Introduction
The purpose of this report is to address the three issues assigned by the Corporate Strategy
Manager of COTS coffee shops. The report presents a project plan for delivering the research project,
and a data analytics framework is used to address the core business questions. The report also
highlights the Key Performance Indicators (KPIs) for COTS's coffee shops, which can be improved
through better analytics.
Project Plan
The project plan consists of five stages, namely:
1. Data Preparation: In this stage, we will collect, integrate, and clean the data to make it
suitable for analysis.
2. Descriptive Analytics: In this stage, we will use descriptive statistics and data visualization
techniques to understand the sales volume and value of the three coffee shops and identify
the best shop to invest in expanding its floor area.
3. Product Offering Analysis: In this stage, we will analyze the product offering to identify the
products with the worst sales performance that are candidates to be removed from the
shops' menu.
4. Impact of Home Delivery Service: In this stage, we will evaluate the impact of home delivery
service on sales performance by analyzing the data from the Blackpool shop.
5. Conclusion and Recommendations: In this stage, we will summarize the findings and provide
recommendations for COTS coffee shops.
Data Analytics Framework
We will use the CRISP-DM (Cross-Industry Standard Process for Data Mining) framework for this
project. The CRISP-DM framework consists of six stages, namely:
1. Business Understanding: In this stage, we will define the business objectives and
requirements and translate them into data analytics questions.
2. Data Understanding: In this stage, we will collect and explore the data to understand its
quality and structure.
3. Data Preparation: In this stage, we will clean, transform, and integrate the data to make it
suitable for analysis.
4. Modeling: In this stage, we will apply statistical and machine learning techniques to the data
to address the business questions.
5. Evaluation: In this stage, we will evaluate the models' performance and validate their results.
6. Deployment: In this stage, we will deploy the models and integrate them into COTS coffee
shops' decision-making processes.
Value of Data Analytics
Data analytics can add value to COTS coffee shops in several ways, including:
1. Identifying the best shop to invest in expanding its floor area by analyzing sales volume and
value.
2. Improving profitability by reducing the cost of ingredients used to produce the menu by
analyzing the product offering and removing the products with the worst sales performance.
3. Evaluating the impact of home delivery service on sales performance and expanding the
service to other shops by analyzing the data from the Blackpool shop.
4. Enhancing the overall customer experience by analyzing customer preferences and behavior.
Task 2: Data Preparation Quality Issues and Remedies
Generic Data Problems: Data analysts encounter several generic data problems, such as:
1. Missing Values: Missing data can reduce the accuracy of the analysis. We can address this
issue by either deleting the rows with missing values, imputing the missing values using
mean or median values, or using advanced imputation techniques like K-Nearest Neighbors.
2. Outliers: Outliers are extreme values that can distort the analysis. We can address this issue
by either deleting the outliers, transforming the data using logarithmic or exponential
functions, or using advanced outlier detection techniques like Z-Score or Box Plot.
3. Inconsistent Data Types: Inconsistent data types can make it difficult to analyze the data. We
can address this issue by converting the data into a consistent data type, such as converting
string data into numerical data.
Specific Data Problems in COTS Dataset: After analyzing the COTS dataset, we have identified the
following
