March 23-24, 2017
Doing It Right
SYMPOSIUM
Big Data and Pricing Analytics
Daniel ReaumeAdvanced Analytics
Dow Chemical Company
4th Big Data & Business Analytics Symposium โ March 23-24, 2017
Dow: An Innovation Company
Founded in 1897
120 years of growth
Largest chemical company in the U.S.
Ranked 56th in the S&P Fortune 500
$48 billion in annual revenue in 2016
~ 56,000 employees worldwide
Integrated value chains aligned to high-growth sectors
7,000 product families manufactured at 189 sites in 34 countries across the globe
24th Big Data & Business Analytics Symposium โ March 23-24, 2017
Delivering Breakthroughs to World Challenges
23.3 million metric tons of CO2emission avoidance
3
40% better purification with 30% less energy
~1.5 billion pounds of trans and saturated fat eliminated
45% reduction in preventable disease from handwashingwith soap
Transportation: BETAMATEโข Structural Adhesives
Water: DOW FILMTECโข ECO Reverse Osmosis Elements
Food:Omega-9 Healthy Oils
Health:POLYOXโข Water-Soluble Polymers in LifebuoyโขSoap
4th Big Data & Business Analytics Symposium โ March 23-24, 2017
Why is Pricing Important?
* Marn and Rosiello, Managing Price, Gaining Profit, Harvard Business Review, 1992.
Price
Quality
Innovation
Customer Service
1% improvement
11.1%profit increase*
Efficiencies
Costs, Volume
1% improvement
2.3% - 3.3%profit increase *
4th Big Data & Business Analytics Symposium โ March 23-24, 2017 4
Fundamental Pricing Relationships
5
Pro
fit M
argi
n ($
/kg)
Price ($)
Dem
and
(kg)
Price ($)To
tal M
argi
n ($
)
Price ($)
Max. Profit
4th Big Data & Business Analytics Symposium โ March 23-24, 2017
6
Dem
and
(kg)
Price ($)
Price Elasticity of Demand
Elasticity = % ๐ถ๐ถ๐ถ๐ถ๐ถ๐ถ๐ถ๐ถ๐ถ๐ถ๐ถ ๐๐๐ถ๐ถ ๐ท๐ท๐ถ๐ถ๐ท๐ท๐ถ๐ถ๐ถ๐ถ๐ท๐ท% ๐ถ๐ถ๐ถ๐ถ๐ถ๐ถ๐ถ๐ถ๐ถ๐ถ๐ถ ๐๐๐ถ๐ถ ๐๐๐๐๐๐๐๐๐ถ๐ถ
4th Big Data & Business Analytics Symposium โ March 23-24, 2017 6
Price Elasticity of Demand
7Insight: Lower price if % profit margin x elasticity > 1
Insight: Lower price if % profit margin x elasticity > 1
Elasticity = % ๐ถ๐ถ๐ถ๐ถ๐ถ๐ถ๐ถ๐ถ๐ถ๐ถ๐ถ ๐๐๐ถ๐ถ ๐ท๐ท๐ถ๐ถ๐ท๐ท๐ถ๐ถ๐ถ๐ถ๐ท๐ท% ๐ถ๐ถ๐ถ๐ถ๐ถ๐ถ๐ถ๐ถ๐ถ๐ถ๐ถ ๐๐๐ถ๐ถ ๐๐๐๐๐๐๐๐๐ถ๐ถ
Dem
and
(kg)
Price ($)
4th Big Data & Business Analytics Symposium โ March 23-24, 2017 7
Price Elasticity of Demand
Elasticity = % ๐ถ๐ถ๐ถ๐ถ๐ถ๐ถ๐ถ๐ถ๐ถ๐ถ๐ถ ๐๐๐ถ๐ถ ๐ท๐ท๐ถ๐ถ๐ท๐ท๐ถ๐ถ๐ถ๐ถ๐ท๐ท% ๐ถ๐ถ๐ถ๐ถ๐ถ๐ถ๐ถ๐ถ๐ถ๐ถ๐ถ ๐๐๐ถ๐ถ ๐๐๐๐๐๐๐๐๐ถ๐ถ
8Do we know these values? Does this demand curve apply?What are our constraints? What will competitors do? โฆ
Insight: Lower price if % profit margin x elasticity > 1
Dem
and
(kg)
Price ($)
4th Big Data & Business Analytics Symposium โ March 23-24, 2017 8
What if Non-Price Factors Affect Demand?
Margin ($/kg)
Dem
and
(kg)
Margin ($/kg)D
eman
d (k
g)
RemoveGDP-driven
demand-inflation
RemoveGDP-driven
demand-deflation
4th Big Data & Business Analytics Symposium โ March 23-24, 2017 9
What if Non-Price Factors Affect Demand?
Margin ($/kg)
Dem
and
(kg)
Margin ($/kg)D
eman
d (k
g)
RemoveGDP-driven
demand-inflation
RemoveGDP-driven
demand-deflationHere, ignoring external factors
overestimates elasticity
4th Big Data & Business Analytics Symposium โ March 23-24, 2017 10
What if Non-Price Factors Affect Demand?
Margin ($/kg)
Dem
and
(kg)
Margin ($/kg)D
eman
d (k
g)
4th Big Data & Business Analytics Symposium โ March 23-24, 2017 11
What if Non-Price Factors Affect Demand?
Margin ($/kg)
Dem
and
(kg)
Margin ($/kg)D
eman
d (k
g)
Overestimated elasticity drivesunderestimated optimal price
4th Big Data & Business Analytics Symposium โ March 23-24, 2017 12
What if Non-Price Factors Affect Demand?
Margin ($/kg)
Dem
and
(kg)
Margin ($/kg)D
eman
d (k
g)
Overestimated elasticity drivesunderestimated optimal price
Big Data Question: How do external factors drive demand?
4th Big Data & Business Analytics Symposium โ March 23-24, 2017 13
Different types of pricing decisions occur at different business levels
Transaction
Market
Industry
14
Establish brand and product family positioning
e.g. price product line X to gain share via world-scale efficiencies
Establish target price and volume for product
e.g. average price for product X set to $Y/kg
Maximize profit from a series of transactions
e.g. sell A units of product X in package B to customer C for $Y/kg
4th Big Data & Business Analytics Symposium โ March 23-24, 2017
Microsegmentation: Optimizing Individual Transaction Prices
15
All Transactions:Avg Price = $X
Region #1:Avg Price = $(X-2)
Package #1:Avg. Price =
$(X-3)
Package #2:Avg. Price =
$(X-1)
Region #2:Avg. Price =
$(X+4)
Big Data Question: How do transaction attributes drive customer value?
4th Big Data & Business Analytics Symposium โ March 23-24, 2017
Pricing to Position Within Industry
Action High Price
Competitive Response
Competitive Response
Low marginsHigh sales
Low marginsModerate sales
High marginsModerate sales
High marginsLow sales
Low Price
High PriceLow Price High PriceLow Price
Big Data Question: How will competitors and customers likely respond?
4th Big Data & Business Analytics Symposium โ March 23-24, 2017 16
80-20 rule still applies: โข 80% of value from 20% of modeling effort
Pricing has enormous leverage on profitโข Added fidelity usually justifies additional modeling effort
Incorporating Big Data is critical to fidelityโข Variety: Many different types of factorsโข Volume: Huge amount of data to considerโข Velocity: Critical for real-time pricing
But simpler models may be initially preferableโข Quick + Broad vs. Slow + Narrowโข User acceptance and time value of money
Summary
174th Big Data & Business Analytics Symposium โ March 23-24, 2017