index (inplace=True) Region_Profit_Pivot

Run to view results

#Visualising the regions in terms of profit since 2014 Region_Profit_Pivot.set_index('MonthYear', inplace=True) Region_Profit_Pivot.plot(kind='barh', figsize=(10, 8)) plt.xlabel('Profit') plt.ylabel('Year') plt.title('Profit by Region Over Time') plt.legend(title='Region', bbox_to_anchor=(1.05, 1), loc='upper left') plt.show()

Run to view results

#Examining the region with the most loss loss_discounts_sales['MonthYear'] = loss_discounts_sales['Order Date'].dt.strftime('%Y') Region_Loss_Pivot= loss_discounts_sales.pivot_table(values= 'Loss', index='MonthYear', columns = 'Region', aggfunc='sum') Region_Loss_Pivot.reset_index (inplace=True) Region_Loss_Pivot

Run to view results

#A brief statistical description of the regions losses from 2014 to 2017 Region_Loss_Pivot.describe()

Run to view results

Customer Analysis: Profit by customers vis -a-vis the quantity purchased.

For Superstore to make informed decisions to enhance profitability and customer satisfaction, it is imperative that the behaviour and financial performance of the customers are understood. The result below shows the top twenty customers that make the most profit. It is observed that the top twenty customers, despite purchasing lesser quantities, are highly profitable for the company. Despite their lower purchase volumes, they contribute significantly to the company's overall profitability. They are all well distributed across all four segments, which indicates that profitability is not limited to a specific customer segment.

A quick comparison of the top twenty customers and the entire customer base of superstore reveals that the top twenty customers have a significantly higher mean profit ($2960.60) compared to the overall dataset ($26.84). There is no difference in the mean sales ($223.40) and discount (0.156) between the top twenty customers and the overall dataset. The top twenty customers purchase a higher mean quantity (64.95) compared to the overall dataset (3.79). This indicates that the top twenty customers contribute substantially more to the overall profit of the Superstore, purchase larger quantities of products, but receive similar discounts and make similar sales compared to the rest of the customers in the dataset.

#Grouping the customers by the number quantity and profitability that they bring to the company Grouped_Customers = Superstore_New.groupby('Customer Name')[['Quantity', 'Profit', 'Category', 'Returned','Segment']].sum().reset_index() # Because I want the profit to be greater than zero as it was returning negative profit for me Grouped_Customers = Grouped_Customers[Grouped_Customers['Profit'] > 0] # Trying to get the top 20 customers in order of their profitability Top_Customers = Grouped_Customers.sort_values(by='Profit', ascending=False).head(20) Top_Customers

Run to view results

#Need to visualise the top twenty customers giving Superstore the most profit in relation to the quantity boughty customer_names = Top_Customers['Customer Name'] profits = Top_Customers['Profit'] quantities = Top_Customers['Quantity'] # Creating the visualisation plt.figure(figsize=(12, 8)) plt.barh(customer_names, profits, color='purple', label='Profit') plt.barh(customer_names, quantities, color='blue', label='Quantity Purchased') plt.xlabel('Amount') plt.ylabel('Customer Name') plt.title('Top 20 Customers by Profitability and Quantity Purchased') plt.legend() plt.gca().invert_yaxis() # Invert y-axis to have the highest profit at the top plt.show()

Run to view results

#Description of the top twenty customer showingb their mean and average Top_Customers.describe()

Run to view results

#module because it provides an implementation of the K-means clustering algorithm. from sklearn.cluster import KMeans

Run to view results

Modelling with Kmeans Clustering

The first point on the Kmeans has a high inertia indicating that the data points in the cluster are distributed. over a larger area, and the cluster center is relatively distant from the data points. This means that they are not tightly grouped around the central point. 5 clusters have been chosen because that indicates a point K where the sum of squared errors begin to decline. From the visualisation below, it can be said that the data points within each cluster are closely packed together in the feature space. There is a little spread of variability among the data points which suggests that Superstores Sales, Discount, Profit and Quantity all share similar characteristics. The mean value of these clusters indivate the relative sizes. Clusters with higher mean value have more datapoints allocated to them while those with lower mean values have fewer data points. In this case, Kmeans_5 followed by KMeans_3 appear to be the largest.

#Creating a new data frame from my original data frame in order to run my modelling with Kmeans from sklearn.cluster import KMeans New_Superstore_Data = Superstore_New[['Sales', 'Profit','Discount', 'Returned']] New_Superstore_Data['Returned'] = New_Superstore_Data['Returned'] * 1 New_Superstore_Data

Run to view results

#generating descriptive statistics of the numerical columns in the DataFrame New_Superstore_Data.describe()

Run to view results

#need to fit the standard scaler to the datanand then transform. this is necessary #in order to reduce the influence that some feature may have over the other to enable #us get an unbiased result from sklearn.preprocessing import StandardScaler scaler = StandardScaler() New_Superstore_Data.loc[:, ['Sales_T', 'Profit_T', 'Discount_T', 'Returned_T']] = scaler.fit_transform(New_Superstore_Data[['Sales', 'Profit', 'Discount', 'Returned']])

Run to view results

#There is a need to visualise the "elbow method" to determine the optimal number of clusters (K) for K-means #clustering on the Superstore dataset and choose the k for which WSS first starts to diminish. # WSS meands within-cluster sum of squares (WCSS) values for each value of K (number of clusters) in the range from 1 to 10, #calculated during the elbow method process. wcss = [] for k in range(1,11): kmeans = KMeans(n_clusters=k, init="k-means++") kmeans.fit(New_Superstore_Data) wcss.append(kmeans.inertia_) #visualizing the within-cluster sum of squares (WCSS) values for different values of K plt.figure(figsize=(12,6)) plt.grid() plt.plot(range(1,11),wcss, linewidth=2, color="red", marker ="8") plt.xlabel("K Value") plt.xticks(np.arange(1,11,1)) plt.ylabel("WCSS") plt.show()

Run to view results

#performing KMeans clustering on the New_Superstore_Data DataFrame with the specified number of clusters kmeans = KMeans(n_clusters = 5) kmeans.fit(New_Superstore_Data)

Run to view results

#assigning the cluster labels obtained from the KMeans clustering algorithm to a new column named 'kmeans_5' #in order to understand how the datapoints are grouped New_Superstore_Data.loc[:, 'kmeans_5'] = kmeans.labels_ New_Superstore_Data

Run to view results

#Initializing a KMeans clustering object #km with n_clusters set to 5, indicating that I want to find 5 clusters in the data. km=KMeans(n_clusters = 5) clusters = km.fit_predict(New_Superstore_Data) New_Superstore_Data.loc[:, "label"] = clusters #Creating a 3D scatter plot using Matplotlib #and plot the data points based on their "Sales", "Profit", and "Discount" values fig = plt.figure(figsize=(20,10)) ax = fig.add_subplot(111, projection='3d') ax.scatter(New_Superstore_Data['Sales'][New_Superstore_Data.label == 0], New_Superstore_Data["Profit"][New_Superstore_Data.label == 0], New_Superstore_Data["Discount"][New_Superstore_Data.label == 0], c='blue', s=60) ax.scatter(New_Superstore_Data['Sales'][New_Superstore_Data.label == 1], New_Superstore_Data["Profit"][New_Superstore_Data.label == 1], New_Superstore_Data["Discount"][New_Superstore_Data.label== 1], c='red', s=60) ax.scatter(New_Superstore_Data['Sales'][New_Superstore_Data.label == 2], New_Superstore_Data["Profit"][New_Superstore_Data.label == 2], New_Superstore_Data["Discount"][New_Superstore_Data.label == 2], c='green', s=60) ax.scatter(New_Superstore_Data['Sales'][New_Superstore_Data.label == 3], New_Superstore_Data["Profit"][New_Superstore_Data.label == 3], New_Superstore_Data["Discount"][New_Superstore_Data.label == 3], c='orange', s=60) ax.scatter(New_Superstore_Data['Sales'][New_Superstore_Data.label == 4],New_Superstore_Data["Profit"][New_Superstore_Data.label == 4], New_Superstore_Data["Discount"][New_Superstore_Data.label== 4], c='purple', s=60) ax.view_init(30, 185) plt.xlabel("Sales") plt.ylabel('Profit') ax.set_zlabel("Discount") plt.show()

Run to view results

##generating descriptive statistics of the numerical columns in the DataFrame New_Superstore_Data.describe()

Run to view results

Conclusion

The analysis of Superstore dataset has yielded useful information into the company's profitability trends, customer segmentation, and regional performance. Despite facing challenges such as fluctuating profitability and increasing losses in certain segments and regions, the company demonstrated overall growth and resilience over the years. While the profitability analysis revealed certain product categories and customer segments performed better than others, there are opportunities for improvement across the board. The correlation between discounts and profitability varied across segments, indicating the need for targeted pricing strategies and cost management initiatives. Regional performance analysis highlighted the West region as a significant driver of profitability, experiencing substantial growth over the years. However, all regions faced increased losses in 2017, suggesting systemic challenges impacting the retail sector.

Recommendation

Given the insights derived from this dataset, it is imperative to develop tailored strategies that will optimise the operations of Superstore. • Develop segment-specific strategies to address the variation in profitability trends observed across different product categories and customer segments. This will entail optimising product offerings and tailoring customer experience to each segment. • Implement cost management initiatives to mitigate the impact of discounts on profitability. This may include evaluating the effectiveness of discounting strategies, optimizing supply chain processes, and minimizing operational inefficiencies. • Prioritize customer satisfaction and retention efforts to cultivate long-term relationships with high-value customers. Leverage data analytics to personalize marketing efforts, tailor product recommendations, and enhance overall customer experience across all touchpoints. • Define clear data quality standards and guidelines for data collection, entry, storage, and maintenance. • Continuous monitoring and optimization of profitability metrics to identify emerging trends, address performance gaps, and seize opportunities proactively. • Focus on regional strategies to capitalize on growth opportunities and address challenges specific to each region. • Invest in market research to understand regional dynamics, consumer preferences, and competitive landscape to inform strategic decision-making.

References

Özemre, M. and Kabadurmus, O., 2020. A big data analytics based methodology for strategic decision making. Journal of Enterprise Information Management, 33(6), pp.1467-1490.

Kanungo, T., Mount, D.M., Netanyahu, N.S., Piatko, C.D., Silverman, R. and Wu, A.Y., 2002. An efficient k-means clustering algorithm: Analysis and implementation. IEEE transactions on pattern analysis and machine intelligence, 24(7), pp.881-892.

Aladžuz, A., Delalić, A. and Šćeta, L., 2022, May. Cluster Analysis in Python: An Example of Market Segmentation. In International Conference “New Technologies, Development and Applications” (pp. 1032-1041). Cham: Springer International Publishing.

Raschka, S. and Mirjalili, V., 2019. Python machine learning: Machine learning and deep learning with Python, scikit-learn, and TensorFlow 2. Packt publishing ltd.

Rogel-Salazar, J., 2023. Statistics and Data Visualisation with Python. Chapman and Hall/CRC.

Aladžuz, A., Delalić, A., & Šćeta, L. (2022, May). Cluster Analysis in Python: An Example of Market Segmentation. In International Conference “New Technologies, Development and Applications” (pp. 1032-1041). Cham: Springer International Publishing.

Baesens, B. (2014). Analytics in a big data world: The essential guide to data science and its applications. Hoboken, New Jersey: John Wiley & Sons, Inc.

Gyamfi, A., & Williams, I. (2019). Big Data and Knowledge Sharing in Virtual Organizations. IGI Global.

Kanungo, T., Mount, D. M., Netanyahu, N. S., Piatko, C. D., Silverman, R., & Wu, A. Y. (2002). An efficient k-means clustering algorithm: Analysis and implementation. IEEE transactions on pattern analysis and machine intelligence, 24(7), 881-892.

Kaufmann, A. B. (2021). DATA ANALYTICS FOR ORGANISATIONAL DEVELOPMENT: Unleashing the potential of your data. S.L.: John Wiley & Sons.

Özemre, M., & Kabadurmus, O. (2020). A big data analytics based methodology for strategic decision making. Journal of Enterprise Information Management, 33(6), 1467-1490.

Raschka, S., & Mirjalili, V. (2019). Python machine learning: Machine learning and deep learning with Python, scikit-learn, and TensorFlow 2. Packt publishing ltd.

Rogel-Salazar, J. (2023). Statistics and Data Visualisation with Python. Chapman and Hall/CRC.

Strydom, M., & Buckley, S. (2020). AI and big data’s potential for disruptive innovation. Hershey, Pennsylvania: IGI Global.

Sujatha, R., Aarthy, S. L., & Vettriselvan, R. (2021). Integrating Deep Learning Algorithms to Overcome Challenges in Big Data Analytics. CRC Press.

.css-15w88e5{color:var(--chakra-colors-fg-neutral-primary);font-weight:inherit;letter-spacing:-0.09px;}

C11BD Coursework

Oluwayomi Fadekemi Ojo

Report title: Profitability Analysis and Insights for Superstore Business Growth

Executive Summary

Introduction

Methodology

Objective

Analysis

Visualisation of Quantity purchased vs Sales, Profit and Discount

Product Analysis: Profitability by Product categories

Segment Analysis: Profitability by Segment in relation to discount

Region Analysis: Profit by Region

Customer Analysis: Profit by customers vis -a-vis the quantity purchased.

Modelling with Kmeans Clustering

Conclusion

Recommendation

References