Big Data presence in the high volume in the data storages can help in various ways to learn more about the need and trends of the current market which will be useful for all type of organizations. Modern information technology used to analyze the relationship between social trends and market insights is a useful way to have indirectly interlinked to customers and their interests from unstructured and semi-structured data. Such analysis will give organizations a broader view towards the practical needs of customers and once banking industry or any industry could know the customers, they can serve better and with more flexibility. In this presentation, team has primarily created the platform and designed the architecture in big data technology for banking industry to maximize the users of credit card.
2. Case Study Problems Highlights
● Maximize the number of active credit card customers
● Isolate the cards that would likely never be activated to
reduce wasted marketing spend.
● Get 360 degree view of the customers
● Reduce loss of customers to their competitors
3. Key Trends - Credit Card
● As per the latest trend, there is a 21% decline in card applications.
● Direct-mail response rates to credit card offers have been declining.
● 44% of credit-card applications are submitted over mobile devices (including smartphones and
tablets).
● The percentage of consumers receiving a credit card offer rose from 54% to 67% over nine months
last year, suggesting that card marketers are making efforts to either reach new demographics or
under-served segments.
● Consumers’ traffic increased in usage of third-party aggregator websites like Credit Karma, The
Points Guy
● The new Ultra FICO score, due to pilot early in 2019, allows consumers to add checking, savings and
other information to potentially boost their existing FICO score. People who previously would have
been locked out of having a credit card because of a lower FICO score might now be a target
audience
5. Key Trends-Big Data
● According to Microsoft, over 60% of banks in North America say that Big Data would offer them an
advantage, but only 37% have experience with live big data technology.
● Big Data and Business Analytics Revenue Worldwide across Industries-2018
6. Approach
● Requirement Analysis- Platform selection- Technical Architecture- Solution
Design and Development- Testing-Deployment
● Social Media Analytics to understand /gain insight to customer sentiments
and customer behavior leading to personalized offerings, identify new
segments
● Click Stream Analytics on Credit card page
● Customer transaction analysis- Customer spending pattern, investment
choices, etc. This with their social data can help to understand the needs,
desires and preferences of the customer and to come up with a detailed
customer profile
7. Proposed Solution
● Recommendation Engine-Right products and services to the right
customers at the right time.
● Targeted ads based on the individual needs of the customer/for a segment
● Efficient Chatbots reducing the call center traffic and enhanced customer
experience
● Fraud Detection
● Personalization-Customized Products/Offerings/Rewards
● Innovative reward strategies to attract the customers.
● Educate customers –Intuitive short videos
8. Proposed Solution ArchitectureDataSource
• Customer
account
data
• Transaction
Data
• Click
stream
Logs/
Social
Media
DataAccess
• Sqoop
• Kafka
• Flume
DataStorage&Processing
• Storm(Real
Time
Processing)
• Spark ML
• HDFS(Stor
age)-(Batch
processing)
Dashboard(Re
al Time)
SiLK
(Search
Analytics)
Tableau
BI Tools
(Analytics
graphs)
Couch
Base
Apache
Solr
(Search
Platform
)
Pig
Hive
ClientAccess
Analytics
Decide/Solution
Fraud
Alerts
ChatBot
Recommen
dation
Engine
360 degree
Customer
Profile
9. Recommended Tech stacks
Data Layer - HDFS
Data Ingestion Layer- Flume, Sqoop, Kafka
Data Processing Layer –
Spark ML, Storm, Pig, Hive, Solr, Couchbase
Operations and Scheduling layer – Oozie, Zookeeper
Data Presentation Layer (Analysis) –BI Tools, Tableau, SiLK
10. Deployment strategy
● Org restructure- --Independent vertical silos to more horizontal structure
● Identify the stakeholders
● Line of Business-Credit Card LOB
● Compliance Team(Data Privacy Rules in the countries)
● Analytics and Engineering Team
● Educate the team
● Choose the deployment Model-IaaS/BDaaS(Big Data Stack ).Providers
include Amazon EMR, Microsoft Azure,Google Cloud Platform
● Pilot a project
● Follow an Agile Methodology for Project Implementation
● Analyze the results, work on the learnings and feedback and execute more
projects.
11. Solution Design Mock Up
AI Chatbot Personalized Offerings-Real Time
AI Chatbot – Initiate with each visitor on
website and leading them to 3-minute
video for their education, learning ,
understanding and motivating to take
step for applying to credit card. Collecting
Information from Customer profile and
highlighting personalized offers.
Example: Summer (Customer, 23 age)-collecting her social engagements from
Twitter/FB analytics and giving her following offers with rewards and gifts; Wedding
personal loan, home loan, kids education loan, Retirement gift
12. Solution design mock up
Visualization:Public Sentiment on Social media-world
wide
14. Proposed Solution Benefits
● Better Recommended products and services to customers
● More personalization/targeted rewards programs/Pushing real time offers-
Happy customers Expand the customer base/identify new consumer
segments by social listening . i.e. Customer Acquisition
● Customer Retention
● Enhanced customer experience
● Increase cross selling opportunities
● Improved risk management by identification of potential defaulters
● Helps in identifying unique value proposition(variation in Annual fees and
reward structures) to keep the customers using the card and engaged
● Helps in Understanding public sentiment towards the brand
● Using Predictive analytics- knowing not only what consumers want now
, but what they are likely to want in the future and make offers available
for them.
15. Proposed Solution Benefits
● Campaign success/failure- Faster feedbacks Based on sentiments from
customer and hence quicker corrections/changes possible without much
harm to the brand
● Helps prevent fraud: Based on 360 degree customer profile and customer
behavior pattern helps in preventing/reducing losses due to fraud
● Reduced call center costs based on the intelligent chatbot
● According to Oracle, providing customers with what they need can bring in
18% increase in annual revenue.
● McKinsey finds using data to make better decisions can save marketing
spend by 15 to 20%. Banks on an avg spend 8% of their overall budget in
marketing. This can be reduced and generate additional revenue by
targeted marketing strategies
16. Summary
● Banking industry can definitely benefit from Big data. They are a storehouse of
data and the information that can be gained from it is tremendous. Big Data
Analytics helps to get a 360 degree view of customer. This in turn aids in building
a better relationship and engagement with the customer leading to a win situation
for both sides.
● Banks have started realizing the value of Big Data Analytics and in the coming
years, will definitely see the wave of change- Personalization, Intelligent
Chatbots, Automatic Fraud Detection, Robotic Process Automation, all helps in
increasing revenue, reducing costs , reduce cycle time and enhanced customer
satisfaction.