CO N SU M ER D ATABASE

AN D D ATA M IN IN G
PRESENTED BY :
BHUPALI CHOUDHRY
NABASMITA BORA
PARIKSHIT SHARMA
HIMANGSHU PATGIRI

Consum er database
Firms maintain a record of all their present and potential
customers using customer database. The database
stores inform like personal details, buying habits, last
interaction with the firm, contact information etc.
Helps understand :
customers buying pattern which is used to design
products and decide on prices.
The companies use the available list of potential
customers to generate higher sales.
Database makes it possible for the firm to provide
personalized service to the customers.

W hat to collect?
Communication details
Customer name
Contact details email id, physical address, phone number
Preferred mode of contact
How was the customer acquired Google search/social media/store
visit/trade show.
signup form and could use the email id or phone number as the primary key for
the record. This data is mainly used for communication and personalization of
the communication.
Transaction details
Items purchased along with their categories
Value of the sale
Time of purchase (will be used to compute frequency)
Method of payment (Paypal, cash,check,credit card)
In case of credit, how soon are the payments made.

 Engagement details
Does the customer respond to the emails and other communication?
What type of messages does the customer respond most often (fun/promotions/informational)?
Previous emails, social media messages and other communication.
Record of calls made to the help desk.

The engagement details can be got by plugging communication channels with

the CRM. It is essential to effectively support the customer and also to customize
the interactions.
Profile details
Birthday of the customer (used for special promotions on birthdays apart from using age
for segmentation).
Region (rural/urban & north/south) this should be computed from the physical address in
the form or can be got from the IP address used during sign up.
Hobbies & Interests hard to get, but can be useful for correct targeting of promotions.
Income levels & family background very hard to get, but could be gleaned from email
interactions.
Technology usage level can be used for customer support.

H ow to collect the data?

Sign up forms.
Messages from communication channels such
as email
Facebook connect and other social media
connections through special apps.
Surveys
Web analytics this will help you track users
browser environment, location and time spent
on the site.
Calls made to the help desk.
Purchase transactions

Data cleaning.
Good Data:CRM reflects relevant, customized and validated
data only
Data normalization:De-duplicate and merge duplicate records
within the contact database Standardize data formats
Maintain data hygiene:Monitor, update and enrich CRM data
on an on-going basis; replenish leads and prevent data decay of
your database
Build relevant new data:match Ideal Prospect Profile (IPP) and
build contact lists matching the required criteria only.
Campaign ready data:easy tagging and list segmentation for
focused nurturing; better lead qualification and conversion

Data warehousing
Data warehousing is basically a process
in which data from multiple
sources/databases is combined into one
comprehensive and easily accessible
database. Then this data is readily
available to any business professionals,
managers, etc. who need to use the data
to create forecasts and who basically
use the data for data mining.

D ata m ining
Data mining is a process used by
companies to turn raw data into useful
information. By using software to look for
patterns in large batches of data,
businesses can learn more about their
customers and develop more effective
marketing strategies as well as increase
sales and decrease costs. Data mining
depends on effective data collection
andwarehousingas well as computer
processing.

Data mining

Term inology
ETL- In computing, Extract, Transform,
Load (ETL) refers to a process in
database usage and especially in data
Enterprise
resource planning (ERP) is business process
warehousing.

management
software that allows an organization to
use a system of integrated applications to manage the
business and automate many back office functions
related to technology, services and human resources.
OLAPis an acronym for Online Analytical
Processing.OLAPperforms multidimensional
analysis of business data and provides the capability
for complex calculations, trend analysis, and
sophisticated data modeling.

Customer relationship management

(CRM) is a term that refers to practices,
strategies and technologies that companies
use to manage and analyze customer
interactions and data throughout the
customer lifecycle, with the goal of improving
business relationships with customers,
assisting in customer retention and driving
sales growth. CRM systems are designed to
compile information on customers across
different channels -- or points of contact
between the customer and the company -which could include the company's website,
telephone, live chat, direct mail, marketing

Third-party data, as the name implies, is data

that a marketer acquires from a multitude of
outside sources. Normally used to help create
consumer segments for targeting ads and
marketing messages, third-party data often refers
to information gathered from internet interactions.
Data-management firms aggregate information
from sites across the web that show interests in
particular topics based on behaviour, and that
third-party data is used to categorize people into
groups such as travellers or sports enthusiasts. But
third-party data isn't just digital. It can represent
real-world interactions such as purchase
transactions or mobile phone locations.

Aflat file databaseis adatabasewhich is

stored on its host computer system as an
ordinary unstructured file called a "flat file". To
access the structure of the data and manipulate
it, the file must be read in its entirety into the
computer's memory. Upon completion of the
database operations, the file is again written out
in its entirety to the host's file system. In this
stored mode the database is said to be "flat",
meaning that it has no structure for indexing
and there are usually no structural relationships
between the records. A flat file can be aplain
textfile or abinary file.

D ata M ining Process

M odels and techniques/D ata

m ining

D ata m ining com m only involves

fi
ve m odels

Association model
Sequential model
Classification model
Clustering model
Predictive model

1. Association m odel

Searches for relationships between variables.

For example a supermarket might gather
data on customer purchasing habits. Using
association rule learning, the supermarket
can determine which products are frequently
bought together and use this information for
marketing purposes. This is sometimes
referred to as market basket analysis.
Example :If a customer buys corn chips,
then 65% of the time, also buys cola Unless
there is a promotion, in which case buys cola
85% of the time

Association in D ata M ining

2.SequentialPatterns

Similar to Association Models, except that

the relationships among items are spread
over time. Sequences are associations in
which events are linked by time
Require data on the identity of the
transactions in addition to details of each
transaction.
Example : time or event based -A customer
orders new sheets and pillow cases followed
by drapes in 75% of the cases

3.Classifi
cation m odel
Most Common Data Mining Model
Describe the group that a member belongs to by
examining existing cases that already have been
classified, and inferring a set of rules.
What are the characteristics of customers who are
likely to switch to a rival telecom service provider?
- Which kinds of promotions have been effective in
keeping which types of customers so that you can
target the right promotion to the right customer?
Example :-Opera ticket buyers are usually young
urban professionals with high income while country
music concert ticket purchasers are typically blue
collar workers

4.Clustering
Discovers different groups in the data whose members
are very similar.
The Clustering model discovers groupings within the
data - One does not know what the clusters will be
when they start, or on what attributes the data will be
clustered. -Hence, a user who is knowledgeable in the
business needs to interpret the clusters.
Example: -Xerox has developed predictive models
using clusters for analyzing usage profile history,
maintenance data, and representations of knowledge
from field engineers to predict photocopy component
failure. -An is sent to the repair staff to schedule
maintenance PRIOR to the breakdown -Root Cause
Analysis enables a prescription for what to do about
a problem

5.Predictive M odels
Relate behaviour of customers (dependent variable)
to predictors (independent variables felt to be
responsible for the dependent one)
Example 1: Predefined Predictors -If the customer is
rural and her monthly usage is high, then the customer
will probably renew. -If the customer is urban and new
feature exploration is high, then the customer will
probably not renew. Example 2: Customer Profiling
-We can tell the profile of someone who is about to
have a baby by what purchases they make We can
then compare that profile with those of others who are
moving into baby space to predict needs. For instance,
such a customer may be a good target for a life
insurance sales pitch.

G O ALS of data m ining

Prediction : how certain attributes within the data
will behave in the future.

Identification : identify the existence of an item, an

event, an activity.

Classification : partition the data into categories.

Optimization : optimize the use of limited resources

Im portance of data m ining

Customer Profiling
Manage customer relationship
Targeted Marketing
Market-basket analysis
Fraud detection
Prevent customer attrition
Mine unstructured data

 Seek and retain most profitable customers

Segment markets for a targeted approach
Predict the future and identify factors to
secure desired effects
Improve customer acquisition and
retention
Increase customer lifetime value
Detect and minimize risk and fraud
Reduce cycle time while maintaining
quality in product development

 Support scientific research

Identify potential respondents
Discover which customer groups
buy specific products
Identify which customers will most
likely defect
Predict which customers are likely to
repay their loan on time

Applyingdatam ining toCRM

In order to build good model of CRM system, there are a
number of steps that need to be followed. Following are
the
basic steps of data mining for effective CRM: Define Business Problem
Build Marketing Database
Explore Data
Prepare Data For Modelling
Build Model
Evaluate Model
Deploy Model and Result

GOOGLE

Three m ain types of data G oogle

collects:
1.
Things you do
When you use our services for example, do a
search on Google, get directions on Google Maps,
or watch a video on YouTube we collect data to
make these services work for you. This can include:
. Things you search for
. Websites you visit
. Videos you watch
. Ads you click on or tap
. Your location
. Device information
. IP address and cookie data

2.

Things you create

If you are signed in with your Google Account, we store

and protect what you create using our services. This can
include:

Emails you send and receive on Gmail

Contacts you add

Calendar events

Photos and videos you upload

Docs, Sheets, and Slides on Drive

3. Things that make you you

When you sign up for a Google account, we keep the basic

information that you give us. This can include your:

Name

Email address and password

Birthday

Gender

Phone number

Country

H ow G oogle autocom pletes your

searches

The spelling correction model uses data from people who have
made the same mistake before to correct it. That is how we know
when you type Barsalona, you most likely meant Barcelona.
Your search history can also help Google autocomplete your
searches. For example, if you have searched for Barcelona
flights before, we might suggest this in the search box before you
even finish typing it. Or if you are a fan of the football club and
often search Barcelona scores, we might suggest that right away.

How YouTube finds videos you want

to watch

We get cues about what is popular and trending based on

what everyone is watching. This helps us suggest the latest
and greatest videos across a variety of genres like top
music tracks, how-to tutorials, and breaking news.
YouTube also recommends videos you may like based on
what you have watched before. If you are obsessed with
watching covers of the theme song from Frozen, we can
suggest more of them. Or if you are a gamer entertained
by PewDiePie, we can suggest more of his videos and
videos like them.

Conclusion
The secret to an effective CRM package is not
just in what data is collected but in the
organising and interpretation of that data.
To manage the relationship with the customer a
business needs to collect the right information
about its customers and organise that
information for proper analysis and action. It
needs to keep that information up-to-date, make
it accessible to employees, and provide the
know how for employees to convert that data
into products better matched to customers
needs.
Data mining is the solution to this.

Snapshot of Sales force application

D ata w are house

Som e CRM system providers

 REAL LIFE EXAMPLE : IDEA CELLULAR

LIMITED

THANK YOU