UNIT V

Comprehensive Notes

1. Social Analytics

• Definition: The broader practice of measuring, analyzing, and interpreting the

results of interactions and associations between people, topics, and ideas. It
often encompasses both quantitative and qualitative data from social media
and other collaborative platforms.
• Focus: Understanding collective human behavior and the impact of social
interactions.

2. Social Network Analysis (SNA)

• Definition: The methodological toolkit used to study social structures through

the use of networks and graph theory. It maps and measures the relationships
and flows between connected entities.
• Key Components:

o Nodes (Vertices): The individual actors within the network (e.g., people,
organizations, websites).
o Ties (Edges): The relationships or connections between the nodes (e.g.,
friendship, communication, trade).
• Key Metrics:

o Centrality: Identifies the most important nodes in a network. A node with high
centrality is often an influencer, information broker, or bottleneck.
o Density: The proportion of actual ties in the network to the total number of
possible ties. Measures how well-connected the network is.
o Clusters/Communities: Groups of nodes that are more densely connected to
each other than to the rest of the network.
3. Social Media: Definitions and Concepts

• Social Media: Web-based platforms that allow the creation and exchange of
user-generated content. They facilitate social interaction online.
• Key Concepts:

o Engagement: Any public interaction with content (Likes, Shares, Comments,

Retweets).
o Reach: The total number of unique users who saw a piece of content.
o Impressions: The total number of times a piece of content was displayed (a
user can see multiple impressions).
o Influencer: A user with a large and engaged following who can shape
audience opinions.

4. Social Media Analytics

• Definition: The process of collecting and analyzing data from social media
platforms to guide business decisions.
• It is the application of Social Analytics and SNA specifically to social
media data.
• Goals: Measure marketing campaign performance, understand brand
sentiment, identify influencers, manage customer service, and track
competitors.
• Examples: Tracking hashtag performance, analyzing sentiment in comments,
identifying key influencers in a topic network.

5. Prescriptive Analytics

• Definition: The final and most advanced stage of analytics, which goes
beyond describing what has happened (Descriptive) or predicting what will
happen (Predictive) to recommend what action to take to make it happen.
• Purpose: To advise on possible outcomes and recommend decisions.
• Uses: Optimization models, simulation, decision analysis.
• Example: Based on predictive models of customer churn, a prescriptive system
might recommend offering a specific discount to a high-value customer to
prevent them from leaving.
6. Optimization and Multi-Criteria Systems

• Optimization: The process of finding the best solution from all feasible
solutions, typically by maximizing or minimizing an objective function (e.g.,
maximize profit, minimize cost).
• Multi-Criteria Decision Making (MCDM): Used when decisions need to be
made with multiple, often conflicting, goals. You cannot optimize for one goal
without sacrificing another.
• Example: Choosing a new office location involves multiple criteria: cost
(minimize), size (maximize), and commute time for employees (minimize).

7. Multiple Goals

• In real-world problems, organizations have multiple goals (e.g., increase

market share, improve customer satisfaction, reduce operational costs).
• Challenge: These goals often conflict. Optimization becomes a balancing act,
requiring trade-offs.
• Solution: Techniques like weighted goal programming are used, where
decision-makers assign importance weights to different goals to find a
satisfactory compromise.

8. Sensitivity Analysis

• Definition: A technique used to determine how different values of an

independent variable impact a particular dependent variable under a given set
of assumptions.
• Purpose: It tests the robustness of an analysis or model. It answers the
question: "If we change an input, how much does the output change?"
• Also known as: "What-if" analysis.
• Example: In a financial model, testing how the Net Present Value (NPV) of a
project changes if the initial cost increases by 10%.

9. What-If Analysis

• Definition: The process of changing the values in cells (inputs) to see how
those changes will affect the outcome of formulas on the worksheet (outputs).
 • It is the practical application of sensitivity analysis.
• Tools: Excel's Data Tables, Scenario Manager, and Goal Seek are built-in tools
for this.

10. Goal Seeking

• Definition: A specific type of what-if analysis that works backwards from the
desired outcome. It calculates the input value needed to achieve a specific goal
for the output.
• Process: The user sets the target value for the formula, and the tool
determines the input value required.
• Example: "What annual sales growth rate do we need to achieve to reach $1
million in revenue next year?" Goal Seek would calculate the required growth
rate.

Assessment Questions & Answers

1. Define Social Network Analysis (SNA). Explain the concepts of

"Nodes" and "Ties" and describe two key metrics used in SNA.
Definition of Social Network Analysis (SNA):
Social Network Analysis (SNA) is the methodological process of investigating social
structures through the use of networks and graph theory. It maps and measures the
relationships and flows between people, groups, organizations, computers, URLs, and
other connected information/knowledge entities. The core premise is that the
patterns of connections (the network structure) are critical for understanding the
behavior of the individuals within the network and the network as a whole.

Core Concepts:

• Nodes (or Actors/Vertices): These are the fundamental units or entities

within a network. They represent the individual players whose connections are
being analyzed.
o Examples: People in a friendship circle, employees in a company,
computers on the internet, websites on the WWW, or airports in a flight
network.
 • Ties (or Edges/Links/Relationships): These are the connections that link the
nodes together. Ties represent the type and strength of the relationship
between nodes.
o Examples: Friendship, professional collaboration, email
communication, financial transactions, hyperlinks, or physical roads.
Ties can be directed (e.g., A follows B, but B does not follow A)
or undirected (e.g., A is co-authors with B).

Two Key SNA Metrics:

1. Centrality Measures (e.g., Degree Centrality & Betweenness Centrality):

o Degree Centrality: This is a simple count of the number of ties a node
has. In a social network, a person with a high degree centrality is a
"connector" or "hub" with many direct contacts. They have access to
and can influence a large part of the network directly.
o Betweenness Centrality: This measures the extent to which a node lies
on the shortest path between other nodes. A node with high
betweenness centrality acts as a "bridge" or "broker" between different
parts of the network. This position gives the node the power to control
the flow of information and is often a source of influence.
2. Density:
o Description: Density measures how connected a network is as a whole.
It is calculated as the ratio of the actual number of ties in the network
to the total number of possible ties. A density of 1 means every node is
connected to every other node, while a density of 0 means no
connections exist.
o Significance: A high-density network implies strong cohesion, rapid
information diffusion, and potentially strong social capital. A low-
density network suggests fragmented groups and may require key
brokers (high betweenness nodes) to connect different clusters.

2. Compare and contrast Social Media Analytics with

Prescriptive Analytics. How do they differ in their purpose and
the questions they answer?
Feature Social Media Analytics Prescriptive Analytics

To listen, understand, and

To recommend actions and
describe customer sentiment,
Primary show the implications of each
brand perception, and emerging
Purpose decision to achieve the best
trends based on social media
possible outcome.
data.

Stage in Descriptive and Diagnostic

The most advanced stage
Analytics Analytics ("What happened?"
("What should we do?")
Maturity and "Why did it happen?")

• What are people saying about • What is the optimal price for
our brand? our product?
Core • How is our campaign • What is the best way to
Questions it performing? allocate our marketing budget?
Answers • Who are the key influencers? • Which customer segment
• What is the public sentiment should we target to maximize
around our new product? ROI?

Insights, Reports, and Actionable Recommendations

Nature of Dashboards. (e.g., sentiment and Decision Options. (e.g.,
Output charts, share of voice, "Set price at $49.99," "Allocate
engagement metrics). 60% of budget to Channel A").

Optimization, Simulation,
Underlying NLP, Sentiment Analysis, Topic
Decision Analysis, Machine
Methods Modeling, Network Analysis.
Learning.

Temporal Primarily focused on the past

Primarily focused on the future.
Focus and present.

Contrast: Social Media Analytics is like a doctor using a thermometer and patient
interview to diagnose an illness (what and why). Prescriptive Analytics is the same
doctor using that diagnosis to prescribe a specific drug and dosage to cure it (what
should be done).
3. A company wants to launch a new product. Describe how
they could use Social Media Analytics for market research and
how they could use Prescriptive Analytics for setting the price.
A. Using Social Media Analytics for Market Research:

The company can use social media as a vast, real-time focus group to understand the
market before the launch.

1. Identify Target Audience & Influencers: Analyze social conversations to

identify which demographics are most actively discussing the product
category. Use SNA to find key opinion leaders and influencers whose
endorsements would be most valuable.
2. Analyze Consumer Needs & Gaps: Use topic modeling and text analysis on
social media posts and reviews of competing products. This reveals unmet
customer needs, common pain points (e.g., "battery life is too short"), and
desired features.
3. Gauge Brand & Concept Sentiment: Launch a "soft" marketing campaign or
teasers for the new product. Use sentiment analysis to measure the initial
public reaction. Is the sentiment positive, negative, or neutral? This helps
refine messaging and positioning before the full launch.

B. Using Prescriptive Analytics for Setting the Price:

After gathering market data, the company must determine the optimal price.

1. Build a Predictive Model: The model would incorporate data from social
media (e.g., sentiment score, volume of conversation), along with internal data
(production costs, desired profit margin) and external data (competitor
pricing, economic indicators).
2. Define Constraints and Objective: The objective is to maximize profit or
market share. Constraints could include: price must be above cost, must be
within 10% of the closest competitor, and must not negatively impact brand
perception.
3. Run Optimization Scenarios: The prescriptive analytics system runs
simulations and optimization algorithms to evaluate thousands of potential
price points. It doesn't just predict sales at a given price; it prescribes
the best price by considering all constraints and the overarching business
objective.
4. Output: The output is a clear, data-driven recommendation, such as: "To
maximize first-year profit while achieving a 15% market share, the optimal
launch price is $129.99."
4. Explain why Multi-Criteria Decision Making (MCDM) is
necessary in business. Provide a business example where
optimizing for a single goal would be insufficient or harmful.
Why MCDM is Necessary:
Business decisions are almost never one-dimensional. Optimizing for a single goal
(e.g., maximizing profit) in a complex environment often leads to unintended
negative consequences in other critical areas. MCDM is necessary because it provides
a structured framework for evaluating alternatives based on multiple, often
conflicting, criteria. It acknowledges that real-world success requires a balance
between various objectives such as cost, quality, risk, and customer satisfaction.

Business Example: Selecting a New Supplier

• Single-Goal Approach (Minimize Cost): A company decides to choose a

supplier based solely on the lowest price per unit.
• Negative Consequences:
o Quality: The cheapest supplier may have poor quality control, leading
to defective products and increased returns.
o Reliability: They may have unreliable delivery times, causing
production stoppages and missed customer deadlines.
o Risk: They might use unethical labor practices, creating reputational
damage and potential legal issues.
o Sustainability: They may have a large carbon footprint, conflicting with
the company's environmental commitments.
• MCDM Approach: The company uses a method like the Weighted Scoring
Model. They define multiple criteria:
o Cost (Weight: 30%)
o Quality & Defect Rate (Weight: 25%)
o On-Time Delivery (Weight: 20%)
o Ethical Compliance (Weight: 15%)
o Environmental Impact (Weight: 10%)
Each potential supplier is scored on each criterion, and the scores are
weighted and summed. The supplier with the highest total score is
selected, ensuring a balanced decision that supports all of the
company's strategic goals, not just cost minimization.
5. Describe the purpose of Sensitivity Analysis. Provide a
concrete example of how a manager might use it to evaluate a
proposed project.
Purpose of Sensitivity Analysis:
Sensitivity Analysis (also called "What-If Analysis") is a technique used to determine
how different values of an independent variable (input) impact a particular
dependent variable (output) under a given set of assumptions. Its purpose is to:

• Identify "Key Drivers": Pinpoint which inputs have the most significant
impact on the outcome.
• Assess Risk and Uncertainty: Understand how vulnerable a project or model
is to changes in its underlying assumptions.
• Support Robust Decision-Making: Test the strength of a decision by seeing
if the preferred alternative remains the best under different scenarios.

Concrete Example: Evaluating a New Product Launch Project

A manager is evaluating a project to launch a new product. The financial model

predicts a Net Present Value (NPV) of $2 million, based on these key assumptions:

• Selling Price: $100

• Sales Volume: 50,000 units/year
• Variable Cost: $60 per unit

The manager performs a Sensitivity Analysis on the NPV by varying each key
assumption:

1. Vary Sales Volume: What if sales volume is only 40,000 units? Or 60,000? The
analysis shows that a 10% drop in volume causes a 25% drop in NPV, making
volume a highly sensitive variable.
2. Vary Variable Cost: What if the cost of raw materials increases, pushing the
variable cost to $65? The analysis might show that a $5 increase reduces the
NPV by 40%, indicating the model is very sensitive to cost.
3. Vary Selling Price: What if competitive pressure forces a price of $95? The
analysis shows a relatively smaller impact on NPV.

Managerial Insight: The manager now knows that the project's success is highly
dependent on achieving the sales volume target and controlling costs. The price is less
critical. Therefore, the manager should:

• Invest more in market research to validate the volume forecast.

• Secure long-term supplier contracts to lock in the $60 cost.
• Develop contingency plans in case volume or cost assumptions prove wrong.
6. Explain the difference between "What-If Analysis" and "Goal
Seeking." Provide a unique example for each.
What-If Analysis and Goal Seeking are both types of sensitivity analysis, but they
work in opposite directions.

Feature What-If Analysis Goal Seeking

Forward-Looking: You change Backward-Looking: You set a

Direction of
the inputs to see the effect on desired output to find the
Analysis
the output. required input.

Core "What would be the result if I "What input value is needed

Question change this variable?" to achieve my goal?"

Input (X) → [Model] → Output Desired Output (Y) → [Model] →

Process
(Y) Required Input (X)

Unique Examples:

• What-If Analysis Example: A Marketing Budget Scenario

o Scenario: A CMO has a total budget of $1 million to allocate between
TV ads and Digital ads. The current plan is $600k for TV and $400k for
Digital, forecasted to generate 20,000 new leads.
o What-If Question: "What if I shift $100k from TV to Digital? How
many leads would I get then?"
o Process: The CMO changes the input (TV=$500k, Digital=$500k) in the
marketing model. The model recalculates and outputs a new forecast:
22,000 leads. This helps the CMO evaluate different allocation
strategies.
• Goal Seeking Example: A Personal Savings Goal
o Scenario: An individual wants to save $50,000 for a down payment on
a house in 5 years. They have $10,000 in initial savings and expect an
annual return of 5% from their investment account.
o Goal-Seeking Question: "To reach my goal of $50,000 in 5 years, how
much do I need to save each month?"
o Process: The person sets the desired output (Future Value = $50,000)
in their financial calculator or spreadsheet. The goal-seeking function
then works backward to solve for the unknown input, which is the
 monthly contribution. The result might be: "You need to save $575 per
month."

Section B: Multiple Choice Questions (1 Mark Each)

1. In Social Network Analysis (SNA), a metric that identifies the most

influential or well-connected individual within a network is known as:
a) Density
b) Centrality
c) Sentiment
d) Engagement
2. The type of analytics that recommends what action should be taken to
influence a future outcome is:
a) Descriptive Analytics
b) Diagnostic Analytics
c) Predictive Analytics
d) Prescriptive Analytics
3. Which of the following is the BEST example of a "What-If" analysis
question?
a) What was our total revenue last quarter?
b) What will our market share be in six months?
c) How would a 5% increase in material costs affect our net profit?
d) What sales figure do we need to break even?
4. A marketing team is analyzing a campaign's performance by looking at
Likes, Shares, and Comments. They are primarily engaged in:
a) Social Network Analysis
b) Sensitivity Analysis
c) Social Media Analytics
d) Goal Seeking
5. The analytical process of working backward from a desired output to
determine the necessary input is called:
a) Optimization
b) Sensitivity Analysis
 c) Goal Seeking
d) Multi-Criteria Analysis
6. When a business decision involves balancing conflicting objectives like
minimizing cost and maximizing quality, it requires:
a) A single objective optimization model.
b) Social media listening.
c) Multi-Criteria Decision Making (MCDM).
d) Descriptive analytics.

Section C: Fill in the Blanks (1 Mark Each)

1. In a social network graph, the individual actors, such as people or companies,

are represented as _______________.
2. The stage of analytics that uses data to forecast what is likely to happen in the
future is called _______________ analytics.
3. _______________ is a technique used to test how sensitive the output of a model
is to changes in its input assumptions.
4. The total number of unique users who saw a social media post is known as its
_______________.
5. Finding the best possible solution to a problem, given a set of constraints, is
the process of _______________.
6. If you use a tool to calculate the required interest rate to double your
investment in 5 years, you are performing _______________.

Answer Key

Section B: Multiple Choice Answers

1. b) Centrality
2. d) Prescriptive Analytics
3. c) How would a 5% increase in material costs affect our net profit?
4. c) Social Media Analytics
 5. c) Goal Seeking
6. c) Multi-Criteria Decision Making (MCDM).

Section C: Fill in the Blanks Answers

1. Nodes (or Vertices)

2. Predictive
3. Sensitivity Analysis
4. Reach
5. Optimization
6. Goal Seeking