Scope Analytics
Scope Analytics
UNIT-I: INTRODUCTION
Analytics
1. Introduction to Data
.1 Meaning andScope ofData Analytics
1.1.1 Data and Data Types
1.1.2 Meaning of Data Analyticsand Data Science
1.1.3 Data and Data Analytics
l.14 Business Data Analyticsand Business Intelligence
1.1.5 Scope ofBusiness Data Analytics (BDA)
1.2 Classification of DataAnalytics
1.2.1 Descriptive Analytics
1.2.2 Diagnostic Analytics
1.2.3 Predictive Analytics
1.2.4 Prescriptive Analytics
1.2.5 Cognitive Analytics
13 Application of DataAnalytics in Business
1.4 Big Data and its Characteristics
1.4.1 Meaning ofBig Data
14.2 Big Data Characteristics
1.4.3 Benefits of Big Data
1.5 Application ofBig Data
1.6 Challenges in Business Data Analytics
Analytics
1.6.1 Overcoming the Challenges in Business Data
Exercises
SUMMARIZATION AND
UNIT-II: DATA PREPARATION,
VISUALIZATIONS USING SPREADSHEET
1.1
Introduction to Data Analytics
Meaning and Scope of Data Analytics
1.1.1 Data and Data Types
1.1.2 Meaning of Data Scicnce and Analytics
1.1.3 Data and Data Analytics
1.1.4 Data Analytics Vs. Data Analysis
1.1.5 Scope of Data Analytics
1.2 Classification of Analytics
1.2.1 Descriptive Analytics
1.2.2 Diagnostic Analytics
1.2.3 Predictive Analytics
1.2.4 Prescriptive Analytics
1.2.5 Cognitive Analytics
1.3 Applicationof analytics in business
1.4 Big Data and its characteristics
1.5 Applications of Big data
1.6 Challenges in data analytics
Exercises
Answers
Dietionary (or Map): an associative data structure that stores collections of key-yalue
pairs; cach key is linked to a single value (e.g. "get(firstname) Anupreet; get(lastname)
Kaur').
Data may be categorised, analyscd and interpreted depending upon its level of measurement, that
deteumiesthe nature of mathematical operations that can be performed on the data. The main
measurement include:
Jvels of data
(a) Nominal: The simplest form of categorical data that does not ascribe any order or ranking to
the data items. Accordingly, nominal data is essentially only a label to accord a distinctive
meaning to the data item without assigning any quantitative value to it. For example, month
January. February, December); gender (male, female, others); movie genre (action, comedy,
drama)etc.
k Ordinal: The next level of categorical data that accords ranks or order to the data items.
However, the intervals between successive ranks may not be of the same magnitude. For
Pxample. in a customer satisfaction survey, the ratings assigned may be 'highly satisfied',
*somewhat satisfied', 'neutral', 'somewhat dissatisfied and 'highly dissatisfied'. While it is
obvious that a rating of 'highly satisfied' ranks higher than other ratings, it is difficult to
auantify the difference in satisfaction levels between successive ratings.
e Scale: The level of categorical data that primarily contains numerical values that accord
meaningful ranks or order to the data items. To state differently, the intervals between successive
values are of equal magnitude. Scala data, therefore, is amenable to statistical analyses and
meaningful interpretations for decision-making. Scala data may be further classified into ratio
data and interval data. Ratio data has atrue zero indicating the absence of avalue. For example,
a zero value assigned to age, weight, speed represents complete absence of age, weight and
speed respectively. Interval data, on the other hand, does not have a true zero indicative of a
complete absence of avalue; the zero point is arbitrary in nature. For example, atemperature
of zero degrees Celsius or Fahrenheit does not signify the absence of heat. Similarly, the year
absence
zero is an arbitrary starting point to measure passage of time and does not indicate the
of time.
affects the
Athorough understanding of data types and levels of measurement is essential since it
way in which data is collected, organised, analysed and interpreted to derive meaningful conclusions
and formulate reliable solutions to real-time problems.
Ouality Control: BDA can be used to improve quality control by identifying trends in e nane
product defects. The manufacBurers, by comparing the actual product data with its Aea wase
specifications,can take corrective action before defects become amajor problem.
(d) Demand Forecasting: The predictive analytics of BDA can be used to forecast demand
for products. As a result, manufacturers can plan their production volumes accordingly
to avoid stockouts or overproduction.
(e) Process Improvement: BDA can be used to identify areas where manufacturing processes
can be improved by identifying bottlenecks and inefficiencies, measuring progress,
improved decision makirg and continuous performance monitoring. Some process
i provement methodologies duly supported by BDA are Six Sigma, which is a data
driven methodology that focuses on eliminating defects in processes; Lean Six Sigma is
a combination of lean manufacturing principles and Six Sigma methodology that focuses
on reducing waste and improving efficiency; and Kaizen, which is a Japenese philosophy
that focuses on continuous improvement. In essence, this can help to reduce costs and
improve product quality.
2 Retail Sector: BDA plays a critical role in the retail sector, byempowering retailers with
valuable insights to make informed decisions and optimize operations by leveraging data.
Some of the main applications of BDA in retail have been explained below:
(a) Customer Segmentation and Targeting: BDA helps the retailers to analyze customer
purchase history, demographics, and loyalty program data, so as to segment their customer
base into distinct groupswith similar characteristics and buying behaviours. This enables
them to target marketing campaigns more effectively, personalize product recommendations,.
and develop targeted promotions that resonate with specific customer segments.
(b) Inventory Management: BDA helps retailers to optimize inventory levels by analyzing
sáles trends, seasonal fluctuations, and supplier lead times. As a result, they can prevent
stockouts, rpduce storage costs, and ensure that the right products are available at the
right time, Predictiveanalytics can be used to forecast future demand and adjust inventory
levels accordingly.
(c) Pricing Strategy: Retailers may leverage BDA to decide optimal prices for their products.
After considering such factors as competitor pricing, product cost, customer demand,
and price elasticity. the retailers can identify the price point that maximizes their
profitability.
(d) Store Operations: BDA can be used to analyze data from point-of-sale systems, foot
fall on counters, and video surveillanceto gain insights into store layout, staffinglevels,
and checkout efficiency. This information may be used to improve store layout, optimize
staffing schedules, and reduce wait times.
(e) Fraud Detection:Retailers may use BDA to detect fraudulent transactions by analyzing
purchase patterns and identifying inconsistenciesand anomalies. Járgepurchases fiequent
returns from the same customer, yasual purchase locations compared to billing address,
or a sudden spike in transactions from a new account can all be red flags as likely
transaction frauds. Unexplained discrepancies between inventory records and actual stock
counts can indicate internaltheft of inventory. Similarly, Analysis of Point-of-sale (POS)
from cash registers can revealunusual activity patterns:such as returnsprocessed outside
Kauen
1.8
BUSINESS ANALYTIC:
of usualreturn windows or by unauthorized personnel. This may help to prevent finanei
losses and protcct customers from identity theft.
() Supply Chain Management: BDAcan be used to improve the efficiency of tthe
chain by tracking inventory levels throughout the supply chain, thereby suppl
identifyin
bottlenccks, and optimizing transportation routes.
(g) Site Selection: Whilc considering opening a new store, retailers can use BDA
to identif
the best locations, The retailers, by analyzing demographic data, traffic patterns, and
3.
presence of competitor can choose locations that are most likely to be successful.
Communication Sector: Business data analytics has become a powerful tool for improving
COmmunication in several ways. Some of the key applications are explained below:
(a) Understanding Customer Behaviour and
Preferences: Business
analyzing data from various communication channels (email, social media,enterprises,
website
by
chat)
can casily understand the way customers prefer to be
content that cchoes well with them. This paves the way for
contacted and also the kind of
strategies that are more likely to get a positive response from targeted communication
(b) Optimizing Communication customers.
Channels: BDA provides a good help in assessing the
effectivencss of different communication channels: email marketing, secial media
campaigns, or çustomer service interactions etc. Business
channels that generate the best results and enterprises can identify the
cfforts. accordingly optinize their communication
(c) Improving Messaging and its Content: The analysis of
communication tends to reveal the kind of content that performscustomer
can accordingly use this
engagement with
well. Business enterprises
information to tailor
communication channels. As a result, it leads totheir
more
messaging for different audiences and
effective communication that drives
conversionsor brand loyalty.
(d) Identifying Communication
which communication Gaps: BDA can be used to identify
and highlight areas in
in customer service, formight be missing or at a rudimentary stage. A slow
may be example, may be revealed
or
low open rates for emailresponse time
identified. Business enterprises can use this campaigns
communicationgaps to bring about an information to identify and
address
(e) Measuring ROIof improvement in overall communicationefficiency.
track key metrics like Communication
Efforts: Business enterprises, using BDA can
by
can help to measure thewebsite traffic, conversion rates, and
data-driven approach allowsreturn on investment (RO)of customer satisfaction. This
their
business enterprises to
strategics that are most efective understand communication
the best
efforts. This
4.
Financial Sectors: BDA plays a criticalin role
achieving their goals. communication
banking and insurance. in main sectors of
the finance
The financial world
is a industry, such as
highlyvolatile
help the
organizations to mancuver their market place, and
BDA is used to extract
turn to business analysts to optimize way through tricky terrains. Financial insights that
portfolio
include: management. SOme of the key
budgeting, banking, financial planning, enterprises
sector
functionalities forecasting, and
provided BDA in the finance sector
by
INTRODUCTIONTO DATA ANALYTICS
1.9
Analytics, by comparing these ratios with industry averages or its own historical performance.
highlights the strengths and weaknesses of the business enterprise. Fourth, descriptive analytics
belps investors to conduct Investment Analysis to evaluate potential investments. This is
achicved by analysing financial statements of the business enterprise, by cxamining its stock
price history, and also considering industry trends. Further, descriptive analytics, by organizing
and summarizing the financial data, helps investors make informed decisions about where to
park their funds. In essence, descriptive analytics is a foundational structure of financial analysis
for a business enterpriseto provide the what' and 'how' behind its financial situation, thereby
enabling financial professionals to make crucial informed decisions.
Operations and Production: Descriptive analytics is a powerful tool used in operational and
productionactivity to understand past performance and identify areas for improvement. First,
it is used for performance monitoring by summarising a lot of data, which is
generated by
the operational activity and production lines. Descriptive analytics helps track key KPls such
as production volume, defect rates, and machine downtime. This enables the managers to
monitor overall performance, identify bottlenecks and the arcas, wherein the capacity stands
exceeded. Sccond, descriptive analytics, by analyzing historical data and their patterns, helps
Identify Trends. For instance, such analytics can reveal a seasonal dip in production or a
gradual increase in defect rates over a specific product line. Third, descriptive analytics performs
the task of Benchmarking, by allowing manufacturing enterprises to compare their
performance against industry standards or their own historical performance over a period of
time. This helps such enterprises to identify arcas where they are lagging behind and have the
opportuniies for improvement. Fourth, this type of analytics can reveal products and services
that are resource-intensive or otherwise, and thereby helps in Resource Allocation through a
better allocation of manpower, materials, and machinery to optimize production efficiency. In
essence, descriptive analytics provides a clear view of actual activities in operations and
production so that adata-driven approach may be adopted as the foundation for making informed
decisions and implementing improvenents in various aspects of manufacturing.
Human Resource Management: Descriptive analytics is a powerful tool in the hands of HR
professionals to understand their workforce requirement by summarizing and analyzing
historical data pertaining to manpower employed. First, The HR team, can identify trends in
historical data pertaining to labour-turnover, absenteeism in specific department, and time to
hire. Further. such trends may be investigated to understand the root causes of absenteeism
and higher labour turnover. Second, descriptive analytics allows Benchmarking by HR
professionals to compare their business enterprise's HR metrics with industry standards or
against their own performance from previous periods. In essence, this helps to assess how
well their HR practices are working in the organization. Third, descriptive analytics can provide
workforce insights with respect to demographics, skills, and experience of the workforce.
This aspect helps in workforce planning, identifying gaps in skill, and also designing targeted
training programs. Fourth, the descriptive analytics may also be used by HR professionals for
reporting and communication by generating reports that summarize key HR metrics for
management. This data-driven approach helps in informed strategic decision making. In essence,
descriptive analytics provides the structural foundation for more advanced HR analytics such
as predictive and prescriptive analytics, which may help HR team to anticipate future trends
and develop strategies to optimize workforce performance.
BUSINESS AHALITIC
1.16
Descriptive
Benefits of business dataAnalytics:
Descriptive analytics provides a clear understanding ofthe busincss hcalth, hrli
goals and identifies areas andissues where deeper
(predictive achievement of
or prescriptive) mightbe useful. Specifically, it offers several benefits to organ17ati
monitor progresstowards
(d) Explore the Data: The data visualization techniques may be used to explore
data so that the
patterns and trends are conveniently and readily identified.
(c Formulation of Hypothesis: An educated guesswork about what might be causing the
anomaly
or a problem may be used for this purpose.
() Testing of Hypothesis: Statistical analysis techniques are used to test the relationships between
the variables that have been identified.
(o) Draw Inferences: On the basis of above analysis, the business data analyst determines the
root cause of the anomaly (or the problem in hand).
(h) Appropriate Action: Once the analyst knows why something undesirable is happening, he
may initiate appropriate steps to fix the problem (or improve the undesirable situation).
Techniques of Diagnostic Analytics: Some of the common techniques used in diagnostic business
data analytics are as given below:
(a) Data Mining: Data mining is a process to extract hidden patterns and trends from a large
dataset, and thereafter explore the relationships within such datasets.
(b) Data Drilling Down: This technique is the process of zooming in on specific data points to
dig deeper into specific data points to understand the underlying causes and also to get a more
detailed view.
(c) Statistical Analysis: Statistical analysis is the use of statistical methods to summarize, analyse,
and interpret data. This technique of statistical analysis is also used to test hypotheses and
identify correlations between variables.
(d) Regression Analysis: This technique is used to model the relationship between a dependent
variable and one or more independent variables.
Benefits of Diagnostic Analytics: Business data analysts may leverage diagnostic analytics to
identify and solve problematic issues, improve efficiency, and make better data-driven decisions.
Some key benefits of using diagnostic business data analytics are as follows:
(a) Identify and Resolve Problems: Diagnostic analytics is a powerful tool for identifying and
resolving problems with targeted solutions by using data to pinpoint the root causes and not
just solutions, and also gives the analysts a general idea of an issue. Insights provided by
diagnostic analysis can help prevent similar problems in the future.
(b) Improved Decision-making: By understanding the root causes of problems and trends,
businesses can make more informed decisions about allocation of resources, improvement in
processes, and also the achievement of enterprises' goals. Decision making is based on evidence,
and not a mere guesswork.
(c) Increased Productivity: Diagnostic analytics can help businesses identify areas where they
are wasting time or money. By eliminating these inefficiencies and unproductive expenses, a
business analyst can improve their overall productivity.
(d) Reduced Costs: Diagnostic analytics, by identifying and solving problems at an early stage,
Can help save money on corrective actions and thereby reduce the costs.
(e) Develop Better Products and Services: Diagnostic analytics helps the analyst to
pinpoint
the root causes of customer behavior, product performance, or operational issues. Understanding
the products and services on a deeper level is the holy grail of business enterprises. Further,
1.18
BUSINESS ANALYTIc
Diagnostic analytics can helpto identify arcas wherethe product or service is falling Short an
and services.
thereby allows for better products Diagnostic analytics can help
Satisfaction and Loyalty: business
() Boost Customer
to understand why customcrs
relatcd to
are satisfied or
enterpr
dissatisfied with the product or services, D
dissatisfaction, the business analysts can imnprove
addressing the issues
satisfactionand develop customer loyalty. wide range of
CUstome
Business Data Analytics: It is applied to a
Application of Diagnostic
situations. For example, a busincss enterprise might use diagnostic analytics in a number of way
declining. In this context, it is imy
busines
understand why customer satisfaction scores are
First,
identify theroot cause ofmanufacturing defects, Analyse the effectiveness of marketing campaign
Second, business enterprises. hy.
and diagnose the reasons for high employee turnover.
diagnostic analytics, can gain a deeper understanding of their operations and make data-drive
aleai
decisions that can improve their bottom line. Third, Diagnostic business data analytics may
prOcesses, such as:
used in busincss functions to improve a wide range of business
Marketing: Diagnose why certain marketing campaigns are not performing well
Sales: ldentify the factors that are influencing sales conversion rates
Customer Service: Understand why customers are churning and take steps to improve custome
retention
Operations: Identify bottlenecks in production processes and improve efficiency.
1.2.3 Predictive Analytics
Predictive business data analytics is a powerful tool that uses historical data, statistical modelling
and machine learning techniques to forecast future trends, patterns and events. Such analytis
essentially allows business enterprises to forecast with reasonable accuracy as to what might happ
in near future. As a result, business enterprises may develop their strategies by taking data-driv
decisions to improve operations, reduce risk, and gain a competitive advantage.
Objectives of Predictive Analytics: The primary goal of predictive analytics is to enable th
forecasting of expected outcomes in the near future, on the basis of historical data. The predicti
analytics, by unveiling the future, goes beyond just understanding what has happened in the pas
by aiming to provide insights into what is likely occur. Predictive analytics uses statistical algorithm
and machine learning techniques to sift through data, identify patterns and
uncover hiddc
patterns that may not be readily apparent through simpler analytical methods.
Techniques of Predictive Analytics: Predictive analytics is a branch of data
statistical techniques, extrapolation, data mining, machine learning and artificial analytics that çs
make predictionsabout future outcomes. The process of predictive Analytics to explore intelligence
and analys
historical data to identify patterns and relationships that can be used for
common techniques that are used in predictive forecasting. Some o
analytics have been explained
(a) Regression Analysis: This is a statistical technique that helps to identify thebelow:
causal relationshi
between a dependent variable (the variable sought to be predicted) and one or more independen
variables (the variables that
degree to which the value ofinfluence
the dependent
variable variable).
registers Ita change understanding thg
the dependent helps in when one or mot-
INTRODUCTION TO DATA ANALYTICS 1.19
independent variables arc varicd. There are different types of regression analysis, including
lincar reyression, multiple regression, logistic regression and polynomial regression, cach suitcd
to differcnt types of data and forccasting. Regression analysis is widely used in various business
functions toidentify patterns,understand the impact of different variables on an outcome and
also to makc predictions. It hclps in making informed decisions, by facilitating forecasting
future trends, besides testing hypotheses based on data.
(h) Decision Irees: Decision trec analysis is avisual tool that helps in making stratcgic decisions.
It involves crcating alowchart-like structure with nodes and branches to represent different
decisionpoints and their potential outcomes. The root node is the starting point, internal nodes
represent decision points, branches represent possible choices, and leaf nodes represent the
final outcomes. By analyzing the decision tree, organizations can identify the most effective
COursc of action.
(c) Neural Networks: Ancural nctwork is amethod in artificial intelligence that guides computers
to proccss data in such a way as human brain does. It is a typeof machine learning process,
callcd deep learning that uses interconnccted nodes or artificial neurons ina layered structure
that rescmbles the human brain, thercby creating an adaptive system. The computers learn
from their mistakes and improve continuously through trial-and-error process. As a result,
artificial ncural networks make repcated attempts to solve complicated problems, like
summarizing documents or recognizing faces, with greater accuracy. Therefore, artificial
neuralnetworks arca type of machine learning algorithm that is inspired by the structure and
functioning of the human brain. Neural networks are used to model complex relationships
betwcen inputs and outputs and can be used for a variety of predictive analytics tasks, such as
fraud detection, image recognition, and natural language processing.
(d) Time Series Analysis: Time series analysis is a specific way of analyzing a sequence of data
points collected over an interval of time consistently. In time series analysis, it is imperative to
recorddata points at uniform intervals over a set period of time rather than just recording the
data points intermittently or randomly. This statistical technique is used to identify trends,
scasonality, andothcr patterns in data. This information may then be used to forecast expected
values of the tíme series.
(e) Survival Analysis: Originally developed in biomedical sciences, survival analysis has been
uscd to understand the onset of certain diseases. Now survival analysis is a branch of statistics
that may bcused to forecast how long it takes for certain instances to occur. It is widely used
in engincering, finance, insurance, risk analysis, and many other disciplines. This statistical
technique is used to analyse and understand insights of data on the time to an event, such as the
time todeath, the time to a customer churn, or the time to a machine failure. Survival analysis
can bc usedto estimate the probability that an event may occur within a certain time period.
These are some of the many techniques that may be used in predictive analytics.The best technique
to be applied will depend on the specific problem that the analyst is trying to solve.
Benefits of Predictive Analytics: The predictive analytics offers the following benefits to the
business enterprises to boost their operational capabilities and business functions:
(a) Identify Future Trends and Opportunities in Real-Time: Advancements in technology and
data processing capability have paved the way for real-time predictive analytics, thereby
allowing for instant predictions, development of strategy and immediate course correction so
as to give business enterprises a significant edge over others.