hexaware webinar series presentshexaware.com/casestudies/knowyourcustomer.pdfindia’s fastest...
TRANSCRIPT
Hexaware Webinar Series Presents:Know your customer better - Insights into CRM Analytics
Sundip Gorai – Vice President, Hexaware TechnologiesDec 4th, 12 pm Eastern Time
Our mission : To build value for clients through innovative use of technology and talent
BusinessAreas
Transportation
ERP/HRIT
InsuranceBFS
IndiaIndia’’s s Fastest Fastest
Growing MidGrowing Mid--Sized Sized
CompanyCompany
32 offices worldwide
18 Global locations
17 years of technology outsourcing expertise
55 Global 500 clients
166 Clients served worldwide
187 USD mil Revenues, 06
6900 Employees worldwide
A Global IT and BPO Service Provider
Strategies and Strengths
Core CompetencyManagement of business-
critical applications offshore
Organization TraitsConsultative approach, Responsive and Result-
oriented
Robust BackboneWorld-class infrastructure, Flexible delivery models, SEI CMMi Level 5, BS7799
Track Record88% Repeat Business
Offshore transition expertiseGlobal Delivery
Leading BFSI service provider with proprietary products (Operational Risk, Collections, Leasing, Wealth Management
# 1 Airlines services provider in India
8 of top 10 airlines are our clients
# 1 provider of HR-IT services in India
500+ projects, 750+ resources
Specialized Insurance service provider Content management, Fraud Mgmt, Work flow, SOX, BPO
LEADERSHIP
THROUGH
FOCUS
LEADERSHIP
THROUGH
FOCUS
ENHANCING
VALUE
ENHANCING
VALUE
The analytic CRM Strategy Choice points
• Prebuilt Marts
• Stats and Mining Models
• Off the Shelf Tools
• Platform of Choice
Analytic CRM Strategy
Hexaware CRM Analytics Domain Coverage
360°Customer
View
Cross Sell Analysis
Campaign Analysis
Customer Profiling
CRM Functions
New Business Analysis
Service Request Analysis
Product Acquisition
Analysis
Behavior Analysis
• Derive Hindsight, Insight and Foresight into your CRM functions
• DW ,OLAP and Reporting –Hindsight
• Statistical and Data Mining Techniques – Foresight and Insight
• Data models
• Dimensions/Hierarchy/Metrics
• Slice/Dice/Roll-up/Roll-down
• Operational Reporting and Ad-hoc Analysis
Campaign Analysis
Campaign Analysis provides an organization with a focal point to analyze the effectiveness of individual marketing campaign performance in terms of responsiveness of the target audience and associated conversion rates.• Campaign Evaluation
– Responses– Customers Acquired– Cost of acquiring a Customer– Value Generated
• Targeting /Contact method evaluation– Measure and manage media effectiveness for products
or regions– Perform cost-benefit analysis of your campaign
Campaign Analysis (continued)
What kind of questions the model will answer for you ?1. Which type of Campaigns generated the maximum
responses and helped in acquiring the maximum customers/no. of Accounts?
2. Which Campaign was most economical based on cost of acquiring a customer?
3. Which product family or product family combinations elicited the maximum responses across campaigns?
4. Which type of Marketing Method should be used for a particular segment of customers based on past behavior?
5. How many Campaigns have been launched for the promotion of a particular product over the last 5 years? What has been the amount invested so far?
Cross-sell Analysis
1. Which is the most popular product family holding and What is theassociated customer profile?
2. How many customers hold product family A and B but do not have C?
3. What is the change in the product family holding over time?4. What is the cross sell opportunity for a particular holding and a
given customer profile?
Cross product holding analysis1. How are the customers distributed across a given product family
combination?2. Which combination of First and Second Product Family is most
common? What is the profile characteristic for such a combination?
3. Which are the clusters across product holding combinations?
Service Request Analysis
• Analyze patterns in Service Requests– Inquiry Channel used– Customers Profile– Products for which Service Request was logged
• Analyze Service Slippages– Service Centers– Type of Requests– Type of Customers– Across Time
What kind of questions the model will answer for you ?1. What is the status of Service Requests? What is the percentage of
Delays in processing Service Requests? Is there any pattern in these slippages?
2. What is the inquiry channel commonly used?3. Which product generated the largest number of service calls?4. Which type of Customers log a large chunk of Service Requests? 5. Which types of service requests do customers frequently make? This
can throw light on the way the calls are attended to or the insufficiency in the information provided to the customers. The Organization can modify its approach either in the content or in the channel.
6. Time lag in closing the service requests –Which types have the maximum time lag?
7. Percentage of service calls meeting the standard service levels.8. Which Service requests exceed the Standard service levels? –
Comparison over different periods would provide improvement in customer service or persistency in lack of improvement.
Service Request Analysis Questions
360° Customer View – Sample Reports
Service call analysis report Customer profile analysis report
Opportunity analysis report ERM enquiry analysis report
360° Customer View – Sample Reports
Number of calls across time and services Total revenue for a profile across channel usage
Total opportunities for a business unit across countries Number of open/closed cases against channel usage
New Business Analysis
What kind of questions the model will answer for you ?1. What is the profile of new Customers acquired? 2. Which is the product by which new customers usually establish a
relationship with?3. Which source channel is the most effective in acquiring
new customers? 4. Is there a relationship pattern between source channels and the profile
of new customers they acquire?5. How many customers has the organization not been able to retain
during the first month?6. Is the relationship value of such customers significant?7. Is there an upward trend in the relationship value of new customers?8. What percentage of new business is because of Cross Selling/Up
selling efforts?9. What percentage of Customers have been Acquired because of
Customer Referrals?
Product Acquisition Analysis
What kind of questions the model will answer for you ?1. Which is the most favored product by which customers usually
establish a relationship?2. Is there a trend between such a product acquisition history and
customer vintage?3. On an average how long does it take for a customer to acquire each of
the product in his portfolio? What is the sequence of such acquisitions?
4. How have the customers who have been acquired because of Campaign, faired over time?
5. What has been the retention rate in Year 2 of Customers who joined in Year 1?
Behavior Analysis
What kind of questions the model will answer for you ?1. Which Customers are dormant across all channels for the
past 2 months?2. What is the preferred channel for a customer based on usage?3. Who are my Profitable customers who also show high Usage –who
could be offered higher facilities?4. Who are my problem customers who log a large amount of complaints?5. Who fall in top 10 percentile of users by channel type?6. Which customers have shown an increase in Internet Usage?
Customer Profiling
What kind of questions the model will answer for you ?
•Questions based on Customer Demographics
– Age
– Income
– Profession
– Marital Status
•Questions based on Relational Characteristics – Product Holding, Vintage, Age on Book
•Questions based on Transactional Characteristics – Channel Usage and Profitability Score
– Industry
– Legal Entity
– Net Worth
– Equity Capital
Hexaware CRM Analytics –Product and Service Offering
What will you get?1. Pre-Built Data Model (E/R and Dimensional) to Jumpstart
your Analytics Implementation2. Pre-Built Catalogue of Metrics and Dimensions to aid your
analytic needs3. Pre-Built Reporting templates instantiated in the technology
of your choice for tactical reporting needs4. More than 100+ Critical Business questions covering
various sub-functions to support you in ad-hoc analysis5. Pre-Built Metadata/Pre-Built Reports 6. Onsite – Offshore execution model to reduce cost7. HR Consulting expertise
Hexaware CRM Analytics Data Model Richness
E/R Model•Approx 100+ entities•Approx 1000+ attributes
Dimensional Model•200 Dimensions•70 Measures•200 Reports
The analytic CRM Strategy Choice points
• Prebuilt Marts
• Stats and Mining Models
• Off the Shelf Tools
• Platform of Choice
Analytic CRM Strategy
Stat Analysis/Mining –Tools and Techniques
Mining Techniques
• Classification/Categorical Analysis• Clustering /Association/Neighborhood
Analysis• Decision Trees• Parameter Selection and Improvement• Forecasting• Simulations• Optimization• Others
Tools
• Statistical Analysis - SAS, SPSS, Statistica, Wincross
• Predictive Modeling - SAS, SPSS, KXEN, EVIEW, Oracle Datamining
• Decision Tree/Segmentation - Knowledge Studio, CART, SAS, SPSS
• Forecasting and Simulation - Crystal Ball, SAS, SPLUS
• Optimization - LINDO, Evolver, RISK Optimizer• Risk and Decision Analysis - @ Risk, SAS,
SPSS • Campaign Management – UNICA
Industry Solutions
• Banks, Capital Markets and Insurance • Retail • Consumer Products • Pharma and Healthcare
• Response propensity
• Purchase score
• Credit score
• Relationship value
• Other business criteria
Business flow of campaign management
Analytic CRM project design and Execution
• Assignment model
• Right product/right customer/right channel using SAS linear programming
Campaign objective Campaign description Customer selection
• Data Mining Techniques• Model Selection
Response propensity score for maximum likelihood of response
Tele marketing
• Campaign effectiveness• Evaluation model
(response and returns to cost)
SAS Datasets
SAS Datasets
SAS Datasets
SAS Datasets
Closing the Loop –Learning Feedback
SAS Datasets
SAS DatasetsSAS Datasets
SAS Datasets
SAS Datasets
SAS Datasets
Reporting AND analysis (use SAS BI/ micro strategy/MS excel)
21 3
4
5
6
7
8
9
10
12
13
14
Owner of campaign budget ,costs andSegmentation other constraints
Customer attributes –age/gender/Relationship details
Customer exclusions criteria
Filtered Customer List
Customer selectionBased on other attributes
Select customers who did not respond to past campaign and are poor on other parameters
Split customer list
• Data Mining Techniques• Model Selection
Response to multiple campaigns stored as history
Siebel schedules
and executes campaign
11
• Cross Sell
• Upsell
• Churn
• Retention
• Segmentation
• Customer transition to higher segment
Direct mail
Walk in to advisors
Other medium
Filtered customer list based how old is relationship
n
CRM Integration Process Flow Expertise
Executing the campaign (manual)
Learning from the campaign•ROI modeling, campaign•Optimization
Reporting and analysis (SASOLAP /microstrategy/excel)
Campaign objective, budget and description
Modeling and feeding scores
Customer exclusion
List generation with optimization model
Customer selection
SourcesMassaged
Data
ETL ETL
Data Preparation
Campaign optimization
Planning the campaign
Campaign management targeting the customer/segmentation manager/
Staging data store
Transition SAS Process
Stat Analysis/Mining –Applications and Techniques
Applications Mining Techniques
• Prospect targeting• Call planning• Marketing optimization• Sales force optimization• Propensity to churn• Customer Segmentation• Performance attribution • Funds/fees analysis• Fund benchmarking• Revenue forecasting• Demand forecasting• Probability of default • Loss given default• Probability of claim• Underwriting Scoring• Fraudulent identification• ALM /FTP models (core segregation,
attrition, matched maturity)• Economic capital modeling• Clinical Intelligence• Resource utilization analysis
• Classification/Categorical Analysis– Logistic Regression, Support Vector
Machine, Naïve Bayes, Adaptive Bayes• Clustering/Association/Neighborhood Analysis
– “K” Means, “K” Nearest Neighbor, “O” Cluster, Association
• Decision Trees– CART/CHAID
• Parameter Selection and Improvement– Factor Selection/Non-Negative Matrix
Factorization/Attribute Importance• Forecasting• Optimization• Others
– Regression, ANOVA
CRM Analytic using Mining models
Customer Segmentation Targeting the “right customers”
Objective• To select the appropriate customers for
selective offer targeting Methodology• Classification using decision
trees/logistic regression• Selection of prospects based on their
response to previous campaigns Benefits• Identification of customers most likely to
respond positively to offers• Reduction in the cost of campaign due
to improved targeting• Higher response rate• Offers to new customers on the basis of
their resemblance to existing customers
Objective• Find out typical customer segments
and their characteristics (profitable, loyal, etc.)
Methodology• Clustering• Classification (if we know the
segments)Benefits• Identification of profitable segments• Insight into customer purchasing
behaviour• Offers to new customers on the basis
of their resemblance to existing customer segments
CRM Analytic using Mining models
Prediction of Credit Card Defaults Reducing Customer Attrition
Objective• To retain profitable customers by
predicting their possibility of attritionMethodology• Classification using decision
trees/logistic regression• Association Rule mining used to identify
right offers that will reduce customer attrition
Benefits• Insight into attrition patterns• Identification of profitable customers
most likely to cancel the credit card services
• Reduction in the cost of retention due to precise offers that are likely to be accepted
Objective• To predict the probability of a credit
card customer defaulting on paymentsMethodology• Logistic Regression• Potential defaulters identified on the
basis of demographic factors and purchasing patterns
Benefits• Minimization of credit card defaulters
by identifying potential candidates at the issuing stage itself
• Credit card limits can be adjusted to minimize amount of default
• Selective procedure for issuing credit cards and upgrades that mitigates default risk
The analytic CRM Strategy Choice points
• Prebuilt Marts
• Stats and Mining Models
• Off the Shelf Tools
• Platform of Choice
Analytic CRM Strategy
Consolidation and Maintenance
BI Infrastructure and Data MiningAnalytics
SAS Enterprise Intelligence Platform SAS
Ban
king
Inte
llige
nce
Solu
tions
SAS
Ban
king
Inte
llige
nce
Solu
tions
SAS
Insu
ranc
e In
telli
genc
e So
lutio
ns
SAS
Insu
ranc
e In
telli
genc
e So
lutio
ns
SAS
for E
nter
pris
e R
isk
Man
agem
ent
SAS
for E
nter
pris
e R
isk
Man
agem
ent
SAS
Cust
omer
Inte
llige
nce
SAS
Cust
omer
Inte
llige
nce
SAS
Dat
a In
tegr
atio
n SA
S D
ata
Inte
grat
ion
SAS
Inte
llige
nce
Stor
age
SAS
Inte
llige
nce
Stor
age
SAS
Ente
rpris
e B
I Ser
ver
SAS
Ente
rpris
e B
I Ser
ver
SAS
Ente
rpris
e M
iner
SAS
Ente
rpris
e M
iner
Hexaware SAS Implementation Capability
The analytic CRM Strategy Choice points
• Prebuilt Marts
• Stats and Mining Models
• Off the Shelf Tools
• Platform of Choice
Analytic CRM Strategy
Tool Evaluation and Selection Strategy
Analytics and Datawarehousing
Decision Support
Source Transformation
Data Source
PerformanceAnalytic Layer
ODS Data warehouse
Feed-back loop
ETL LayerStaging Area
ETL Layer
Source Systems –Domain Apps, Legacy Systems
Transaction Applications Layer
Analytics Reporting Mart Creation Mining Models
Analytic Layer
Direct connection Met
a da
ta la
yer
How we differentiate?
• A BI strength spanning the past ( the data warehouse), present (patterns in data) and future (prediction) of data and support and maintaining the same
• Capability to build BI COE and articulate enterprise intelligence strategy Driven by strength of
– Analytic IPs
– Accelerator toolkits
– Agile Framework
– Domain and Technology thought leadership driven consulting
• Relationship based data strategy for client from conception to maintenance
Analytics and Data Mining
Implementation Methodologies
Cred
it R
isk
and
Cred
it S
corin
g
Enterprise Risk Management
Analytics
Mar
ket R
isk
Ope
ratio
nal R
isk
Stru
ctur
al R
isk
Regulatory compliance
Prebuilt Data Model
Prebuilt Metrics and Analytical structures
Prebuilt Predictive and Quantitative Models
CRM Analytics
Cust
omer
Loy
alty
Cam
paig
n M
anag
emen
t
Cust
omer
Acq
uisi
tion/
Ret
entio
n
Cros
s Se
ll /U
p-se
ll
Human Capital Management
Dem
ogra
phic
s
Staf
fing
and
recr
uitin
g
Trai
ning
and
Dev
elop
men
t
Prod
uctiv
ity
Com
pens
atio
n an
d be
nefit
sData Mining, Statistics, and Quantitative Models
Specialized Financial Analytics
Cred
it Ca
rd A
naly
tics
Mor
tgag
e An
alyt
ics
Ente
rpris
e Pe
rfor
man
ce A
naly
tics
Capi
tal M
arke
t Ana
lytic
s
Airline Analytics
Perf
orm
ance
Ana
lytic
s
MR
O A
naly
tics
Rou
te O
ptim
izat
ion
Carg
o An
alyt
ics
BI and A Innovation Lab – Accelerators
Some of the tools developed by the Innovation team that would be part of all projects areSolution Accelerators
Other solution accelerators in usage• TRACE IT• Teradata Analytic • Excel Comparator • SAS Web Compiler
• Web Services Manager• Excel Formatter• Mapping Change Propagator
XML-based query engine for Data Integration environment to ensure ETL coding standards and locate coding errors; saves up to 90% time
A tool for determining duplicate reports, identifying set-subset related reports, location of dead variables, etc., within any BI environment; saves up to 70% time
BI environment metadata foundation tool to create semantic layers such as universes and models in an automated fashion; saves up to 40% time
BI and A Innovation Lab – Jump Start Analytics
Some of our prepackaged analytics includes prepackaged data model, prepackaged reports, prepackages cubes, metric and dimensions. Hexaware has prepackaged analytics in following areas
Jump Start Analytics
Our HR analytics enable organizations to get meaningful insights from HR data collected from various enterprise-wide HR and Non HR systems.
Our CRM analytics enable organizations to get deep insights on customer data collected across various channel systems.
Our niche analytics covers various niche areas that includes Credit Card analytics, Mortgage analytics, Credit Risk Analytics and Capital Market Analytics.
Our Airline Analytics enables organizations to strategize their business activities by analyzing various data such as customer loyalty, routes, and cargo.
Our Enterprise Risk Management subsidiary Risk Tech provides Analytics implementation and services related to the banking risk domain
Customers
Thank You For Attending
http://www.hexaware.com/webcastarchive1.htmlFor a recording of this webinar please visit:
Hexaware Webinar SeriesUpcoming Webinars
Download Case Studies & Whitepapers at: www.hexaware.com
Register Today!
Data Mining- Seeing the Future and Knowing the Patterns Of Your Business Using Your Organization DataHow can banks leverage analytics across various perspectives protecting their current investment in technology?