Volume 6, November 2010

The Analytics Insider

Knowledge Services - Turning Data into Information

Business Metamorphosis
‘A Change for better’

Analytics can make a seamless transformation – Are you ready for the change?

IN THIS EDITION:

 From the Leaders Desk

 Hidden truth-Market Mix Modeling
 Client Case Study
 In the lime light – Inspiration Unlimited
 DYK, Brain Teaser
 From the Leader’s Desk
Analytics - MindTree’s Road Ahead
The best part of starting a new job is when six months
into it, everything that you were expecting is happening.
When I started at MindTree back in February, I
anticipated working in a challenging, start up division for
a great, global company that was going to play a big role
in changing the way the industry viewed analytics. It has
indeed exceeded my anticipation. I could see significant
focus on all aspects of analytics. The right blend of data,
business and statistical analysts coupled with effective
grooming proves our seriousness in the business.
I see the analytic landscape evolving dramatically in two
major areas. The first is the one that gets the most
attention, the social media perspective. A few years ago
the rage was email marketing. This was a cheaper and
faster alternative to paper mail pieces, and it also
allowed for almost instantaneous information to be
gathered on the clicks, converts and behaviors. The
result was the creation of a boom in the online analytics
field.
Email is still very prevalent today, as a communication
tool rather than a marketing tool, but I still get the same
amount of offers and catalogues in my “snail mail” as I
did years ago. The sudden change in business dynamics
and introduction of new tools like Facebook, Twitter for
interacting and marketing has brought social media to
the forefront. With these tools creeping into our lives,
I’m a witness of these within my own family that has
changed the way things work. Emails sent for marketing
purposes are wasted. It is easy to see why social media
is so important.
Here at MindTree we are working on new, innovative
ways to track all of the new social media terms: likes,
friends, followers, fans, tweets, etc. By capturing this
data and combining it with sentiment analysis, we are
able to provide our clients and prospects with an even
more complete view of their customer in today’s day
and age. Our capabilities as a startup are already
months, maybe even years, ahead of my previous
companies.
The second impact that I see analytics having on
companies today is how they use not only their customer
data, but also all of the other data that they have within
their organization. I came from a world where my clients’
main focus was solely on their customers. While we can
all agree that the customer is the most important piece of
the puzzle, yet areas like vendors, stores, and employees
play a lesser, but still important role in the clients’
business.
I believe the next focus for our clients and prospects will
be to utilize all of their data, not just customer data. Doing
so will enable them to make holistic decisions around
things like: store placement, shelf space utilization, hiring,
tracking suppliers, and more. Companies are coming to
realize that almost all of the data they have, in some way,
could be valuable. MindTree is in a perfect position to
help take those companies to the next level with their
utilization of analytics.
“While we can all agree that the customer is the
most important piece of the puzzle, yet areas like
vendors, stores, and employees play a lesser, but
still important role in the clients business ”
Since starting with MindTree, it has been exciting to see
Knowledge Services evolve within the analytics space with
HR Analytics as one of our key offerings. I feel the HR
Analytics work we have delivered and the ones we are
doing will be areas where our expertise would serve as a
great strategic path and a door opener for Knowledge
Services especially in a very sensitive area like HR.
I’m sure there is a long way to go in terms of the areas we
would like to ponder, clients we want serve and to add
value at every stage of our engagement. As the saying
goes “To get through the hardest journey we need take
only one step at a time, but we must keep on stepping”
and that’s exactly what we at MindTree are doing.
- Scott Nugent
 Hidden Truth
Market Mix Modeling
Market Mix Modeling is a process that uses data,
analysis and technology to help companies plan,
measure, validate and optimize their marketing
spend.
Why Market Mix Modeling?
 New media, new markets, new demographics are
creating greater opportunities and greater
challenges for marketers.
 CEOs are demanding greater accountability of
marketing spend.
 The impact of multi-channel marketing is difficult
to quantify using traditional reporting techniques.
 Marketers are being asked to make choices
regarding allocation of scarce resources.
 Assessing the relative ROI of these choices enables
the marketer to make optimize activities and
allocations of resources in order to maximize sales,
return and profitability.
The Objectives of Market Mix Modeling
 Identify and quantify sales, marketing and
advertising activities and spending that drive sales
 Determine the impact of each activity on sales
 Determine the ROI for each activity
 Determine the optimal mix of media that will drive
new customer acquisition costs down while
optimizing your total return on customers
 Establish a data driven process to monitor and
track the effectiveness of future sales, marketing
and advertising activities
Real Time Implementation of Market Mix Modeling:
Project objective:
 Estimate the impact of various marketing activities
on sales then forecast the impact of future sets of
promotional tactics
 Optimize investment allocation across competing
marketing activities to help diverting limited
resources among several marketing activities
Three Pronged Approach:
1. Econometrics modeling
2. The Delphi method
3. Optimization
Econometrics modeling: Historical data
analysis was used to establish cause and
effect relationships to predict revenue.
Consider the top 95% marketing activities
which have high and medium effect on the
revenue.
Delphi: The Delphi method pioneered by
the RAND Corporation is a technique that
aims at building consensus among several
experts, about opinions or views. This
helped in choosing critical 5% marketing
drivers which make less impact on the
revenue.
Optimization: The top 95% of the market
drivers are identified creating the
optimization tool with giving ±5% of the
fluctuation window
Solution: In the real world scenario it’s difficult to
remove any marketing driver from the promotional
activity. The major challenge was to short list 30
marketing drivers from 35, by using predictive analysis &
modeling coupled with discussion and key inputs from
decision makers.
Recommendations: Market Mix Modeling is an
evolutionary process that may challenge existing
paradigms. Preparing a delivery model that
recommends measurable action that can be taken on
the highest-value activities first. Deliver in phases that
support the addition of data elements to optimize
models to end state while ensuring that immediate
actions delivery promised results.
- Shivani Gupta
 Deep Dive: Customer Case Study
Background •Domestic Transfers,, Regional
Corridor Analysis Transfers,International Transfers
 Our Client is a global market leader in money transfer
•Pricing Analyis, Promotion Analysis,
business, providing high-value consumer services for Operational Analysis Campaign Analysis
over 150 years
•Analysis of revenue over same period
 Handling on an average one million transactions Times Series Analysis for corridor
daily, or 12 per second •Identifying High Value Customers, Big,
 Ability to transfer money in over 120 currencies Customer Analysis Medium and Small (BMS)

 Multiple money transfers modes through agents, •Profiling of Agents (Around 410,000
Agent Profiling agents
online and mobiles
 Extensively used for Consumer to Consumer money •SMS Auto notification
Marketing Campaign
transfer •Contacting Customers
Gold Card Penetration by Country
Challenges faced by the customer
 Querying Database containing over 800 million
records
 Retrieve Data for visualizing concepts like
Cannibalization, Penetration, Retention etc.,
 Huge data outputs and Multiple Output Tabs (Excel)
 Data Validation on the production Tables Customer Retention for a Corridor
 Loading Unstructured data into the Database

What MindTree offered?

 MindTree's Analytics team works as a trusted partner
in association with the CS&A (Customer Strategy &
Analytics) team in analysis and presenting
information
 SQL (Analytic Functions) Expertise to retrieve data
 Microsoft Excel Macro Coding
 Quick Response Time
 Converting Unstructured India Census Data to
Structured Data and preparing it for analysis
 Understanding the Business Terminologies and
definitions in the customer organization
 Translate Business questions to Database code
 Extracting, consolidating and presenting the key data
elements in Excel Pivots
 Enabling decisions made by Analysts and Marketing
team
Tools and Technology
 Database: Oracle 10g
 Aqua Data Studio and SQL Plus
 Reporting Tool: Microsoft Excel 2007 and Microsoft
Access 2007
 Core FTP: Transferring Huge data files
Business Insights/Impact:
Based on the data retrieved, Client has been impacted in
the following ways,
 Business analyses the impact by introducing a new
agent location
 Analyze customers who are cannibalized by these new
agents
 The existing agents impacted by the introduction of
new agents
 Analyze Customers who are lapsed and their impact to
the business
 Retaining Existing Customers by considering them as
High Value Customers based on their transactions
 Contacting customers through SMS, Email and Landline
during promotional periods
 Based on the Gold Card Penetration, analyze the
growth in Gold Card Market
 Decision to change the fee to attract the lapsed
customers
 In the Limelight – Inspiration Unlimited

Recognition - 3 year service completion

Anil Kumar
Ashwini Nama
Geetha M S
Ruchi Singh - 2nd November Jayanth Siddaramaiah
Neha Nagpal
Prathibha T N
Anusha Santhanakrishnan - 3rd November Praveena Prasannakumari
Reshma Dash
Santhosh Kumar U
Jaykishan Agarwal - 8th November

Appala Raju Muvvala - 10th November

Pat on the Back
Shalini Shroff - 12th November
Jayanth Siddaramaiah
th
Richard A. - 12 November

Rahul Eugine - 19th November

Jyothi Jayakrishna - 22nd November

Ten on Ten
Abhilasha Singh - 22nd November
Amit Bandil
Scott D Staples - 28th November
Santhosh Kumar U
Sreejit Andrews - 7th December

Sohil Patel - 14th December

Aashima Madan - 21st December

Spot Award
Sangeeth Kumar
nd
Wyatt Hirsch - 22 December
(As part of Oman e-visa RFP Team)
Gaurav Mahajan - 30th December

Answer: From Top Right corner, clock wise 10,8,3,5,6,1,7,2,4,9

 xponential times
Did You Know - We are living in e

There are over 2.7billion searches performed on Google each month.

The number of text messages sent and received every day exceeds the total population of the planet.

There are about 540,000 words in the English language, about 5x as many as during Shakespeare’s time. More than 3,000
new books are published every day.

It is estimated that a week’s worth of New York Times contains more information than a person was likely to come across in
a lifetime in the 18th century.

It is estimated that 40 Exabyte’s (4.0x1019) of unique new information will be generated worldwide this year. That is more
than in the previous 5,000 years.

The amount of new technical information is doubling every 2 years. For students starting a four–year technical or college
degree, this means that half of what they learn in their first year of study will be outdated by their third year of study.

By 2011, it is predicted to double every 72 hours. Third generation fiber optics have recently been separately tested by
NEC and Alcatel that carry 10 trillion bits per second down a single strand of fiber. That is 1900 CDs or 150 million phone calls
every second. It is tripling every 6 months and is expected to do so for the next 20 years.

Predictions are that by 2013 a supercomputer will be built that exceeds the computation capability of the human brain.
And while technical predictions further out than about 15 years are hard to do. Predictions are that by 2049 a $1000 computer
will exceed the computational capabilities of the entire human species.

Brain Teaser
For the answer please scroll up to the previous
page b ‘day column 

Fill in the boxes with numbers from 1-10.

One number can be used only once.