smarter predictions dean abbott co-founder and chief data
TRANSCRIPT
![Page 1: Smarter Predictions Dean Abbott Co-Founder and Chief Data](https://reader035.vdocuments.us/reader035/viewer/2022081323/589d7d981a28ab0d4a8ba37a/html5/thumbnails/1.jpg)
Smarter Predict ions
Dean Abbott Co-Founder and Chief Data Scient ist , SmarterHQ
President, Abbott Analyt ics
[email protected] Twit ter : @deanabb
![Page 2: Smarter Predictions Dean Abbott Co-Founder and Chief Data](https://reader035.vdocuments.us/reader035/viewer/2022081323/589d7d981a28ab0d4a8ba37a/html5/thumbnails/2.jpg)
2
SmarterHQ - Why
Built For Marketers, By Marketers 40+ years multi-channel retail
25+ years machine learning
20+ years digital marketing
Value Delivered, Clients Proven time to value
Self funding platform
20x ROI
Thought Leadership, Funding 2015 Sherpa campaign of the year
8 eTail award winning clients
Funded by Battery Ventures & CNB
![Page 3: Smarter Predictions Dean Abbott Co-Founder and Chief Data](https://reader035.vdocuments.us/reader035/viewer/2022081323/589d7d981a28ab0d4a8ba37a/html5/thumbnails/3.jpg)
3
SmarterHQ - Who
• A platform that unites customer intelligence and cross-channel marketing.
• Focuses on attainable innovation that grows with retailers.
• Delivers on the promise of unifying marketing channels to provide consistent messaging, offers
and promotions…and more.
• Places shoppers at the center of the conversation to provide a full understanding of a shoppers
relationship with a brand, from online to in-store.
• Enables marketers to take action on this intelligence within their existing marketing partner
investments; from email, to display to onsite personalization.
![Page 4: Smarter Predictions Dean Abbott Co-Founder and Chief Data](https://reader035.vdocuments.us/reader035/viewer/2022081323/589d7d981a28ab0d4a8ba37a/html5/thumbnails/4.jpg)
4
Website Mobile
App In-Store Call
Center 3rd Party
Data for Customer Intelligence Customer-centric, Multi-channel
B R A N D L E V E L
E N T E R P R I S E L E V E L
O T H E R B R A N D S
#3314
#9102
#2456
#5306
Droid
iOS
Mobile S
ite
Desktop S
ite
Am
azon
eBay
Web S
upport
Spring C
atalog
![Page 5: Smarter Predictions Dean Abbott Co-Founder and Chief Data](https://reader035.vdocuments.us/reader035/viewer/2022081323/589d7d981a28ab0d4a8ba37a/html5/thumbnails/5.jpg)
5
Website Mobile
App In-Store Website Mobile
App In-Store
B R A N D L E V E L B R A N D L E V E L
iOS
Desktop S
ite
Call Center 3rd Party
#3314 #2456
iOS
Desktop S
ite
Mobile S
ite
Customer Intelligence – Why It Matters
Droid
Am
azon
eBay
Web S
upport
Spring C
atalog
#9102
E N T E R P R I S E L E V E L
O T H E R B R A N D S
He is a different shopper at every level…
Integrating his interactions tells a complete story.
W H O I S F R E D ? #5306
Droid
#9102
#3314
E N T E R P R I S E L E V E L
O T H E R B R A N D S
![Page 6: Smarter Predictions Dean Abbott Co-Founder and Chief Data](https://reader035.vdocuments.us/reader035/viewer/2022081323/589d7d981a28ab0d4a8ba37a/html5/thumbnails/6.jpg)
6
H O W
• How are we communicating?
W H A T
• What are we talking about?
It’s A Conversation
G O A L
Each interaction, or lack thereof, with a brand is a shopper trying to have a conversation.
W H O
• Who am I talking to?
Email Display
Complete The Conversation
Customer Intelligence
W H O W H A T
Cross-Channel Marketing
H O W
![Page 7: Smarter Predictions Dean Abbott Co-Founder and Chief Data](https://reader035.vdocuments.us/reader035/viewer/2022081323/589d7d981a28ab0d4a8ba37a/html5/thumbnails/7.jpg)
7
Tuning Abandonment - Conversation
G O A L
Recover a shopper that abandoned their shopping cart .
W H O W H A T H O W
• Anyone • Recent Visit To Website
• Carted Items
• Did Not Purchase
• Targeted Messaging
• Automated Email
Version 1…
What is the right action?
W H O I S K R I S T E N ?
![Page 8: Smarter Predictions Dean Abbott Co-Founder and Chief Data](https://reader035.vdocuments.us/reader035/viewer/2022081323/589d7d981a28ab0d4a8ba37a/html5/thumbnails/8.jpg)
8
Tuning Abandonment - Conversation
G O A L
Recover a shopper that abandoned their shopping cart that was acquired through paid media.
W H O W H A T H O W
• Guest
• High Marketing Cost
• First Visit To Website
• Carted Items
• Did Not Purchase
• Targeted Messaging
• Targeted Offer
• Automated Email
Version 2…
What is the right action?
W H O I S K R I S T E N ?
![Page 9: Smarter Predictions Dean Abbott Co-Founder and Chief Data](https://reader035.vdocuments.us/reader035/viewer/2022081323/589d7d981a28ab0d4a8ba37a/html5/thumbnails/9.jpg)
9
Tuning Abandonment - Conversation
G O A L
Recover a shopper that abandoned their shopping cart that was acquired through paid media.
W H O W H A T H O W
• Guest
• High Marketing Cost
• First Visit To Website
• Carted Items
• Did Not Purchase
• Targeted Messaging
• Targeted Offer
• Automated Email
• Onsite Messaging
Email Onsite
![Page 10: Smarter Predictions Dean Abbott Co-Founder and Chief Data](https://reader035.vdocuments.us/reader035/viewer/2022081323/589d7d981a28ab0d4a8ba37a/html5/thumbnails/10.jpg)
10
Optimizing Margin – The “Non-Conversation”
G O A L
I want to optimize the value of my High AOV, engaged customer.
W H O W H A T H O W
• Customer
• High Dollar AOV
• Highly Engaged
• Browsed Premium Product several
times in past 3 days
• High predicted propensity to purchase
within the next 3 days
• Get out of his way and let him
purchase.
• Save margin and do not make an offer
• if he doesn’t purchase within 3 days,
then make an offer if purchase propensity
still high
Email Onsite Display
![Page 11: Smarter Predictions Dean Abbott Co-Founder and Chief Data](https://reader035.vdocuments.us/reader035/viewer/2022081323/589d7d981a28ab0d4a8ba37a/html5/thumbnails/11.jpg)
11
Predictive Models -> Insights
Asset Quality Visit Quality Engagement (with
Visit Quality Decay)
Checkout Abandon Causality
Marketing Attribution
Category Browse Abandonment
Customer Attrition Risk
Purchase Replenishment
Predicted Next Product/Category
to Purchase
Cross-Sell / Up-Sell /
Recommendations
Predicted Days to Next Purchase
Predicted Days to Next Visit
Customer (Lifetime) Value
![Page 12: Smarter Predictions Dean Abbott Co-Founder and Chief Data](https://reader035.vdocuments.us/reader035/viewer/2022081323/589d7d981a28ab0d4a8ba37a/html5/thumbnails/12.jpg)
12
Days To Next Purchase: Single Regression Model
15 days
All customers
1day 7days 30 days
![Page 13: Smarter Predictions Dean Abbott Co-Founder and Chief Data](https://reader035.vdocuments.us/reader035/viewer/2022081323/589d7d981a28ab0d4a8ba37a/html5/thumbnails/13.jpg)
13
Days To Next Purchase: Single Regression Model, One Customer
15 days 1day 7days 30 days
Single customer
6.8 days +/-‐ 1 day • Uni-‐modal • One value per customer
![Page 14: Smarter Predictions Dean Abbott Co-Founder and Chief Data](https://reader035.vdocuments.us/reader035/viewer/2022081323/589d7d981a28ab0d4a8ba37a/html5/thumbnails/14.jpg)
14
Days To Next Purchase: Set of Binary Classification Models
15 days 1 day 7days 30 days 3 days
15-‐30 days
7-‐15 days
3-‐7 days 2-‐3 days
Single customer
![Page 15: Smarter Predictions Dean Abbott Co-Founder and Chief Data](https://reader035.vdocuments.us/reader035/viewer/2022081323/589d7d981a28ab0d4a8ba37a/html5/thumbnails/15.jpg)
15
Days To Next Purchase: Set of Binary Classification Models
15 days 1 day 7days 30 days 3 days
15-‐30 days 7-‐15 days 3-‐7 days 2-‐3 days
Single customer
![Page 16: Smarter Predictions Dean Abbott Co-Founder and Chief Data](https://reader035.vdocuments.us/reader035/viewer/2022081323/589d7d981a28ab0d4a8ba37a/html5/thumbnails/16.jpg)
16
Days To Next Purchase: Set of Binary Classification Models
15 days 1 day 7days 30 days 3 days
15-‐30 days 7-‐15 days 3-‐7 days
2-‐3 days Single customer
![Page 17: Smarter Predictions Dean Abbott Co-Founder and Chief Data](https://reader035.vdocuments.us/reader035/viewer/2022081323/589d7d981a28ab0d4a8ba37a/html5/thumbnails/17.jpg)
17
Website Mobile
App In-Store Call
Center 3rd Party
Models for Customer Intelligence How Many Do We Need?
B R A N D L E V E L
E N T E R P R I S E L E V E L
O T H E R B R A N D S
#3314
#9102
#2456
#5306
Droid
iOS
Mobile S
ite
Desktop S
ite
Am
azon
eBay
Web S
upport
Spring C
atalog
![Page 18: Smarter Predictions Dean Abbott Co-Founder and Chief Data](https://reader035.vdocuments.us/reader035/viewer/2022081323/589d7d981a28ab0d4a8ba37a/html5/thumbnails/18.jpg)
18
What Data Best Predicts Behavior?
hDp://www.kaushik.net/avinash/compeHHve-‐intelligence-‐analysis-‐tools-‐metrics-‐reports-‐techniques/?utm_campaign=viralheat&utm_medium=social&utm_source=twiDer
![Page 19: Smarter Predictions Dean Abbott Co-Founder and Chief Data](https://reader035.vdocuments.us/reader035/viewer/2022081323/589d7d981a28ab0d4a8ba37a/html5/thumbnails/19.jpg)
19
Data Preparation: Feature Creation
![Page 20: Smarter Predictions Dean Abbott Co-Founder and Chief Data](https://reader035.vdocuments.us/reader035/viewer/2022081323/589d7d981a28ab0d4a8ba37a/html5/thumbnails/20.jpg)
20
Data Preparation: Variable Reduction
![Page 21: Smarter Predictions Dean Abbott Co-Founder and Chief Data](https://reader035.vdocuments.us/reader035/viewer/2022081323/589d7d981a28ab0d4a8ba37a/html5/thumbnails/21.jpg)
21
Data Preparation: Stratified Sampling (for speed only)
![Page 22: Smarter Predictions Dean Abbott Co-Founder and Chief Data](https://reader035.vdocuments.us/reader035/viewer/2022081323/589d7d981a28ab0d4a8ba37a/html5/thumbnails/22.jpg)
22
Modeling—Lots of Preparation to Get to a Single Model
![Page 23: Smarter Predictions Dean Abbott Co-Founder and Chief Data](https://reader035.vdocuments.us/reader035/viewer/2022081323/589d7d981a28ab0d4a8ba37a/html5/thumbnails/23.jpg)
23
A Quick Aside on Target Shuffling
![Page 24: Smarter Predictions Dean Abbott Co-Founder and Chief Data](https://reader035.vdocuments.us/reader035/viewer/2022081323/589d7d981a28ab0d4a8ba37a/html5/thumbnails/24.jpg)
24
A Quick Aside on Target Shuffling: 500 runs
![Page 25: Smarter Predictions Dean Abbott Co-Founder and Chief Data](https://reader035.vdocuments.us/reader035/viewer/2022081323/589d7d981a28ab0d4a8ba37a/html5/thumbnails/25.jpg)
C U S T O M E R I N T E L L I G E N C E U N I T E S C R O S S - C H A N N E L M A R K E T I N G
Thank You!!