2. Table of Contents
Introduction............................................................................................................................................................. 2
What is Business Analytics?............................................................................................................................... 2
Why business Analytics is important?................................................................................................................ 3
Role of Analytics in business.................................................................................................................................. 5
Business Analytics - Industry focus areas .............................................................................................................. 7
Business Analytics Reference Architecture............................................................................................................ 8
Data Layer........................................................................................................................................................... 9
Why data layer is so important?...................................................................................................................... 9
Data Layer Systems ...................................................................................................................................... 10
Structured Data ............................................................................................................................................. 10
Unstructured Data ......................................................................................................................................... 10
Analytics Access Layer..................................................................................................................................... 11
Historical Analysis........................................................................................................................................ 12
Current Analysis ........................................................................................................................................... 12
Future Analysis (Analytics) .......................................................................................................................... 12
Performance Analytics Applications (Current and Future) .............................................................................. 13
Delivery Medium.............................................................................................................................................. 14
Security ............................................................................................................................................................. 14
High Level Analytic Capabilities – Industry Agnostic......................................................................................... 15
How do we Implement Business Analytics Solutions? ........................................................................................ 16
Why Cognizant?.................................................................................................................................................... 17
References :........................................................................................................................................................... 18
3. Introduction
Businesses Analytics have evolved rapidly over the last few years. On one side there have been many
technological advancements and generation of lot of data, on the other side it has given in the following
challenges:
Too much data, but no insight.
Delay in availability of information when needed.
Toomany variety of data to be integrated – structured, logs, web, email, machine data, social media
feeds, and blogsacross different sources with the right set of rules.
This mandates the need for a good enterprise Information system to be able to answer the following
questions:
What has happened?
What are the factors that led it to happen?
When and where has it happened?
What is happening now?
Do we see a trend or pattern emerging?
What factors can influence a better outcome for business?
After making the changes, are the desired outcomes as expected?
What is Business Analytics?
The true encashment of data as an asset lies in delivering the right set of information through right
visualization to the right internal and external users. This enables realization of business objectives by the
right set of information visualization techniques, measuring and monitoring the business drivers via these
information sets. Any impediment here results in improper, inaccurate and untimely visibility into vital metrics
and key performance indicators. This leads to losing market share (customers), poor performance (financial,
operational), inability to service external demands for information (regulatory/compliance) and losing
competitive advantage. Therefore, it is extremely essential that overall integration, presentation and delivery
of information is required in a process, architecture and technology. This combination is what we refer to as
“Business Analytics”
Business Analytics isabout discoveringand delivering facts, insights and patternsto business users. All this is
based on an integrated enterprise data set. Business Analytics have two main characteristics. It is based out of
data and converts the data to information, relevantknowledge and intelligence useful to the business.
Analytics tools can help us visualize and make the right business choices.
4. Why Business Analytics is important?
We spend valuable time and money everyday in ensuring that appropriate business decisions are based upon
solid, accurate data and information. Analysts typically spend 80% of their time retrieving and
manipulatingdata and only 20% of their time using and analyzingthe information for decision making.
The table below highlights the current state and desired state from an Analytics standpoint for most of the
organizations.
Current State
Time Spent
Acquiring Information
Desired State
Time Spent
Analysis and Making Decisions
Extrapolate or guess at missing data Consistent information across organization
Extraction processes are inconsistent and
require incremental changes
Extraction is planned, organized, structured and
primarily automated
Resulting numbers don’t “tie”
Ability to view data from many angles - slice
and dice, consolidated system and drill down to
location, etc…
Filtering of data done manually at many levels Drill down to pinpoint problems/opportunities
Data integration is time consuming and leaves
limited time for analysis
Efficiently arrive at sound business decisions
based on facts validated through analysis
Businesses should invest in Business Analytics applications for the following reasons:
High visibility
Competitive advantage
Better and faster decisions
Industry leaders rely on analysis of data
High rate of return
Operational efficiency
High Visibility
Business Analytics applications provide a very high level of visibilityof businessperformance. A company could
visualize its strengths, weaknesses, opportunities and threats far better.Good business analytics solutions
provide visibility at operational level. For example,the seniormanagement could look at its operating margins
based on finance, drill it down to operational drivers which drive the expense and further drill down to factors
which contribute to that expense drivers.
Competitive Advantage
Business Analytics solutions convert the data to information, knowledge and intelligence.
Companiesthatheavily invest on Business Analytics solutions will have the correct information through the
correct people and at the correct time. Cutting edge knowledgeover their peers in the industry provides a
competitive advantage as well.
5. Better and Faster Decisions
Companies that have heavily invested in Business Analytics base their decision on data.
Data driven decisions are rational, better and faster. Data driven operational decisions could be easily
automated. Since the executives will be free from the operational decision making, they can spend their
valuable time more on analysis and jobs which add value to the organization. Better decision making
throughout the organization provides the ability to respond to changing market trends.
Top Performers go for Business Analytics
A recent survey conducted by IBM and MIT Sloan School of Management Review across 3000 executives
spread over 100 countries and 30 industries found that top performers in the industryuse the Business
Analytics five times more than their peers in the industry. Top performers agree on the fact that business
information and analytics makes a huge difference on their decisions and more and more decisions are driven
by data. (3)
Operational Efficiency
Business Analytics also results in operational efficiencies in the organization leading them to be more nimble
at information management and deliver the following results:
Delivers a scalable reporting platform where business users are able to run canned and ad hoc reports.
Eliminates the cost of custom programming for each report; eliminates business user dependency on
programmers for data access.
It also delivers the following benefits:
o Aggregated data not available in transactional systems
o Trend analysis — by calendar year, quarter, month, and day of week
o What-if scenario analysis
o Report scheduling
o Various output and display formats , such as Excel, PDF and graphs/charts
o Different manifestations via mobile, tablet and desktop.
Return on Investment
Nucleus Research stated in their survey that organizations driven by analytics garner high return on
investment when compared to their peers. Higher return on investment was achieved across industries as well
as government sector. According to Nucleus Research, every dollar spend on analytics, gives a return of 10.66
dollars. (4)
6. Role of Analytics in business
The core principle of business remains the same irrespective of the industry.
Every business works on the following principles and Business Analytics helps to achieve them:
Maximize its revenue
Minimize its operating expense
Maximize return on long term assets
Manage its risk better.
Business Analytics providesbusiness performance visibility at three levels. (1)
Financial performance level
o Revenue Growth, Operating Margin(%), Operating Margin (%), Risk Exposure Index etc.
Operational Drivers Level
o Revenue Dollar, Overhead Cost Index, Staff Productivity Index, Operational etc.
Operational Factors level
o Pricing, Inventory Management, Investment, Risk Assessment etc.
Financial performance level visibility is at macro level and operational level visibility is at the micro level. At
each level the focus will be on revenue maximization, expense reduction, maximize return on long term assets
and manage risk. At each level analysis on history, current and future is very important. Business Analytics
helps to do suchanalysis. A good Business Analytics solution provides a good visibility at all three levels and
helps to make better decisions.
Figure1: Visibility Diagram
Financial level performance provides information about financial indicators. Financial Planning and Analysis
(FP&A) function manages this level visibility.The most common analysis is Budget vs. Actual, profit and loss
analysis at a particular level. A mere financial visibility is just not enough. Operational driversdrive
7. theoperational efficiency, which helps the financial performance to achieve its goals.Operational factors
contribute to the operational drivers which in turn helps the business to achieve its goals.
Figure 2provides an idea about three levels of visibility (1). Please note that Figure 2 is a common
representation and not an industry specific representation. Verticals and clients could build the visibility
diagram based on industry and business requirement. Specific mapping such as Operational Drivers to Factors
is not done in this diagram since it is beyond the scope of this document and the diagram.
Figure 2: Different Levels of Business Analytics
8. Business Analytics - IndustryFocus Areas
Industry Focus areas
Banking Integrated enterprise view of risk and finance and a single view of the customer
Federal Performance optimization, open government, fraud, and risk management
Insurance
Customer transformation, sales force effectiveness, distribution strategy,
underwriting, and claims excellence
Life Sciences
Physician targeting, managed markets, safety analytics, and generic drug
competition
Healthcare Healthcare reform, fraud, quality of care improvement, and compliance
State
Government
Performance optimization, cost reduction, fraud management, and workforce
planning
9. Business Analytics Reference Architecture
Business Analytics Reference Architecture focusses on integration and transformation of the data, conversion
of the data to information, knowledge, business intelligence and useful analytics. This document does not
discuss about “Data in” (Data Write) systems. Performance Analytics (Planning) isan exception to this.
Figure 3 provides a high leveloverview of Business Analytics Architecture
Figure 3: Business Analytics Reference Architecture Diagram
Business Analytics architecture has the following main components:
Data Layer (Read)
o Structured Data , Unstructured Data
Analytics Access
o History , Current and Future Analysis
Performance Analytics
o Financial Planning Application
Delivery Medium
o Intranet, Email, Mobile, Internet, File Transfer etc.
Security
o Object level and Data Level Security
10. Data Layer
A good Business Analytics solution starts with the Data Layer. Most of the business rules are applied in the
Data Layer.
Why isData Layer important?
Foundation and Structure
When building a house, we begin with the foundation followed by the structure, walls and finally the painting.
Simultaneously, in a Business Analytics system, Data Layer corresponds to the foundation and thestructure.
Most of the majorwork is done in the Data Layer.
Data Layer is the repository of data integrated from various sources which could be internal as well as
external. Data sourced and stored could be structured or unstructured. It holdsa huge volume of historical
data for doing analysis and designed for Data Read. Data layer provides the consistent results and provide the
single version of Truth.
Integrate Relevant Data
The source system contains huge volume of structured data. Socialnetworks contain huge volume of
unstructured data. We do not need all thedatathat comes to the Data Layer.Chances are that there might be
duplication of the same data in multiple places and the grain of such data could be in too detail in the source
systems. Such kind of data may not be required or helpful in decision making.
When data is acquired in the Data Layer, duplication of data has to be prevented and the data should be
summarized to the required level and stored.This will provide a single version of the ‘Truth’. In the case of
unstructured data, all the data is not relevant for business. The first level of separation of the unwanted data
happens while acquiring in the Data Layer.
Build for Insight
Data acquired is not presented in the same form in the Information Access layer. Acquired data is cleaned,
conformed, business rules are applied and transformed. Data models are built in the Data Layer to suite the
requirements of the business function. The conversion of Data to Knowledge happens in the Data Layer. Data
is kept in a highly suitable form (de-normalized) for analysis.
Critical for Success
In some cases the Source System may not have the required data for the information. Data has to be derived
out of the existing data. For example,a company that wishes to look at its profit and loss by product line
level,capture its revenue in the Source System at the product level and captures the expensesin the source
system at a higher level than the product level. To derive the expensesat product line an accounting allocation
has to be done. Performance analytics application cannot handle the huge volume of data. The allocation
process is done in the Data Layer using allocation business logic.
Another company has built its Data Layer for each functional area like data silos. In this case, cross functional
analysis cannot be done easily here.
11. The types of Data Gap mentioned above have to be carefully addressed while designingthe DataLayer. A good
Data Layer is very important for the success of analytics. Any gap in the Data Layer cannot be filled in the
Analytics Access Layer. So business may not be able to do the critical analysis required for success.
Data Layer Systems
Data Read systems can be categorized into two major systems based on the structure of the data.
• Structured Data (Data warehouse Data Mart)
• UnstructuredData (Social Media)
Structured Data
Data Warehouse Data Mart
Data Warehouses Data Mart source most of the data from internal systems and provide insight on strengths
and weakness of a business. It contains data for historical and current analysis. Data Warehouse contains huge
volumes of historical data and is capable of tracking the history with the context at that point of time. In a
Data Warehouse, data from various source systems are integrated, cleaned, transformed and kept in a
structured form, which is easy to read. Typically a Data Warehouse Data Mart contains structured andsemi
structured data.
Unstructured Data
Lots of businessrelevant information originates in the unstructured format such as text, email, call logs, web
logs, social media feeds, blogs, tablets, sensors and mobile data. Businesses have recently discovered that
there are valuable insights that can be unlocked from these data sets to accurately drive business decisions.
Unstructured data could lead to the following trials:
The timeliness of delivery: Is this information delivered immediately to the right business group for
visibility?
The action ability of insights: Is the information sufficient to enable a business decision for a better
outcome?
12. Analytics Access Layer
Users access the data through the AnalyticsAccess Layer. Even though some business logic is applied in the
Data Layer, majority of the business ruleis applied in the Analytics AccessLayer. Data conversion of
information, knowledge, business intelligence and useful analytics happens in this layer.
Figure 4 provides an overview of Analytics Access Layers.
Figure 4: Analytics Access Diagram
13. Historical Analysis
Historical Analysis answers things that happened in the past. It is similar to the rear view mirrorin a car.
Historical Analysis normally answers thewhat, whenandwhere of a query. Canned reports and adhoc analysis
is used for Historical Analysis.
Canned Reports
ITbuilds the Canned Reports based on the requirement of the user. Typically, canned reports are list or cross
tab reports build with prompts and provides low level information. Most of the canned reports are very detail
in nature.
Adhoc Analysis
Adhoc analysis provides the end user ability to slice and dice the data. Theycan drill down the data to a
detailed level. Adhoc capability provides better ability to analyze the business. IT builds a user friendly data
model for AdhocAnalysis. It benefits both the users as well as IT. Userscan get more analytical capability and
the burden on IT comes down since they do not have to build the canned reports and maintain them.
Current Analysis
Current Analysis projects the current status. It is similar to the car windshield while driving. Key Performance
Indicator (KPI) dashboards, business score cards are used to get the current analysis.
Business Score Cards
Score cards are strategic tool for the top management to monitor business performance.Business works to
achieve a particular target or goal. Business score card monitorsthe performance of the business. In a business
there is always the comparison of the current performance against the predefined target. Scorecards are the
best tools, which is strategic in nature and focuses on indicators and trends.
Key Performance Indicator (KPI) Dash Boards
KPI dashboards help to monitor business performance in key areas. It monitors a specific functional area and
measures the currentvalue against the expected value. A KPI dashboardis tactical in nature and focusses more
on Measure Values.
Future Analysis (Advanced Analytics)
Future Analysis (Advanced Analytics) uses statistical and complex mathematical techniques to see a pattern or
trend to predict the future.The functionality is similar to the GPS used while driving the car to understand the
traffic ahead.
Advanced Analytics usesthe predictive model to derive knowledge out of data. Secondly, it uses the derived
knowledge for further action or decision making (6).
Analytics uses structured data from Data Warehouses as well as unstructureddata. Banks generally use it for
fraud detections, and City Planning usesit for traffic pattern analysis. Retail companies use it to understand
consumer buying pattern, and Energy companies use it gauge the energy utilization pattern.
Future Analytics focusses on the following areas:
Why it happened?
What will happen in future if the current trend continues?
What will be the future outcome?
Howcan we change the outcome?
14. The following are the most commonly used analytics:
Predictive Analytics
Based on historical known trends and current transaction data, it predicts the trends in future. For example,a
weak U.S Baseball Team “Oakland A” hired undervalued players purely based on their pastperformance data.
Predictive analytics predictedstatistically that they will do very well and will return good value for their money.
Oakland A finished that year league in second place, with very less spending when compared to other players
who were more costly. The story of the movie Money Ball is basedon Predictive Analytics. (5)
Data Mining
Data Mining focusses on exploring unknown trends from huge volume of data. For example, while using our
credit or debit card in a new city, the Bank sends us an alert to confirm the transaction. Through Data Mining,
Banks get to know where we spend our money generally. As soon as they detect that our card is being used in
a different place, it triggers an alert to detect any fraud.
Simulation
Simulation replicates a business problem, process, and system. Based on what if analysis, we can predictthe
future behaviour or understand potential bottlenecks. For example, BusinessSimulation software could be
used to analyse a particular strategy,or factors that affect business, and predict the final outcome. Business
school students play a similar business game as a part of curriculum.
Segmentation Analytics
Segmentation Analytics focusses on consumer needs to determine how the market could be divided into
different segments. Segmentation could be based on geography, demographics, or behavioural. For example,
the retailers in the neighbouring areas of Hispanic store specific brands of beers in their grocery stores.. This is
to address a specific market segment based on demographics.
Risk Analytics
Risk Analytics focusses on risk identification, assessment, mitigation and monitoring. It is heavily done in
financial industry. For example, Banks do the credit risk analysis for a loan based on the client’scredit history,
ability to pay the loan, current economic condition and collateral.
Performance Analytics Applications (Current and Future)
Performance Analytics Applications monitor the business performance against pre-defined goals. Business
performance management, corporate performance management,enterprise performance management are
synonym to Performance Analytics.
Performance Management involves three main activities:
Set up goals
Collect important measures and monitor against set goals
Take action to improve the future outcome
Performance Analytics Application is used in financials, sales, project management and Human Resources. But
it is heavily used in Financial Planning and Analysis (FP&A) ,which is the first level of visibility to top
management. Normally FP&A monitor the performance based on Profit& Loss (P&L) and Budget or Forecast
(Future). As the period progresses,management will monitor the actual P&L measures (Current) againstBudget
15. or Forecast P&L. Depending on the variance, management will take action to increase revenue or decrease the
expense.
Performance AnalyticsApplication is a write-back application. Users can write-back data in these applications.
A performance application stores the data in files or in database, which is later loaded into the Data ware
house and the subsequent reportsaccessed throughthe analytics access layer.
.
Delivery Medium
Information is delivered to users through one of the following mediums:
Integrated Corporate Portal / Intranet
Analytics access is done through a web portal. This portal can be a common corporate portal like the
SharePoint or application specific portal like SAP or Cognos. Business users can access the portal while they
are within the corporate network.
Email
Reports, dashboards, score cards can be delivered via corporate email. But most of the corporate emailshave
limited size to accommodate heavy file attachments, thereforereports with less file size only can be
deliveredvia email.If the report is available in a portal, users can access such report by emailing the path to the
portal.
Mobile Analytics access could be done through handheld mobile devices such asiPad, iPhone, and Black Berry
etc. Users can access these reports offline as well as read the interactive version on their mobile devices.
Normally the content delivered in mobile will have less data volume because of the limitations in the
hardware. Dashboards are normally delivered through the mobile platform.
File System
The output of a report could be saved as Excel, PDF, CSV, or Text Format in a file system, which is shared by
group of users. Users can access the files through a web portal also. Normally reports delivered in a file system
are consumed by another application as data source.
Internet
We can access the Analytics access through the internet. Some application in Analytics Access Layer can be
used by users who are outside the company network, such as Bank or Insurance clients, vendors, suppliers etc.
They can access information of their need through the internet or in a de-militarized zone outside the
corporate network.
Security
Information security means securing the sensitive information. Sensitive information is sensitive data like
Social Security Number or a specific report. If personal information like social security or credit card is misused
in wrong hands it can cause a huge problem to the consumer in the form of identity theft/ fraud. Consumers
can file legal lawsuits against the company which is supposed to safeguard such sensitive data of its
customers.
Information security is applied at two levels:
Authentication to use the Business Analytics
Authorization to use specific object or data within Business Analytics.
Authentication
16. Authentication is a mechanism through which the user identity is conformed to access the business analytics
application. For example, a company may have thousands of employees but all the employees are not
authenticatedto access the business analytics application. Only a group of users are authenticated to access
the application.
Authorization
Authorization is next level of security. Authorization follows authentication. Authorization gives the right to
the user to access specific object or data within the object.
Authorization is applied at two levels:
Object level security secures the objects in the Data Layer or Analytics Access Layer.
Data Level Security secures the data in Data Later or Analytics Access Layer.
Object level Security
Object could be a database table, view, report, folder, dashboard, metrics, analyticsmodel. Object level
security is applied in the database or in BI &Analytics applications based on the requirement. For example, a
user from the finance team will be allowed to use objects (reports, dashboards, databasetables) which are
specific and required for his work. He cannot access objects pertaining to marketing or human resources.
Data Level Security
“All the sales Professionals should be able to run the sales report. But they must see only their sales area
data”.
The above business scenario requires a data level security. Data level security could be applied in database
layer or in the BI application Layer. Based on the user login credentials, the users can view data limited to their
area. Data level security could be at the row or column level.
Note: Server or network security is beyond the scope of this document.
High Level Analytic Capabilities – Industry Agnostic
The table below represents the modes of unstructured and structured data, its analytic capabilities and
impacts.
Data Analytics Capability Value
Email Text Analytics
Sentiment Analytics
Customer experience management
Call logs
Web logs
System logs
Sentiment Analytics
Customer Interaction Analytics
Customer experience management
Effective operations management
Social media feeds
Blog feeds
Sentiment Analytics
Marketing Analytics
Text Analytics
Monitor the marketing, brand and service effectiveness
Device data
Sensor data
Usage Analytics
Propensity Analytics
Better consumer tracking and offer management
CRM Marketing
Sales Force
Customer Service
Pricing
Better consumer tracking and offer management
17. Churn
Human Resources Workforce Analytics Effective management of your human talent.
Finance Regulatory reporting
Asset Valuation
Revenue recognition
Risk Analytics
Asset analytics
Manage assets, revenues, risks and liabilities for a stable
financial performance.
Supply Chain Inventory analytics
Planning & Forecasting
Sourcing & Procurement
Logistics & Distribution
Effect channel management and delivery of product
distribution to end consumer.
Product Product Margin
Product distribution
Effective productive management for better profitability
and revenue
How do we Implement Business Analytics Solutions?
The following steps are involved in the implementation of Business Analytics:
Analyse the Current Position
Figure 5 maps the current Business Analytics visibilityand maturity.
Figure 5: Business Analytics Position Diagram
Visualize the Future Position
Future position is basically envisioning your Business Analytics.Please mark the positions also in Figure 5.
Do the Gap Analysis
Analyze the Gap between the current and future positioning. Gap Analysis should focus on information gap as
well as technical gap.
Build the Road Map
This phase will address the current gap and guide us to the future.
18. Implement the Road Map
Prioritize your requirements and implement the road map.
Re-evaluate the Implementation
After a period of time, evaluate the implementation, evaluate the social and technological changes that affect
your business model and if required re-align your road map. Re-evaluation is a continuous process.
Why Cognizant?
Cognizant is a business technology solution company.
The following competencies have given us an edge over our competitors:
Business Driven
A client manger will be placed onsite to ensure seamless alignment with your business process. Senior delivery
manager will help the client manager from near shore or offshore. We have a Two-in-One box client
engagement model. Our focus for our client is to increase their revenue, reduce their operating cost, maximize
the return on long term assets and help them manage their risk better.
Vertical Competence
We specialize in twelve major industry sectors and can understand the businessprocess chain in these
industries and business requirements at enterprise, functional and sub-functional level. Many of our senior
people originate from the businesses we serve, so we are able to offer deep experience across almost all
major industries.
Skilled Workforce
We have trained and certified consultants in each of the technical area of Business Analytics.
Our architects and consultants are spread across the world. We are born global and will help you to design,
implement a Business Analytics solution based on business function and technical requirements.
Delivery Competence
So far we have delivered 4900 projects for 700 blue chip companies.
o Our Business Analytics solutions leverage Cognizant’s proprietary PLATINUM information management
solution. PLATINUM can deliver tangible performance improvements and automate processes at every
stage of your solution: conception, design, development and deployment.
o Our internal Cognizant 2.0 system unites our entire workforce, business partners and clients. This
enables us to share our knowledge and expertise in real time and able to produce better results.
Standards
Our culture is values based, you are assured of the highest ethical standards of integrity, transparency and
corporate governance.
Success stories
We have lot of success stories in Business Analytics space. The following list will speak of our work.
Industry Category Solution Delivered Value to Client
Life Sciences Profitable Promotion Mix Increase Profit
Pharmaceutical giant Better Value from Marketing
Investments
Increase Revenue and Market
Share
19. Insurance Sales Effectiveness Increase Revenue
Information, Media and
Entertainment
Customer Centric Focus Increase customer satisfaction and
Revenue
Oil and Gas Control its Drilling and Production
in Real Time
Manage Operation Better and
effective cost control
Retail Increase Revenue without
additional Promotional spending
Increase Revenue and Profit
Reference:
1. The performance Manager, Proven Strategies for turning information into Higher Business Performance by
Roland Mossiman, Patrick Mosimann and Mug Dussalt.
2.Identifying the Path to Profitability by EVAN STUBBS
3.http://public.dhe.ibm.com/common/ssi/ecm/en/gbe03371usen/GBE03371USEN.PDF
4.Nucleus Research. Research Note: Analytics Pays Back $10.66 for EveryDollar Spent. Document L122,
November 2011, page 1.
5.http://practicalanalytics.wordpress.com/predictive-analytics-101/
6.http://en.wikipedia.org/wiki/Analytics