1.

Evolution of Analytics:
Introduction to Business Analytics

Business analytics refers to the practice of using data analysis, statistical techniques, and Software
tools to improve business decision-making. It aims to enhance operational efficiency, identify market
trends, optimize various business processes, and support strategic decisions through data-driven
insights.

Evolution Timeline

1970-1980: Decision Support Systems (DSS)

During the 1970s, Decision Support Systems (DSS) emerged as computerized systems designed to
assist in complex decision-making. These systems comprised three primary components: database
management, model management, and user interfaces.

Database management facilitated the storage and retrieval of relevant data, while model
management involved the application of analytical models to the data. The user interface allowed
users to interact with the system, retrieve information, and analyze data. Technologies of this era
included mainframes and minicomputers, which enabled the processing and storage of data,
alongside query languages like SQL for database interactions. DSS were utilized in areas such as
inventory control, scheduling, and resource allocation.

1980-1990: Executive Information Systems (EIS)

The 1980s saw the advent of Executive Information Systems (EIS), which were designed to provide
top executives with easy access to critical data. EIS featured user-friendly interfaces, typically in the
form of dashboards and graphical displays. These systems offered drill-down capabilities, allowing
users to explore data hierarchically, and trend analysis for identifying and visualizing business trends.

Technologies in this period included graphical user interfaces (GUIs) and early forms of data
visualization, which made data analysis more intuitive for executives. EIS were employed in strategic
planning, financial analysis, and competitive intelligence to aid in making informed strategic decisions.

1990-2000: Business Intelligence (BI)

Ramaraju Poosapati Business Analytics For year 2024

The 1990s marked the rise of Business Intelligence (BI), a domain focused on transforming raw data
into actionable insights for strategic decision-making. BI systems encompassed data warehousing,
online analytical processing (OLAP), and data mining.

Data warehousing involved the centralized storage of large volumes of data, while OLAP tools
facilitated multidimensional data analysis. Data mining utilized statistical and analytical tools to
extract patterns and knowledge from large datasets. Technologies of this period included ETL (Extract,
Transform, Load) processes for integrating data, relational databases, and data marts. BI was applied
in performance management, market analysis, and customer insights to drive business strategies.

2000-2010: Analytics

The early 2000s introduced advanced analytics, focusing on deriving actionable insights using
sophisticated statistical techniques. Analytics evolved into three main categories: descriptive,
predictive, and prescriptive analytics.

Descriptive analytics aimed to understand past performance by analyzing historical data to determine
what happened. Predictive analytics used statistical algorithms and machine learning to forecast
future events and trends, addressing questions like what will happen. Prescriptive analytics went a
step further by recommending actions based on predictions, suggesting what should be done to
achieve desired outcomes.

Technologies during this period included data mining algorithms, statistical software such as SAS and
SPSS, and advanced Excel functions. These tools supported applications in marketing campaigns, risk
management, and operational efficiency by providing deeper insights and enabling better decision-
making.

2010-2020: Big Data

The 2010s were characterized by the emergence of Big Data, involving the management and analysis
of vast and complex datasets. Big Data introduced the concept of the four Vs: volume, velocity,
variety, and veracity.

Volume referred to the large amounts of data generated, velocity to the speed at which data was
produced and processed, variety to the diverse types of data (structured and unstructured), and
veracity to the quality and reliability of the data. Key technologies in this era included the Hadoop
ecosystem (with components like HDFS and MapReduce), NoSQL databases such as MongoDB, and
cloud computing platforms like AWS and Azure, which provided scalable solutions for data storage
and processing.

Big Data applications spanned social media analysis, Internet of Things (IoT), and real-time analytics,
allowing businesses to uncover hidden patterns and insights from massive datasets.

2020 Onwards: AI-Based Analytics

The 2020s have seen a shift towards AI-based analytics, leveraging artificial intelligence to automate
and enhance data-driven decision-making processes. This approach integrates machine learning, deep
learning, natural language processing (NLP), and reinforcement learning to provide more
sophisticated analytics solutions.

Ramaraju Poosapati Business Analytics For year 2024

Machine learning (ML) involves algorithms that learn from data and make predictions, while deep
learning uses advanced neural networks to identify complex data patterns. NLP focuses on
understanding and generating human language, and reinforcement learning enables decision-making
based on interactions with the environment. Technologies supporting AI-based analytics include AI
frameworks like TensorFlow and PyTorch, and cognitive computing systems.

AI-based analytics are applied in autonomous systems, personalized recommendations, and

intelligent customer service solutions such as chatbots, pushing the boundaries of what analytics can
achieve.

Modern Business Analytics

Modern business analytics represents the integration of Business Intelligence/Information Systems

(BI/IS), statistics, and modelling and optimization. While the core topics of modern business analytics
have been in use for decades, their uniqueness lies in their intersections.

Data mining focuses on understanding characteristics and patterns among variables in large
databases using a combination of statistical and analytical tools. It employs both standard statistical
tools and more advanced ones to extract useful information from data. Simulation and risk analysis
use spreadsheet models and statistical analysis to examine the impacts of uncertainty on various
estimates and their potential interactions, enabling businesses to explore different scenarios and
assess risks effectively.

Spreadsheets and formal models facilitate what-if analysis, allowing manipulation of data to perform
scenario testing. This type of analysis helps businesses understand how specific combinations of
inputs, reflecting key assumptions, will affect model outputs. Additionally, what-if analysis is used to
assess the sensitivity of optimization models to changes in data inputs, providing better insights for
making informed decisions.

Ramaraju Poosapati Business Analytics For year 2024

2. Scope of Business Analytics:
Business analytics encompasses a wide range of activities and methods designed to enhance
decision-making by leveraging data and statistical techniques. The scope of business analytics
can be broadly categorized into three primary types: descriptive, predictive, and prescriptive
analysis.

1. Descriptive Analysis
Definition: Descriptive analysis involves examining historical data to understand and describe
past events. It answers the question, “What has happened?” by summarizing data and
identifying patterns or trends. It uses techniques like Statistical Summaries, Data Aggregation
and Data Visualization.
Key Features:
• Data Summarization: Descriptive analysis summarizes large volumes of data into
understandable formats using measures like mean, median, mode, and standard
deviation.
• Trend Identification: It identifies patterns and trends over time, such as sales trends,
customer behavior patterns, or market performance.
• Visualization: Data visualization tools such as charts, graphs, and dashboards are
crucial in descriptive analysis for presenting data insights in a comprehensible
manner.

2. Predictive Analysis
Definition: Predictive analysis uses statistical models and algorithms to analyze historical data
and make predictions about future events. It answers the question, “What is likely to
happen?”. It uses techniques like Regression Analysis and Time Series Analysis.
Key Features:
• Trend Forecasting: Predictive analysis forecasts trends by analysing past data patterns
to anticipate future outcomes.
• Risk Assessment: It assesses potential risks and opportunities by predicting
probabilities of various outcomes.
• Scenario Planning: Helps in creating and analysing different future scenarios based on
predictive models.

3. Prescriptive Analysis
Definition: Prescriptive analysis provides recommendations for actions to achieve desired
outcomes. It answers the question, “What should we do?” by suggesting the best course of
action based on data insights and predictions. It uses techniques like Optimization models,
Simulation and Heuristic Algorithms.
Key Features:

• Optimization: Prescriptive analysis seeks to optimize outcomes by suggesting actions

that will yield the best results.
• Decision Support: It provides decision support by evaluating multiple scenarios and
recommending the most effective strategies.
• Real-Time Recommendations: In some applications, prescriptive analytics can offer
real-time recommendations based on current data.

Ramaraju Poosapati Business Analytics For year 2024

3. Models in Business Analytics:
A model is an abstraction or representation of a real system, idea, or object. Models capture
the most important features of a problem and present them in a form that is easy to
interpret.
Decision Model:
A decision model is a logical or mathematical representation of a problem or business
situation that can be used to understand, analyze, or facilitate making a decision. Most
decision models have three types of input:
1. Data, which are assumed to be constant for purposes of the model. Some examples
would be costs, machine capacities, and intercity distances.
2. Uncontrollable variables, which are quantities that can change but cannot be directly
controlled by the decision maker. Some examples would be customer demand,
inflation rates, and investment returns. Often, these variables are uncertain.
3. Decision variables, which are controllable and can be selected at the discretion of the
decision maker. Some examples would be production quantities, staffing levels, and
investment allocations.

Ramaraju Poosapati Business Analytics For year 2024

 4. Types of Data in Business Analytics
Data can also be classified by the type of measurement scale used:

Categorical (Nominal) Data

Categorical data is sorted into distinct categories based on specific characteristics. These
categories have no inherent order or quantitative relationship.
• Examples:
o Geographical Regions: North America, South America, Europe, Pacific.
o Employee Roles: Managers, supervisors, associates.
• Properties: Categorical data are often represented using numbers for ease of
management but do not imply any order. They are typically counted or expressed as
proportions or percentages.

Ordinal Data
Ordinal data can be ranked or ordered according to some relationship but does not provide
specific numerical differences between ranks.
• Examples:
o Rankings: College sports teams ranked based on performance.
o Survey Scales: Service ratings such as poor, average, good, very good,
excellent.
• Properties: Ordinal data allows for ranking and comparison but lacks fixed units of
measurement, making numerical differences between ranks less meaningful.

Interval Data /Discrete Data

Interval data has ordered categories with constant differences between observations, but
with arbitrary zero points, making ratios meaningless.
• Examples:
o Temperature: Measured in degrees Fahrenheit or Celsius.
o Test Scores: SAT or GMAT scores.
• Properties: Interval data allows meaningful comparison of differences but not ratios.
Time and temperature are common examples, where comparisons of differences (not
absolute values) are meaningful.

Ratio Data
Ratio data is continuous and has a natural zero, making both differences and ratios
meaningful.
• Examples:
o Dollars: Financial figures like sales revenue or expenses.
o Time: Duration of processes or events.
• Properties: Ratio data is the most informative type of measurement, allowing for
meaningful computation of averages, ranges, and ratios. It includes all the
information of the other data types and can be converted to them.

Ramaraju Poosapati Business Analytics For year 2024