9
GUILT & REDEMPTION Phase Two Kelly Foy, Tucker Hammel, Dan McKenzie & Charles Whelan

Analytics Competition Continued

Embed Size (px)

Citation preview

Page 1: Analytics Competition Continued

GUILT & REDEMPTION

Phase Two

Kelly Foy, Tucker Hammel,Dan McKenzie & Charles Whelan

Page 2: Analytics Competition Continued

Agenda

Q1:DemographicsGeographic AnalysisCluster AnalysisQ2:

Optimization FunctionPrice Elasticity/Sensitivity

Q3:Pricing Strategy

Page 3: Analytics Competition Continued

Q1: Customer Demographics

$70,000-$79,999 income

73% Are “asian/pacific islander”

39% 4 year education

60% Employed for wages

66% Shopping up to 2-3 times a month

Page 4: Analytics Competition Continued

Q1:Demographics-Geographic

Presence highest in Northeast, Great Lakes,Florida, and Northern & Southern California metro regions

Modest in Piedmont-Atlantic

Low everywhere else

Low/Modest & high demand Texas Triangle, Piedmont, Cascadia

GILT- WTP. & PURCH INT.

Non Gilt- WTP. & PURCH INT.

Page 5: Analytics Competition Continued

1: Younger, Low Income2: High PI, Low Income3: Younger, High Income4: Low PI, High Income5: Older, High Income

Q1: Clustering of survey participants

Page 6: Analytics Competition Continued

How to determine maximum revenueStep 1: Determine unique WTPStep 2: Calculate total revenue for each WTP in the following functionPrice * Number of people = RevenueStep 3: Find maximum revenue and set optimal priceStep 4: Repeat for each look

Page 7: Analytics Competition Continued

Q2: Price Sensitivity and the Elasticity of Demand

The demand elasticity attained for any look can be attained by the function (%change in demand)

The range of elasticities we attained was between -.87 and -1.47

These elasticities are determined between the values of optimal price and its associated number demanded, and the number demanded at gilt’s price

(%change in price)

Page 8: Analytics Competition Continued

Q3: Maximized values within the given datasetProduct: Optimal Price; Optimal Revenue; Revenue in sample at Gilt price

Product 1: 99; $29601; $4792Product 2: 24.99; $8821.47; $4275Product 3: 49.99; $15746.85; $2933Product 4: 29.99; $8487.17; $7020Product 5: 75; $23700; $20493Product 6: 50; $9850; $2904Product 7: 49.99; $13597.28; $6557Product 8: 39.99; $9917.52; $3490

Amount gained$24809

$4546.47$12813.85

$1467.17$3207$6946

$7040.28$6427.52

Page 9: Analytics Competition Continued

Q3: Pricing Strategy

Continue gathering info on what maximum prices customers will purchase a product

Survey customers before an item goes on sale

Implement optimization function

Guarantees minimum level of sales and would increase the amount of money Gilt earns

All while decreasing the price of goods