Copyright © 2014, Oracle and/or its affiliates. All rights reserved. | CON 8965 Jim Acker Industry...
-
Upload
beatrice-james -
Category
Documents
-
view
214 -
download
0
Transcript of Copyright © 2014, Oracle and/or its affiliates. All rights reserved. | CON 8965 Jim Acker Industry...
Copyright © 2014, Oracle and/or its affiliates. All rights reserved. |
CON 8965
Jim AckerIndustry Solutions ManagerOracle Global Business Unit, Financial Services
Customer Profile in a Big Data Client Solution Approach: Monetizing Customer DNA
Copyright © 2014 Oracle and/or its affiliates. All rights reserved. | 3
Trends in Consumer ExperienceUsing Customer Analytics to Create More Personalized CX
Customers will make web / mobile their primary interaction with the financial institution
All interactions of each individual customer are turned into a personalized experience:
Those channels are already heavy personalized and the customer will expect the same from the financial institution
Brands will use more differentiating content or offers to acquire and retain customers, to up-sell and cross-sell
Copyright © 2014 Oracle and/or its affiliates. All rights reserved. | 4
Status of Personalization
However, few companies have been able to implement
Source: Econsultancy, Digital Marketing Exchange
of those surveyed believe that "personalization is critical to our current and future success”
94%
Copyright © 2014 Oracle and/or its affiliates. All rights reserved. | 5
Barriers to Customer Experience Management
regard IT roadblocks and lack of technology as barriers to adopting or improving personalization
Source: Econsultancy, Digital Marketing Exchange
No Solutions – No Automation – Manual Work – Low ROI
84%
Copyright © 2014 Oracle and/or its affiliates. All rights reserved. | 6
I have a customer - what are the top 3 products he is likely to buy?
Answering the Tough Questions…
Which top hundred customers are likely to buy my product X today?
What is the best channel to connect with my customer, and when? Can I turn around my most valuable potential churners?
Copyright © 2014 Oracle and/or its affiliates. All rights reserved. | 7
Getting to Actionable Customer Insights
Getting from Raw Data to Individual Preferences
Traditional Data Warehouse based solutions (DW/BI) are costly, slow to implement and change, work with sample data and provide limited insight
Big Data and advanced analytics provide an ideal solution for predictive customer insight that is more cost effective, easier to implement and change, and operates in real-time on ALL your data
Copyright © 2014 Oracle and/or its affiliates. All rights reserved. | 8
• Male, born in 1948
• Grew up in England
• Married twice, children
• Successful, wealthy, celebrity
• Loves dogs and the Alps
8
Challenges with Traditional ApproachEffective Customer Treatment Requires 1:1 Personalization
Copyright © 2014 Oracle and/or its affiliates. All rights reserved. | 9Oracle Confidential
real-time decisionengine
data
in
tegr
ation
Oracle / NGData Customer Analytics Solution
stream organize analyze decide respond
internal
external
batch
real-time
master data
marketing automation
contentsites
customer service
ecommerce and sales
BI and analytics tools
acquire learn
identity
enrich
advertising platforms
Big Data ApplianceCloudera
data management platform (DMP)
Lily Enterprise
Copyright © 2014 Oracle and/or its affiliates. All rights reserved. | 10
Turning Data into Valuable Customer DNAIntroducing NGData and Lily Enterprise
Identify unique customer behaviors and preferences in real timeView thousands of metrics for each customer
Continuously monitor customers’ evolving preferences to identify opportunitiesBring Analytics to the data – Open towards DW/BI
Copyright © 2014 Oracle and/or its affiliates. All rights reserved. | 11
Lily Delivers Next Generation PersonalizationFrom Raw Data to Individual Preferences
• Listen Better - Lily works with all types of data - all transactions, all behavior, all context - continuously capturing and automatically making real time observations
• Learn Faster - Lily delivers behavior- based models that take into account all context at various levels of granularity, automatically delivering micro-segmentation to the individual customer and multi-contextual recommendations based on predicted customer needs
• Execute Smarter – easily integrates with marketing and BI platforms, allowing companies to deliver offers based on smarter dynamically updated predictions for better customer experiences
Copyright © 2014 Oracle and/or its affiliates. All rights reserved. | 12
Customer DNA
See everything together – comparisons with a Set defined by you, and evolving trend scores for each customer
From Data to DNA – 1000s of metrics determine individual DNA – common, industry and customer metrics
Copyright © 2014 Oracle and/or its affiliates. All rights reserved. | 13
Customer DNADynamically created Sets defined by your own rules
More effective Alerts based on real-time customer metrics
Models available, or easily and dynamically add new models from all available metricsManage Big Data -
Breaking down data silos to gain insights on all customer interactions in one place
With Lily’s Customer DNA and Machine Learning Engine, individual product Preferences are available each moment
Copyright © 2014 Oracle and/or its affiliates. All rights reserved. | 1414
Real Time Delivery Engine
Recommendations improved in real time during interaction Real Time Delivery Engine – Intelligent Interactions
• Automating decision-making in any channel
• I-CX engine recommendations modified based on data collected during the interaction
• Self-learning process determines propensity to do something for each customer
• Prioritizes and triggers events.
Website Mobile SocialIVR
Digital Interactions
Human Interactions
BranchContact Center Sales
• Digital DNA & 360 view• Predictive Analytics• Next Best Action• Next Best Product• Most Relevant Experience
Lily Enterprise
Copyright © 2014 Oracle and/or its affiliates. All rights reserved. | 1515
Deliver Offers in Real-Time
Marketing Automation
Content Sites
Advertising Platforms
eCommerce and Sales
Customer Service
Predictive Models
Business Rules Performance Goals
Real-time Offers
Decision
Self-learning Feedback Loop
Lily Enterprise
Copyright © 2014 Oracle and/or its affiliates. All rights reserved. | 16
?
?
?
Website Real Time Offer Personalization
Copyright © 2014 Oracle and/or its affiliates. All rights reserved. | 17
Mobile Customer ExperienceLocation-Based Real Time Offer Personalization
Mobile Information Mobile Wallet Mobile Redemption
Joe can view and look up favorite shops, restaurants,...
Joe receives merchant offers in his Bank’s Mobile wallet
Joe can redeem coupons through his mobile wallet
Copyright © 2014 Oracle and/or its affiliates. All rights reserved. |
Implementing the Solution at HDFCRussell SangsterVice President, Professional ServicesNGData
Copyright © 2014 Oracle and/or its affiliates. All rights reserved. | 19
HDFC Bank : Background HDFC Bank wants to offer their customers personalized offers, but only at a time when
they are most likely to make a relevant spend at the nearest accessible outlet. The approach was to collect more detailed data about an individual customer’s
spending habits, lifestyle choices and combine this with their propensity to buy and factor in the situational variables.
The challenge is assimilating high-volume/high-velocity data streams quickly to be able to take decisions and implement decision on real-time basis.
HDFC wanted a solution to derive real business value from a wide variety of data types from different sources, and to be able to easily analyze it within the context of all their enterprise data.
Copyright © 2014 Oracle and/or its affiliates. All rights reserved. | 20
HDFC BANK USE CASE: REAL TIME OFFERS
OBJECTIVE To provide real time offers to HDFC credit card customers based on
propensity, geo-location and offer palette Increase customer spend by providing relevant, targeted offers
Copyright © 2014 Oracle and/or its affiliates. All rights reserved. | 21
HDFC Bank Real Time Offer Project
HDFC is looking to enrich their traditional enterprise data with non-traditional yet potentially valuable data for decision making.
At the core of this project HDFC Bank is gaining Customer Intelligence and making relevant Merchant Funded Offers to the banks Customers in ‘Real Time’ for maximum impact
HDFC Bank is presenting their Credit / Debit Card Customers with applicable Bank and Merchant Offers, based upon the Customer buying behavior, by: Real time integration of Customer Credit / Debit Card transaction data Real time analytics to identify and present, to the banks Customers the Merchant and Bank
Offer that has been determined to be of the most interest to them Deliver the relevant offer in real time for maximum impact
Copyright © 2014 Oracle and/or its affiliates. All rights reserved. | 22
Real-Time Offer FlowConceptual View
Copyright © 2014 Oracle and/or its affiliates. All rights reserved. | 23
Real Time Offer - Process Flow
<ADV>Dear Preferred customer, We have exclusive offer of 20% savings at Gucci and Sephora near your location!
Card transaction made at a shopping mall
Transaction data transfer in real-time
A real time calculation linking type of transaction, location information,
offers in vicinity and the propensity associated with the next best action is
done.
Bank’s Data Center
Send real-time offer via SMS based on time,
customer’s location and propensity model
Real-time/batch based understanding of offer acceptance/rejection and subsequent tweaking
of models
Use offer presented at merchant
Copyright © 2014 Oracle and/or its affiliates. All rights reserved. | 24
Architecture and Roles
1.Approved credit card transactions are captured and replicated to RTD Database.
3.RTD looks up the List of Offers, closest merchant to customer location, checks if customer on DNC list, mobile number is available and the best offer is sent to Customer. If any check fails no offers is made.
2. Customers past 1 year transactions details are provided to NGDATA Lily. NGDATA Lily creates Propensity Model for the customers/ the NBO model. Lily does customer identification and location identification to identify the next best spend categories and the merchant categories for this spend.
4. The next best offer is presented via a text message on their registered mobile number.
*
Copyright © 2014 Oracle and/or its affiliates. All rights reserved. | 25
Pilot Timelines
Week Week 1Week
2Week
3Week
4Week
5Week
6Week
7Week
8Week
9Week
10Week
11Week
12
Use Case Confirmation
Infrastructure availability and connectivity
Software installation and configuration
Business discovery
Test cases planning
Development \ Deployment
Pre-Production system testing
Data Preparation
Data Loading and Model Tuning
POC Go Live
Downstream processes from the inferences are not factored in the timelines