Lecture 4

Business Intelligence; Data Analytics in the Retail Sector; Data Analytics vs. Data Science; RStudio

Byeong-Hak Choe

bchoe@geneseo.edu

SUNY Geneseo

September 4, 2024

A Framework for Business Intelligence (BI)

What is BI?

• Business Intelligence (BI) is a process that involves the collection, integration, analysis, and presentation of business data to support better decision-making.

• How Does It Work?

  • Data Collection: Gathering data from various sources, like databases.
  • Data Integration: Combining data into a unified view.
  • Data Analysis: Identifying trends and insights from data.
  • Reporting and Visualization: Displaying data via dashboards, reports, and visualizations.

Popular BI tools

• Python and R can also be BI tools.

What is a Dashboard?

• A dashboard is a visual display of key information, data, and metrics, presented in a way that is easy to read and interpret at a glance.

• Dashboard Examples:

  • Microsoft Power BI
  • Tableau
  • Looker
  • Python and R

Purpose of ML and BI in Business

• Informed Decision-Making: ML and BI provides insights from data to support strategic and operational decisions.

• Performance Monitoring: Tracks key performance indicators and metrics to ensure goals are met.

• Data Visualization: Transforms complex data into easy-to-understand visual formats like dashboards and reports.

• Operational Efficiency: Identifies inefficiencies and areas for improvement within business processes.

Data Analytics in the Retail Sector

Data Analytics in the Retail Sector

• The retail sector is heavily invested in analytics due to large volumes, thin margins, and rapidly changing customer preferences.

• Retailers like Amazon, Walmart, and Target have made significant investments in analytics to optimize their value chains.

• Analytics are crucial for understanding suppliers, customers, employees, and stakeholders to make better data-driven decisions.

• The extensive use of retail analytics by companies like Amazon highlights the potential for numerous applications, many of which are discussed throughout the source material.

Data Analytics Applications in the Retail Sector

1. Inventory Optimization

• Key Questions:
  1. Which products are in high demand?
  2. Which products are slow to sell or becoming outdated?
• Business Benefits:
  1. Predict demand for fast-moving products and ensure adequate stock to prevent shortages.
  2. Accelerate the turnover of slow-moving products by pairing them with high-demand items.

Data Analytics Applications in the Retail Sector

2. Price Elasticity

• Key Questions:
  1. What is the net margin for this product?
  2. What discount can be offered on this product?
• Business Benefits:
  1. Optimize markdowns for each product to minimize margin losses.
  2. Determine the best price for product bundles to protect margins.
  • While online retailers can tailor user experiences upon login, physical stores struggle with this at the entrance.

Data Analytics Applications in the Retail Sector

3. Market Basket Analysis

• Key Questions:
  1. Which products should be combined to create a bundle offer?
  2. Should bundles mix slow-moving and fast-moving items?
  3. Should bundles consist of products from the same or different categories?
• Business Benefits:
  1. Machine learning uncovers hidden product correlations, enabling:
  1. Strategic product bundling focused on inventory management or margin optimization.
  2. Enhanced cross-selling or upselling through bundling products from different or the same categories, respectively.

Data Analytics Applications in the Retail Sector

4. Shopper Insight

• Key Questions:
  1. Which customers are buying which products at which locations?
• Business Benefits:
  1. By segmenting customers, businesses can create personalized offers, leading to improved customer experience and retention.

Data Analytics Applications in the Retail Sector

5. Customer Churn Analysis

• Key Questions:
  1. Which customers are likely not to return?
  2. How much revenue could be lost?
  3. How can these customers be retained?
  4. What demographic is most loyal?
• Business Benefits:
  1. Identify customer and product relationships with high churn to focus on improving product quality and addressing churn reasons.
  2. Use customer lifetime value (LTV) to target marketing efforts, enhancing customer retention.

Data Analytics Applications in the Retail Sector

6. New Store Analysis

• Key Questions:
  1. Where should a new store be located?
  2. What and how much initial inventory should be stocked?
• Business Benefits:
  1. Leverage best practices from other locations and channels to ensure a strong start.
  2. Compare competitor data to develop a unique selling proposition (USP) that attracts new customers.

Data Analytics Applications in the Retail Sector

7. Store Layout

• Key Questions:
  1. How should the store be laid out to maximize sales?
  2. How can in-store customer experience be enhanced?
• Business Benefits:
  1. Analyze product associations to optimize store layout for customer satisfaction.
  2. Plan workforce deployment to enhance customer interactions and improve overall experience.

Data Analytics Applications in the Retail Sector

8. Video Analytics

• Key Questions:
  1. What demographic is entering the store during peak sales periods?
  2. How can a high LTV customer be identified at the store entrance for a personalized experience?
• Business Benefits:
  1. Plan in-store promotions and events based on the demographics of incoming traffic.
  2. Enhance customer experience through targeted engagement and instant discounts, leading to higher retention rates.

Data Analytics vs. Data Science

Data Analytics vs. Data Science

Focus and Key Activities

• Data Analytics:
  • Primarily focuses on examining existing datasets to generate insights.
  • It involves exploratory data analysis, which help in understanding historical data and identifying patterns or trends.
• Data Science:
  • Encompasses a broader scope, including data analytics, but extends to predictive analysis.
  • Data science aims to uncover insights from data through advanced statistical methods and machine learning (ML) methods.

Data Analytics vs. Data Science

Key Activities

• Data Analytics:
  • Involves tidying, transforming, and visualizing data.
  • The goal is to provide actionable insights based on existing data, often through dashboards, reports, or visualizations.
• Data Science:
  • Involves data collection, exploration, and the use of advanced techniques like ML.
  • Data scientists often work on building ML models that predict future trends or optimize decisions.

Data Analytics vs. Data Science

Skill Sets

• Data Analytics:
  • Requires skills in tools like Excel, SQL, and BI software.
  • It may also involve basic statistical knowledge and proficiency in creating data visualizations.
• Data Science:
  • Requires a deeper understanding of programming languages like Python or R, advanced statistical methods and machine learning algorithms.
  • Data scientists often need to have strong mathematical and computational skills.

Data Analytics vs. Data Science

Tools & Techniques

• Data Analytics:
  • Commonly uses tools like SQL, Tableau, Power BI, and Excel.
  • The focus is on querying data and creating reports or visualizations.
• Data Science:
  • Utilizes more advanced tools and techniques such as Python and R with ML-related tools.
  • Techniques include data analytics, machine learning, and algorithm development.

Data Analytics vs. Data Science

Applications

• Data Analytics:
  • Often used in business intelligence, reporting, trend analysis, and operational decision-making.
• Data Science:
  • Applied in areas requiring predictive insights, such as recommendation systems, fraud detection, and customer segmentation.

Data Analytics vs. Data Science

Job Titles

• Data Analytics:
  • Roles often include Data Analyst, Business Analyst, BI Analyst, or Reporting Analyst.
• Data Science:
  • Roles include Data Scientist, Machine Learning Engineer, Data Engineer, and AI Specialist.

Data Analytics vs. Data Science

Fuzzy Distinctions

• The difference between professionals focusing on descriptive analytics and those engaged in all types of analytics is not clear-cut.

• Graduates of data-related programs often perform tasks aligned with data science.