Post on 26-Feb-2019
Pricing Optimization Unsecured Lending
Jane Zhong
Customer Knowledge & Insights
Scotiabank
Toronto Data Mining Forum
May 15, 2013
Page 2
Agenda
Who we are
What is pricing optimization
Project Initiative
Stakeholders
Analytical Solution
Pricing Solution
Lessons Learned
Questions
Page 3
Who We Are
Customer Knowledge & Insights
Enterprise
Data
Warehouse
Decision
Support &
Management
Market
Research
Predictive
Analytics
Customer
Analytics
Customer
Interaction
Management
Predictive Modeling:
Business/Sales Growth
Asset/Liability Management
Optimization:
Marketing Optimization
Pricing Optimization
Responsibility
Page 4
What is Pricing Optimization?
Upon product application, specific customer attributes are potentially used to determine a customer’s optimal price based on business objectives.
o Different pricing strategies could exist depending on the business
objectives. o Optimizing the trade-off between volume and profit within a set of
business constraints.
o Re-optimizing the pricing whenever necessary to maximize net revenue and manage the sales conversion.
A strategy of customizing pricing for each customer based on an understanding of its sensitivity to pricing.
This means…
Page 5
How Can Pricing Optimization Help?
Pricing optimization is
enabling bank executives to use pricing as a strategic weapon to meet their corporate objectives and to become more customer-centric in their decision making.
Dr. Robert L. Phillis
Founder &
Chair Science Officer
Nomis Solutions
Page 6
Project Initiative
Background
Unsecured lending is a key component of the retail bank.
Our existing approach relied on broad risk-based pricing and discretion.
However, this approach had resulted in fewer sales and lower profit
Business Case
Similar to the best practices in the industry, apply dynamic pricing optimization technique for pricing decision.
Leverage our internal expertise to build a pricing optimization framework in house.
Build a process that regularly monitors segment level sales to re-optimize our pricing and maximize the overall product profit.
With a modest investment in IT, complex pricing optimization techniques can be simplified for practical deployment within our sales channels.
Until we incorporate customer relationship fully within the pricing model, pricing discretion capabilities will remain an option.
Page 7
How will it Happen?
Enterprise Data Warehouse
Builder of data mart to store
pricing solutions and interactions
for monitoring, reporting and
further enhancements
Information Technology Solutions
Builder of the “generic” pricing engine and
system enhancements that will provide individual
customer pricing decision
Branches
Decision Support & Management
Provider of pricing solutions through
predictive analysis & optimization to
determine the factors & parameters which
will drive the pricing decision
Call Center
Online
Page 8
Stakeholders
Project Sponsor: Unsecured Lending
Solution Provider: Decision Support & Management
Data Provider: Enterprise Data Warehouse Business Requirement: Integrated Business Solution
System Development: IT&S
Price Execution: Channels
Customer Experience
Risk Management
Audit
Compliance
Page 9
Scenario
Creation
Scenario/Pricing
Signoff
Price
Publishing
Price
Execution
Performance
Monitoring
Pricing
Strategy
Analyze planned
vs. actual results,
update models and
reports to for better
performance
Performance
Monitoring
?
ANALYTICS
Simulate, optimize,
and collaborate to
develop pricing
scenarios for
evaluation
Scenario
Creation
?
Business objective
Strategic Focus
Market Change
Competitors
Pricing
Strategy
?
Data driven
proposals to adjust
pricing for changes
in the market,
and/or our
business goals
Scenario and
Pricing
Signoff
?
Automate the
publishing of prices
and store an
auditable record of
pricing decisions
Price
Publishing
?
Optimize
negotiation ranges,
and offers and
tracking and
incorporating
pricing discretion
Price
Execution
?
Pricing Process
Page 10
Analytical Framework
Booking Models
Attrition Models
Utilization Models
Loss Models
Customer data
Account data
Pricing
Optimization
Platform
Reporting
Monitor/Track
App Data
Risk Data
Pricing Solution
Business Line Signoff
Price Engine
Page 11
Analytical Methodology
To identify major pricing factors by building different customer behavior models
Booking Sensitivity Model
Account Attrition Model
Limit Utilization Model
Loss Model
To build sensible pricing segments on business objective
Combine business sense and scientific study
Customer homogeneity
To deliver better pricing solution by using optimization
Maximize life-time product profit and satisfy certain constraints:
Improve booking ratio
Maintain loss ratio
Improve booking mix of profitable customers
Improve profit from certain customer segments
Page 12
Pricing Solution
Price
H H M M L L H H M M L L H H M M L L H H M M L L H H M M L L Y N Y N Y N Y N Y N Y N Y N Y N Y N Y N Y N Y N Y N Y N Y N 5 5 5 5 5 5 4 4 4 4 4 4 3 3 3 3 3 3 2 2 2 2 2 2 1 1 1 1 1 1
New Price for A Old Price for A New Profit for B Old Profit for B New Price for C Old Price for C
New Price vs Old Price
Profitability orientation not volume orientation
Multiple pricing factors vs. sole pricing factor
Stay competitive within the unsecured lending market
Page 13
Lessons Learned
Pricing evolution as opposed to revolution – make small steps
Time to market is a key
Phased approach provides immediate benefit while gaining executive supports
Need more experts with end-to-end knowledge
Build robust methodology and flexible infrastructure
Work around limitations and control the controllables
Data doesn’t lie and let the number talk first
Identify data gap and collect more relevant data
Need sufficient price driven behavior data to quantify price elasticity
Marry business sense and scientific study for customer segments
Models will need periodically calibration for market volatility
Re-define pricing strategy to align with business objectives and market
Pricing is a key strategic weapon and should continue to be invested