![Page 1: Eric Siegel, Ph.D. Program Chair Predictive Analytics World · Predict influence with uplift modeling – Direct marketing – ... Approach: Target campaign with an uplift model Resulting](https://reader031.vdocuments.us/reader031/viewer/2022011910/5f7a2e1a6a4561662173f034/html5/thumbnails/1.jpg)
Predictive Analytics World
Eric Siegel, Ph.D. Program Chair Predictive Analytics World
![Page 2: Eric Siegel, Ph.D. Program Chair Predictive Analytics World · Predict influence with uplift modeling – Direct marketing – ... Approach: Target campaign with an uplift model Resulting](https://reader031.vdocuments.us/reader031/viewer/2022011910/5f7a2e1a6a4561662173f034/html5/thumbnails/2.jpg)
Twitter hashtag:
#pawcon
DDBW: • Predictive Analytics World • Text Analytics World • eMetrics • Conversion
![Page 3: Eric Siegel, Ph.D. Program Chair Predictive Analytics World · Predict influence with uplift modeling – Direct marketing – ... Approach: Target campaign with an uplift model Resulting](https://reader031.vdocuments.us/reader031/viewer/2022011910/5f7a2e1a6a4561662173f034/html5/thumbnails/3.jpg)
Predictive Analytics World
Eric Siegel, Ph.D. Program Chair Predictive Analytics World
![Page 4: Eric Siegel, Ph.D. Program Chair Predictive Analytics World · Predict influence with uplift modeling – Direct marketing – ... Approach: Target campaign with an uplift model Resulting](https://reader031.vdocuments.us/reader031/viewer/2022011910/5f7a2e1a6a4561662173f034/html5/thumbnails/4.jpg)
Agenda
§ Predict behavior § Predict influence with uplift modeling
– Direct marketing – Retention – Other applications
§ Analytical methods
![Page 5: Eric Siegel, Ph.D. Program Chair Predictive Analytics World · Predict influence with uplift modeling – Direct marketing – ... Approach: Target campaign with an uplift model Resulting](https://reader031.vdocuments.us/reader031/viewer/2022011910/5f7a2e1a6a4561662173f034/html5/thumbnails/5.jpg)
Data driven technology that produces a predictive score for each customer.
Predictive Analytics:
34 12 27 79 42 59 34 12 34 27 12 34 79 27 12 34 42 59 79 27 12 34 12 34
![Page 6: Eric Siegel, Ph.D. Program Chair Predictive Analytics World · Predict influence with uplift modeling – Direct marketing – ... Approach: Target campaign with an uplift model Resulting](https://reader031.vdocuments.us/reader031/viewer/2022011910/5f7a2e1a6a4561662173f034/html5/thumbnails/6.jpg)
Predictive modeling Predictive modeling Predictive model
Customer data
![Page 7: Eric Siegel, Ph.D. Program Chair Predictive Analytics World · Predict influence with uplift modeling – Direct marketing – ... Approach: Target campaign with an uplift model Resulting](https://reader031.vdocuments.us/reader031/viewer/2022011910/5f7a2e1a6a4561662173f034/html5/thumbnails/7.jpg)
Today Tomorrow
![Page 8: Eric Siegel, Ph.D. Program Chair Predictive Analytics World · Predict influence with uplift modeling – Direct marketing – ... Approach: Target campaign with an uplift model Resulting](https://reader031.vdocuments.us/reader031/viewer/2022011910/5f7a2e1a6a4561662173f034/html5/thumbnails/8.jpg)
1%
3%
Today Tomorrow
Today Tomorrow
3%
1%
![Page 9: Eric Siegel, Ph.D. Program Chair Predictive Analytics World · Predict influence with uplift modeling – Direct marketing – ... Approach: Target campaign with an uplift model Resulting](https://reader031.vdocuments.us/reader031/viewer/2022011910/5f7a2e1a6a4561662173f034/html5/thumbnails/9.jpg)
![Page 10: Eric Siegel, Ph.D. Program Chair Predictive Analytics World · Predict influence with uplift modeling – Direct marketing – ... Approach: Target campaign with an uplift model Resulting](https://reader031.vdocuments.us/reader031/viewer/2022011910/5f7a2e1a6a4561662173f034/html5/thumbnails/10.jpg)
“You didn't have to be so nice; I would have liked you anyway.”
– The Lovin' Spoonful, 1965
Predict:
![Page 11: Eric Siegel, Ph.D. Program Chair Predictive Analytics World · Predict influence with uplift modeling – Direct marketing – ... Approach: Target campaign with an uplift model Resulting](https://reader031.vdocuments.us/reader031/viewer/2022011910/5f7a2e1a6a4561662173f034/html5/thumbnails/11.jpg)
![Page 12: Eric Siegel, Ph.D. Program Chair Predictive Analytics World · Predict influence with uplift modeling – Direct marketing – ... Approach: Target campaign with an uplift model Resulting](https://reader031.vdocuments.us/reader031/viewer/2022011910/5f7a2e1a6a4561662173f034/html5/thumbnails/12.jpg)
Predict customer behavior. marketing influence on
^
![Page 13: Eric Siegel, Ph.D. Program Chair Predictive Analytics World · Predict influence with uplift modeling – Direct marketing – ... Approach: Target campaign with an uplift model Resulting](https://reader031.vdocuments.us/reader031/viewer/2022011910/5f7a2e1a6a4561662173f034/html5/thumbnails/13.jpg)
Uplift modeling: Analytically modeling to predict the influence on a customer's buying behavior that results from marketing contact.
“How much more likely does contact make the desired outcome?”
![Page 14: Eric Siegel, Ph.D. Program Chair Predictive Analytics World · Predict influence with uplift modeling – Direct marketing – ... Approach: Target campaign with an uplift model Resulting](https://reader031.vdocuments.us/reader031/viewer/2022011910/5f7a2e1a6a4561662173f034/html5/thumbnails/14.jpg)
Will the customer buy if contacted? only
Will the patient become healthy if treated?
only
![Page 15: Eric Siegel, Ph.D. Program Chair Predictive Analytics World · Predict influence with uplift modeling – Direct marketing – ... Approach: Target campaign with an uplift model Resulting](https://reader031.vdocuments.us/reader031/viewer/2022011910/5f7a2e1a6a4561662173f034/html5/thumbnails/15.jpg)
Trials of B
Trials of A Trials of A
Trials of B Trials of B Trials of B
Test
Control
![Page 16: Eric Siegel, Ph.D. Program Chair Predictive Analytics World · Predict influence with uplift modeling – Direct marketing – ... Approach: Target campaign with an uplift model Resulting](https://reader031.vdocuments.us/reader031/viewer/2022011910/5f7a2e1a6a4561662173f034/html5/thumbnails/16.jpg)
Buy
if d
o re
ceiv
e an
offe
r
Buy if don’t receive an offer
Response Uplift Modeling
No Do-Not-Disturbs Lost Causes
Yes Sure Things
Yes No
COST CUTTER: Don't contact those who'd respond anyway.
Persuadables
![Page 17: Eric Siegel, Ph.D. Program Chair Predictive Analytics World · Predict influence with uplift modeling – Direct marketing – ... Approach: Target campaign with an uplift model Resulting](https://reader031.vdocuments.us/reader031/viewer/2022011910/5f7a2e1a6a4561662173f034/html5/thumbnails/17.jpg)
Respondents “Sure things�”
“Do-not-disturbs�”
“Uplift modeling empowers your organization to capture more than 100% of responses by contacting less than 100% of the target population.”
Kathleen Kane Principle Decision Scientist Fidelity Investments
Predict:
![Page 18: Eric Siegel, Ph.D. Program Chair Predictive Analytics World · Predict influence with uplift modeling – Direct marketing – ... Approach: Target campaign with an uplift model Resulting](https://reader031.vdocuments.us/reader031/viewer/2022011910/5f7a2e1a6a4561662173f034/html5/thumbnails/18.jpg)
Uplift Gains
![Page 19: Eric Siegel, Ph.D. Program Chair Predictive Analytics World · Predict influence with uplift modeling – Direct marketing – ... Approach: Target campaign with an uplift model Resulting](https://reader031.vdocuments.us/reader031/viewer/2022011910/5f7a2e1a6a4561662173f034/html5/thumbnails/19.jpg)
US Bank Business case: Direct mail for a home-equity line of credit Approach: Target campaign with an uplift model Resulting improvements over prior approach: § Campaign ROI increased over 5 times § Costs down 40% § Lift up 2 times
Leading financial institution Business case: Direct mail for a financial product Approach: Target campaign with an uplift model Resulting improvements over prior approach: § Increased revenue per contact by a factor of 20
![Page 20: Eric Siegel, Ph.D. Program Chair Predictive Analytics World · Predict influence with uplift modeling – Direct marketing – ... Approach: Target campaign with an uplift model Resulting](https://reader031.vdocuments.us/reader031/viewer/2022011910/5f7a2e1a6a4561662173f034/html5/thumbnails/20.jpg)
Mining for the Most Truly Responsive Customers Kathleen Kane, Fidelity
Session: Today, 4:15pm
Hi Low
![Page 21: Eric Siegel, Ph.D. Program Chair Predictive Analytics World · Predict influence with uplift modeling – Direct marketing – ... Approach: Target campaign with an uplift model Resulting](https://reader031.vdocuments.us/reader031/viewer/2022011910/5f7a2e1a6a4561662173f034/html5/thumbnails/21.jpg)
Uplift Model Deciles
PERSUADABLES LOST CAUSES/ SURE THINGS
DO-NOT-DISTURBS
![Page 22: Eric Siegel, Ph.D. Program Chair Predictive Analytics World · Predict influence with uplift modeling – Direct marketing – ... Approach: Target campaign with an uplift model Resulting](https://reader031.vdocuments.us/reader031/viewer/2022011910/5f7a2e1a6a4561662173f034/html5/thumbnails/22.jpg)
Response modeling: à Model those contacted
Churn modeling: à Model those not contacted
Test BOTH treatments and incorporate ALL results
with uplift modeling
![Page 23: Eric Siegel, Ph.D. Program Chair Predictive Analytics World · Predict influence with uplift modeling – Direct marketing – ... Approach: Target campaign with an uplift model Resulting](https://reader031.vdocuments.us/reader031/viewer/2022011910/5f7a2e1a6a4561662173f034/html5/thumbnails/23.jpg)
Agenda
§ Predict behavior § Predict influence with uplift modeling
– Direct marketing
– Retention
– Other applications
§ Analytical methods
![Page 24: Eric Siegel, Ph.D. Program Chair Predictive Analytics World · Predict influence with uplift modeling – Direct marketing – ... Approach: Target campaign with an uplift model Resulting](https://reader031.vdocuments.us/reader031/viewer/2022011910/5f7a2e1a6a4561662173f034/html5/thumbnails/24.jpg)
Today Tomorrow
![Page 25: Eric Siegel, Ph.D. Program Chair Predictive Analytics World · Predict influence with uplift modeling – Direct marketing – ... Approach: Target campaign with an uplift model Resulting](https://reader031.vdocuments.us/reader031/viewer/2022011910/5f7a2e1a6a4561662173f034/html5/thumbnails/25.jpg)
Optus (Australian telecom) Doubled churn model performance
with social data
Major N. American Telecom 700% more likely to cancel if
someone in your network does
Major N. American Telecom 10%-15% improvement to churn
model via online behavior, including reviewing contract period
Contract expires à Cancel à Friends cancel
![Page 26: Eric Siegel, Ph.D. Program Chair Predictive Analytics World · Predict influence with uplift modeling – Direct marketing – ... Approach: Target campaign with an uplift model Resulting](https://reader031.vdocuments.us/reader031/viewer/2022011910/5f7a2e1a6a4561662173f034/html5/thumbnails/26.jpg)
![Page 27: Eric Siegel, Ph.D. Program Chair Predictive Analytics World · Predict influence with uplift modeling – Direct marketing – ... Approach: Target campaign with an uplift model Resulting](https://reader031.vdocuments.us/reader031/viewer/2022011910/5f7a2e1a6a4561662173f034/html5/thumbnails/27.jpg)
Leav
e if
do
rece
ive
a re
tent
ion
offe
r
Leave if don’t receive a retention offer
Churn Uplift Modeling
Yes Sleeping Dogs Lost Causes
No Sure Things
No Yes
"Leave well enough alone." "If it ain't broke, don't fix it."
"Do not disturb!"
Persuadables
Sleeping Dogs
![Page 28: Eric Siegel, Ph.D. Program Chair Predictive Analytics World · Predict influence with uplift modeling – Direct marketing – ... Approach: Target campaign with an uplift model Resulting](https://reader031.vdocuments.us/reader031/viewer/2022011910/5f7a2e1a6a4561662173f034/html5/thumbnails/28.jpg)
COST CUTTER: Don't trigger those
who'd otherwise stay.
![Page 29: Eric Siegel, Ph.D. Program Chair Predictive Analytics World · Predict influence with uplift modeling – Direct marketing – ... Approach: Target campaign with an uplift model Resulting](https://reader031.vdocuments.us/reader031/viewer/2022011910/5f7a2e1a6a4561662173f034/html5/thumbnails/29.jpg)
70%
70%
90%
Today Tomorrow
Today Tomorrow
90%
![Page 30: Eric Siegel, Ph.D. Program Chair Predictive Analytics World · Predict influence with uplift modeling – Direct marketing – ... Approach: Target campaign with an uplift model Resulting](https://reader031.vdocuments.us/reader031/viewer/2022011910/5f7a2e1a6a4561662173f034/html5/thumbnails/30.jpg)
• Campaign ROI increased 11-fold • Reduce churn a further 36% relative to baseline • Targeted volume reduced by 40%
Telenor: world’s 7th largest mobile operator - 159 million subscribers
![Page 31: Eric Siegel, Ph.D. Program Chair Predictive Analytics World · Predict influence with uplift modeling – Direct marketing – ... Approach: Target campaign with an uplift model Resulting](https://reader031.vdocuments.us/reader031/viewer/2022011910/5f7a2e1a6a4561662173f034/html5/thumbnails/31.jpg)
Agenda
§ Predict behavior § Predict influence with uplift modeling
– Direct marketing – Retention
– Other applications § Analytical methods
![Page 32: Eric Siegel, Ph.D. Program Chair Predictive Analytics World · Predict influence with uplift modeling – Direct marketing – ... Approach: Target campaign with an uplift model Resulting](https://reader031.vdocuments.us/reader031/viewer/2022011910/5f7a2e1a6a4561662173f034/html5/thumbnails/32.jpg)
![Page 33: Eric Siegel, Ph.D. Program Chair Predictive Analytics World · Predict influence with uplift modeling – Direct marketing – ... Approach: Target campaign with an uplift model Resulting](https://reader031.vdocuments.us/reader031/viewer/2022011910/5f7a2e1a6a4561662173f034/html5/thumbnails/33.jpg)
![Page 34: Eric Siegel, Ph.D. Program Chair Predictive Analytics World · Predict influence with uplift modeling – Direct marketing – ... Approach: Target campaign with an uplift model Resulting](https://reader031.vdocuments.us/reader031/viewer/2022011910/5f7a2e1a6a4561662173f034/html5/thumbnails/34.jpg)
Applications of Uplift
Response upli+ Contact? Churn upli+ Reten*on offer? Content targe2ng With which ad, crea*ve, etc.? Channel selec2on Which channel? Dynamic pricing Which price?
![Page 35: Eric Siegel, Ph.D. Program Chair Predictive Analytics World · Predict influence with uplift modeling – Direct marketing – ... Approach: Target campaign with an uplift model Resulting](https://reader031.vdocuments.us/reader031/viewer/2022011910/5f7a2e1a6a4561662173f034/html5/thumbnails/35.jpg)
Uplift modeling: Analytically modeling to predict the influence on a customer's buying behavior that results from choosing one marketing treatment over another.
“How much more likely is this treatment to generate the desired outcome than the alternative treatment?”
![Page 36: Eric Siegel, Ph.D. Program Chair Predictive Analytics World · Predict influence with uplift modeling – Direct marketing – ... Approach: Target campaign with an uplift model Resulting](https://reader031.vdocuments.us/reader031/viewer/2022011910/5f7a2e1a6a4561662173f034/html5/thumbnails/36.jpg)
Trials of B
Trials of A Trials of A
Trials of B Trials of B Trials of B
Test
Control
Test of treatment A
Test of treatment B
![Page 37: Eric Siegel, Ph.D. Program Chair Predictive Analytics World · Predict influence with uplift modeling – Direct marketing – ... Approach: Target campaign with an uplift model Resulting](https://reader031.vdocuments.us/reader031/viewer/2022011910/5f7a2e1a6a4561662173f034/html5/thumbnails/37.jpg)
“Weigh your options.”
![Page 38: Eric Siegel, Ph.D. Program Chair Predictive Analytics World · Predict influence with uplift modeling – Direct marketing – ... Approach: Target campaign with an uplift model Resulting](https://reader031.vdocuments.us/reader031/viewer/2022011910/5f7a2e1a6a4561662173f034/html5/thumbnails/38.jpg)
If banner is seen: • Related searches
+61% • Paid clicks +249%
![Page 39: Eric Siegel, Ph.D. Program Chair Predictive Analytics World · Predict influence with uplift modeling – Direct marketing – ... Approach: Target campaign with an uplift model Resulting](https://reader031.vdocuments.us/reader031/viewer/2022011910/5f7a2e1a6a4561662173f034/html5/thumbnails/39.jpg)
39
[Client]
![Page 40: Eric Siegel, Ph.D. Program Chair Predictive Analytics World · Predict influence with uplift modeling – Direct marketing – ... Approach: Target campaign with an uplift model Resulting](https://reader031.vdocuments.us/reader031/viewer/2022011910/5f7a2e1a6a4561662173f034/html5/thumbnails/40.jpg)
40
Results:
25% increased “take” rate; 3.6 – 5% revenue increase
A-B test deployment:
A. Legacy system based on acceptance rates across users
B. Model-based ad selection
Almost $1 million per year in additional revenue, given the existing $1.5 million monthly revenue.
![Page 41: Eric Siegel, Ph.D. Program Chair Predictive Analytics World · Predict influence with uplift modeling – Direct marketing – ... Approach: Target campaign with an uplift model Resulting](https://reader031.vdocuments.us/reader031/viewer/2022011910/5f7a2e1a6a4561662173f034/html5/thumbnails/41.jpg)
More Applications of Uplift
Collec2ons Offer a deeper write-‐off? Credit risk Offer a higher credit limit or APR? Electoral poli2cs Campaign in a par*cular state? Personalized medicine Apply which medical treatment?
![Page 42: Eric Siegel, Ph.D. Program Chair Predictive Analytics World · Predict influence with uplift modeling – Direct marketing – ... Approach: Target campaign with an uplift model Resulting](https://reader031.vdocuments.us/reader031/viewer/2022011910/5f7a2e1a6a4561662173f034/html5/thumbnails/42.jpg)
Agenda
§ Predict behavior § Predict influence with uplift modeling
– Direct marketing – Retention – Other applications
§ Analytical methods
![Page 43: Eric Siegel, Ph.D. Program Chair Predictive Analytics World · Predict influence with uplift modeling – Direct marketing – ... Approach: Target campaign with an uplift model Resulting](https://reader031.vdocuments.us/reader031/viewer/2022011910/5f7a2e1a6a4561662173f034/html5/thumbnails/43.jpg)
![Page 44: Eric Siegel, Ph.D. Program Chair Predictive Analytics World · Predict influence with uplift modeling – Direct marketing – ... Approach: Target campaign with an uplift model Resulting](https://reader031.vdocuments.us/reader031/viewer/2022011910/5f7a2e1a6a4561662173f034/html5/thumbnails/44.jpg)
Predictive model for B
Mary’s profile
Trials of B
Trials of A Modeling
Modeling
Predictive model for A Trials of A
Trials of B Trials of B Trials of B
Predictive score: Mary responds to A?
Predictive score: Mary responds to B?
“Two-Model” Approach
![Page 45: Eric Siegel, Ph.D. Program Chair Predictive Analytics World · Predict influence with uplift modeling – Direct marketing – ... Approach: Target campaign with an uplift model Resulting](https://reader031.vdocuments.us/reader031/viewer/2022011910/5f7a2e1a6a4561662173f034/html5/thumbnails/45.jpg)
Mary’s profile
Trials of B
Trials of A
Uplift modeling
Uplift model
Trials of A
Trials of B Trials of B Trials of B
Predictive score: Is A better than B?
Uplift Modeling
![Page 46: Eric Siegel, Ph.D. Program Chair Predictive Analytics World · Predict influence with uplift modeling – Direct marketing – ... Approach: Target campaign with an uplift model Resulting](https://reader031.vdocuments.us/reader031/viewer/2022011910/5f7a2e1a6a4561662173f034/html5/thumbnails/46.jpg)
§ Engagement § Recency § Age
![Page 47: Eric Siegel, Ph.D. Program Chair Predictive Analytics World · Predict influence with uplift modeling – Direct marketing – ... Approach: Target campaign with an uplift model Resulting](https://reader031.vdocuments.us/reader031/viewer/2022011910/5f7a2e1a6a4561662173f034/html5/thumbnails/47.jpg)
Response target segment: High response rate Uplift target segment: High impact
Multi-Var Models
![Page 48: Eric Siegel, Ph.D. Program Chair Predictive Analytics World · Predict influence with uplift modeling – Direct marketing – ... Approach: Target campaign with an uplift model Resulting](https://reader031.vdocuments.us/reader031/viewer/2022011910/5f7a2e1a6a4561662173f034/html5/thumbnails/48.jpg)
US Bank – Direct Mail
Has paid back more than 17.3% of current loan – AND –
Is using more than 9.0% of revolving credit limit – AND –
Is designated within a certain set of lifestyle segments
Purchase rate: 1.83% if contacted 1.07% if not contacted
Segment:
![Page 49: Eric Siegel, Ph.D. Program Chair Predictive Analytics World · Predict influence with uplift modeling – Direct marketing – ... Approach: Target campaign with an uplift model Resulting](https://reader031.vdocuments.us/reader031/viewer/2022011910/5f7a2e1a6a4561662173f034/html5/thumbnails/49.jpg)
Will customers buy without contact? Can retention offers backfire? Are both treatments non-passive?
Will Uplift Modeling Help Me?
![Page 50: Eric Siegel, Ph.D. Program Chair Predictive Analytics World · Predict influence with uplift modeling – Direct marketing – ... Approach: Target campaign with an uplift model Resulting](https://reader031.vdocuments.us/reader031/viewer/2022011910/5f7a2e1a6a4561662173f034/html5/thumbnails/50.jpg)
Download the free white paper:
“Uplift Modeling: Predictive Analytics Can't Optimize Marketing Decisions Without It”
by Eric Siegel, Ph.D.
www.pawcon.com/signup-uplift-whitepaper.php
PAW Workshop - March 9-10, 2012 in San Fran:
‘Net Lift Models: Optimizing the Impact of Your Marketing”
by Kim Larsen
www.pawcon.com/uplift
![Page 51: Eric Siegel, Ph.D. Program Chair Predictive Analytics World · Predict influence with uplift modeling – Direct marketing – ... Approach: Target campaign with an uplift model Resulting](https://reader031.vdocuments.us/reader031/viewer/2022011910/5f7a2e1a6a4561662173f034/html5/thumbnails/51.jpg)
Predict:
influence persuasion impact
Conclusions – Uplift Modeling
![Page 52: Eric Siegel, Ph.D. Program Chair Predictive Analytics World · Predict influence with uplift modeling – Direct marketing – ... Approach: Target campaign with an uplift model Resulting](https://reader031.vdocuments.us/reader031/viewer/2022011910/5f7a2e1a6a4561662173f034/html5/thumbnails/52.jpg)
52
© Copyright 2006. All Rights Reserved Prediction Impact, Inc. © Copyright 2008. All Rights Reserved
Predictive Analytics World Conference
Nov 30-Dec 1, 2011: London, UK
2012: San Francisco, Toronto, Chicago
Bigger wins!
Strengthen the business impact delivered by predictive analytics.
www.pawcon.com
"Predictive Analytics World was probably the best analytics conference I have attended... turned into
my new must-go-to conference." Dennis Mortensen
Director of Data Insights, Yahoo!
March 2011 drew over 500 attendees. PAW has included case studies from: Acxiom, Amazon.com, Bella Pictures, Charles Schwab, ClickForensics, Google, The National Rifle Association, Pinnacol Assurance, Reed Elsevier, Sun Microsystems, TaxBrain, Telenor, Wells Fargo, Yahoo! and many more.
![Page 53: Eric Siegel, Ph.D. Program Chair Predictive Analytics World · Predict influence with uplift modeling – Direct marketing – ... Approach: Target campaign with an uplift model Resulting](https://reader031.vdocuments.us/reader031/viewer/2022011910/5f7a2e1a6a4561662173f034/html5/thumbnails/53.jpg)